What would Loop do?
-It’s 8pm and your predicted BG curve looks like the following. What do you expect Loop to recommend/enact at 8pm? +
It’s 8pm and your predicted BG curve looks like the following. What do you expect Loop to recommend/enact at 8pm? A. Zero (suspend) temp basal B. Scheduled basal from your settings C. High temp basal D. Lower temp basal (between zero and scheduled) -
-
-
What would your answer be? Before you give your final answer...consider this next twist. Would you give the same answer to this graph as you gave the graph above? If not, what would the answer be for this graph and why? (ignore the timestamp mismatch
-
-
-
This second case has a dramatic drop happening. Suspend threshold is still at 60 mg/dL, the correction range is still 90-110, the lowest value on the predicted BG chart is 75 mg/dL, and the eventual BG is 171 mg/dL for this example. So, pretty similar to the first example except for this precipitous drop going on right now.
Let me tell you some of the common pitfalls we all can easily slip into when trying to answer these "Why is Loop giving me this basal?" questions.
Wondering about IOB, COB, or DIA in order to answer...those aren't a factor in answering the question as they have already been used to make the predicted BG curve. In other words, they are accounted for already in the information presented. All you need to answer this question is provided by the predicted BG curve, your suspend threshold, and your correction range. diff --git a/search/search_index.json b/search/search_index.json index b772617..72183a8 100644 --- a/search/search_index.json +++ b/search/search_index.json @@ -1 +1 @@ -{"config":{"lang":["en"],"separator":"[\\s\\-]+","pipeline":["stopWordFilter"]},"docs":[{"location":"","title":"Introduction","text":"
Welcome to LoopTips! The purpose of these pages is to help you make the most of your DIY closed-looping experience. In particular, these pages support users of the Loop app. These documents have been a long time in the making and as always, your feedback is welcome. If you have comments, please feel free to provide input at the Looped Facebook Group or in Loop's Zulipchat.
LoopTips is a companion to LoopDocs. LoopTips does not cover the build or installation of Loop app. If you have problems with building your Loop app or troubleshooting the technical aspects of Loop operations, please head over to LoopDocs for help. The LoopTips pages are specifically related to understanding how your loop is recommending the actions being taken, and how you can improve your blood glucose outcomes on Loop. Some of the pages that originated in LoopTips are now included in LoopDocs. Don't be surprised when you find links between the sites. (The color bar is deliberately a different shade to help you know where you are.)
"},{"location":"#what-is-a-closed-loop","title":"What is a closed loop?","text":"The U.S. Food and Drug Administration (FDA) has a general definition of an \"artificial pancreas device system\" (aka closed loop system) on their website, quoted below.
\"Most Artificial Pancreas Device Systems consists of three types of devices already familiar to many people with diabetes: a continuous glucose monitoring system (CGM) and an insulin infusion pump. A blood glucose device (such as a glucose meter) is used to calibrate the CGM.
A computer-controlled algorithm connects the CGM and insulin infusion pump to allow continuous communication between the two devices. Sometimes an artificial pancreas device system is referred to as a \"closed-loop\" system, an \"automated insulin delivery\" system, or an \"autonomous system for glycemic control.\"
An Artificial Pancreas Device System will not only monitor glucose levels in the body but also automatically adjust the delivery of insulin to reduce high blood glucose levels (hyperglycemia) and minimize the incidence of low blood glucose (hypoglycemia) with little or no input from the patient.\"
"},{"location":"#what-is-loop","title":"What is Loop?","text":"The Loop app is a do-it-yourself closed loop algorithm and user interface, developed through the work of community volunteers. You can read about the history of Loop development in LoopDocs. Loop predicts future blood glucose based on basals, carbohydrate intake, insulin deliveries, and current CGM readings. These blood glucose predictions provide Loop with the information needed to recommend a temporary basal rate to attain a targeted glucose range in the future. The system can either operate as an \u201copen loop\u201d by making recommendations to the user for their approval before enacting or as a \u201cclosed loop\u201d by automatically setting the recommended temporary basal rate.
As exciting as this sounds...we should first get the caveats out of the way. There are some limitations on what kind of pumps/equipment are required. This system doesn't work with most pumps and requires you to do some work. The pumps that are supported and the directions for building are thoroughly covered in LoopDocs.
"},{"location":"#disclaimer","title":"Disclaimer","text":"While it may seem obvious, please consult with your health care professional regarding your diabetes management. The suggestions and discussion in LoopTips are not a one-size-fits-all nor intended to replace the input from your doctor. You take full responsibility for building and running this system and do so at your own risk.
"},{"location":"translate/","title":"Translation","text":""},{"location":"translate/#google-translate-links","title":"Google Translate Links","text":"Click on a language from the list below to turn on Google automatic translation.
- Every LoopTips page gets automatically translated to the selected language as do links to other websites from LoopTips
- The Google Translate Tool will appear at the top of each page
- LoopTips: Google Translate Tool Instructions
- Google: Google Translate Help Link
Automatic Translation
These links connect this site to the Google Translation service.
- No human has reviewed the translated information for accuracy
- Please use the translation with care
- Not available in all regions
- Some links might not work as expected
- Any \"code\" not protected by blocks may not appear correctly - be sure to click on Original to make sure you a viewing code properly
\u0639\u0631\u0628\u064a
\u0411\u044a\u043b\u0433\u0430\u0440\u0441\u043a\u0438
\u010ce\u0161tina
Deutsch
Dansk
\u0395\u03bb\u03bb\u03b7\u03bd\u03b9\u03ba\u03ac
Espa\u00f1ol
\u65e5\u672c
Suomi
Fran\u00e7ais
\u05e2\u05d1\u05e8\u05d9\u05ea
Hrvatski
Italiano
\ud55c\uad6d\uc5b4
Norsk
Nederlands
Polski
Portugu\u00eas
Rom\u00e2n\u0103
\u0420\u0443\u0441\u0441\u043a\u0438\u0439
Sloven\u010dina
Svenska
Turkish
\u4e2d\u6587\uff08\u7b80\u4f53)
\u4e2d\u6587\uff08\u7e41\u9ad4)
"},{"location":"translate/#google-translate-tool-instructions","title":"Google Translate Tool Instructions","text":"Once Google Translate has been turned on, clicking on a link above shows a \"Google Translate: Can't translate this page error\".
- To modify the language for a single page, use the Google Translate tool
- To modify the language for the entire site, see Change Language
The graphic below shows the Google Translate Tool when maximized (default) for a browser and mobile display. The tool can be minimized by tapping on the up/down carets at the right of the tool. This is very useful if the tool obstructs part of the original screen. Additional options can be selected with the three-dot menu as shown in the graphic.
"},{"location":"translate/#change-language","title":"Change Language","text":"To modify the language choice for the whole site, copy the line below and paste it into the URL and then choose the desired language from the list
https://loopkit.github.io/looptips/translate\nOR
Use the Google Translation three-dot menu and select Go to Original URL while on the Translation page.
Perceptus is a website and set of free tools developed by Kenneth Stack and Gustavo Munoz, two active members of the DIY community. Perceptus provides two tools particularly helpful for Loop users; Dash and GlucoDyn. As a side note, Gustavo has an excellent Spanish Tedx Talk about building his own DIY looping algorithm. Kenneth is also the brains behind NSApple watch app.
"},{"location":"data/glucodyn/#dash","title":"Dash","text":"For Loop users with Nightscout integration running, Dash can provide an alternate view and data analysis for the Loop data stored in your Nightscout's mLab database. (Note: mLab used to be called MongoLab, still the same intended reference)
To set up your Dash integration, you will need to:
-
Signup for a Perceptus account here
-
Provide your mLab database information
Assuming you have a basic Heroku account to host your Nightscout website, you can access your mLab information as shown in the screenshots below. (Users who manually setup their own mLab databases as part of Azure accounts will need to access their mLab database manually using their login and account information.)
Login to your Heroku account to start the process, and then that will take you to your mLab information.
Your mLab database name and API Key can be found on the same page, as shown below. Copy and paste these bits of information into your User Settings in Dash.
The format for the MongoLab API URL string is: https://api.mlab.com/api/1/databases/your-database-name/collections/entries where you are replacing the your-database-name part with your info from the first circled area shown below.
Dash provides multiple views of your mlab information, and the data is updated live as your Nightscout data updates too. While there are some constraints (cannot set your own low/high range limits), the data views are very useful for identifying difficult times of the day or difficult days of the week.
"},{"location":"data/glucodyn/#glucodyn","title":"GlucoDyn","text":"GlucoDyn was created to educate people with T1D and their caregivers about blood glucose dynamics. Each day people with T1D make decisions about what and when to eat, and how much insulin to use to cover carbohydrates effect. It can be difficult to visualize what is happening throughout the day as carbohydrate absorption raises blood glucose and insulin reduces it \u2013 particularly when multiple events interact with one another.
GlucoDyn uses calculations similar to an insulin pump\u2019s bolus wizard to create graphs of blood glucose versus time, based on your inputs of carb and insulin events. The user can enter carbohydrates, boluses, and temporary basals, and GlucoDyn calculates the effect on blood glucose over time. GlucoDyn also allows the user to \u201cslide\u201d the events around once they have been entered \u2013 allowing you to see what would have happened to your blood glucose curve if you had bolused earlier or eaten a different amount of carbohydrates \u201con the fly\u201d \u2013 the curves change as you move the sliders.
Seeing the curves and understanding the effects of bolus timing, carbohydrate absorption rate glycemic index (GI) value, temporary basal timing and rates, etc. can help users with their overall understanding of T1D and its management.
Originally, Loop used the same carbohydrate absorption and insulin models as Glucodyn uses. Those have since been updated in Loop with dynamic carbohydrate absorption and exponential curves for insulin. Even with those changes, the GlucoDyn tool can still provide users an excellent tool to visualize how small changes in bolusing can affect trends in blood glucose. For example, it\u2019s quite interesting to see the effect of pre-bolusing. By delaying the carb intake by say 20 minutes, you can visualize the reduction of the maximum blood glucose. But, you have to be careful \u2013 before the carbohydrates have time to absorb insulin is working, and the minimum blood glucose will also drop. These are the kinds of tradeoffs that are critical to the management of T1D, and thankfully are automated by Loop's smart bolusing based on carbohydrate absorption estimates.
"},{"location":"data/health/","title":"Health App","text":""},{"location":"data/health/#overview-of-healthkit-and-health","title":"Overview of HealthKit and Health","text":"There's often confusion over the terminology surrounding Apple's health data storage so let's start there.
- HealthKit -- Apple's iOS database for storing health-related data. App developers can choose to use HealthKit to store information from their apps such as heart rate, blood glucose, insulin use, and a long list of other parameters
- Health App -- Apple allows iOS users to manage HealthKit permissions and view stored HealthKit data through the use of Apple's Health app (standard on iPhones and iPods, but not available on iPads).
When you first install Loop onto your iPhone, you will be asked to setup Loop's access to the HealthKit database. Loop uses the iPhone's HealthKit to read and write blood glucose, insulin, and carbohydrate data depending on how you initially setup this section.
At a minimum, you will need these HealthKit settings:
Loop
- insulin data: read and write
- carbohydrate data: read and write
- blood glucose data: read and write
Dexcom App
- blood glucose data: write (if you have this disabled, Loop will still function but will not display blood glucose values older than 3 hours old)
You can always check your HealthKit settings by opening the Health App, clicking on Sources at the bottom bar, and then clicking on the app's name you are interested in, for example, Loop.
Potential conflicts can arise when third-party apps are granted access to HealthKit permissions that may interfere with Loop's specified data permissions. For example, carbohydrate data is stored in Nightscout and Spike for some users...you wouldn't want to enable Spike app to write duplicate carbohydrate entries that Loop would be reading. Therefore, you should disable other apps from writing carbohydrate data to HealthKit so that Loop does not read those other entries unintentionally. Also, good practice, because carbohydrate entries in HealthKit that were created by non-Loop apps will not be able to customize carbohydrate absorption times nor be edited later if needed.
"},{"location":"data/health/#your-use-of-health-app","title":"Your use of Health App","text":"Summaries of your carbohydrates, insulin, and blood glucose results can be found by clicking on the Health Data at the bottom bar, and then selecting either the large Nutrition box (for carbohydrates) or smaller Results line (for insulin deliveries and blood glucose results).
If you toggle on the \"add to favorites\" slider for the individual data categories (insulin, blood glucose, carbohydrates), the data from those categories will be added to your Today view for easy quick reference and access.
The summary data for the categories can help you follow monthly trends, help identify periods of insulin sensitivity/resistance, evaluate total daily insulin use, breakdown of basal rate vs bolus insulin, and carbohydrate consumption. You can sort your data trends by day, week, month, or year views and scroll back through time in each of those data trends. You can even quickly use these data for endocrinology appointment discussions...as they provide the endocrinologist with a very quick and useful set of data points directly from your Loop.
If you drag two fingers separately like you are spreading them apart, you can get averages for the data set your fingers are covering, as shown below.
"},{"location":"data/nightscout/","title":"Nightscout","text":"Nightscout (NS) is a cornerstone of the DIY type 1 community. It is an excellent tool to remotely view Loop's actions and access insulin and blood glucose data. It allows for easy remote monitoring of Loop activities, troubleshooting of Loop errors, history of CGM, carbs, boluses, & temporary basal rates, and reports to analyze data trends and patterns. There is a Nightscout app in your iPhone App Store, or you can use a web browser to view the data. Setting up Nightscout for Loop is free and fairly quick. Once set up, the site can be accessed by anyone that you share your unique Nightscout URL with.
"},{"location":"data/nightscout/#nightscout-display","title":"Nightscout Display","text":"Nightscout is highly recommended for Loop users, especially those using Loop as caregivers. Nightscout displays are often the easiest way to troubleshoot Loop settings if you are having problems and seeking input from others. Below is some discussion about the general Nightscout display, as well as some Loop-specific display information.
"},{"location":"data/nightscout/#blood-glucose","title":"Blood Glucose","text":"Your blood glucose readings from your CGM are shown in green, yellow, or red in the main graph area of NS. You can adjust your high and low blood glucose targets in NS (when NS alerts will start for high/low BGs), but those will not affect your Loop performance. Loop only uses the blood glucose targets given in the Loop app settings. The main screen displays the time duration you have selected (in the example above, 12 hours of BG history). The very bottom of the screen shows the last 48 hours of blood glucose trends. You can scan backward by dragging the bottom timeline to the left if you want to review specific Loop actions or data in the last two days.
"},{"location":"data/nightscout/#carbohydrates","title":"Carbohydrates","text":"Carbohydrates are automatically uploaded to NS by the Loop app (when configured). The amount of carbohydrates on board (active carbs or COB) can be seen by clicking the Loop pill. The size of a white carbohydrate dot on the graph is proportional to the amount of carbohydrates entered...bigger meals get bigger dots. Loop does not read carbohydrates from NS (nor from the pump's bolus wizard) for use in looping calculations. Loop only uploads carbohydrates to NS that have been entered in the Loop app and recorded to HealthKit.
Insulin boluses are also automatically uploaded to NS by the Loop app. The bolus is shown as a filled-in blue lower half of the dot, and the specific amount of the bolus is also shown. There may be a slight delay between when the carbohydrates for a meal are uploaded to NS to when the meal bolus appears. This is because depending on the pump model and size of the bolus, the bolus may take a few minutes to actually be given and then Loop has to perform a pump read to verify the insulin delivery (i.e., the bolus was given). As shown in the example above, the meal carbohydrates have been uploaded, but the bolus delivery is still in progress. Insulin on board (active insulin or IOB) is also shown in the Loop pill.
Within your NS profile, you will enter in the scheduled basal rates from your Loop settings. This is not automatically updated when you update the Loop app; you will have to manually update your basal profile in your NS site if you change your basal profile in Loop. If you don't update the NS basal profile, it doesn't cause any problems in Loop function...you simply won't have an accurate VISUAL representation of high/low temporary basal rates within NS. The dashed blue line represents the scheduled basal profile (as entered in your NS profile). The solid blue lines indicate the actual basal amounts set for a given time...so as Loop sets temporary basal rates higher or lower than your scheduled basal rate, the solid blue line will jump above or below your dashed blue line. There is a slight delay (up to 2-5 minutes) in the rendering of basal changes on the blue lines, so if you want the accurate current basal rate, you should check the Loop pill.
"},{"location":"data/nightscout/#predicted-blood-glucose","title":"Predicted Blood Glucose","text":"The purple line to the right of the blood glucose readings is Loop's predicted blood glucose curve. Watching the behavior of that purple line can help you understand why Loop is making decisions regarding high or low temporary basal rates. You can read more on that topic in the Algorithm section of LoopDocs.
"},{"location":"data/nightscout/#loop-pill","title":"L\u200boop pill","text":"The Loop pill is the little display box that when hovered over or clicked, will provide additional information about recent Loop activities and status. The information included is the last time Loop ran, the temporary basal rate set, IOB, and COB. Looking at the Loop pill is a quick method for assessing if your Loop is currently active, as well. The small symbols to the right of the word Loop have a specific meaning. The small horizontal lightning bolt, shown in this example, is an active loop that recently enacted a temporary basal rate.
The SAGE, BAGE, and CAGE pills are for sensor age, pump battery age, and cannula age. These optional pills can track the time since your CGM sensor, pump battery, and pump site were last changed. You can set up custom alerts to remind you when it is time to change the devices, or simply use the visuals to keep track of your particular timing for site/sensor changes.
"},{"location":"data/nightscout/#nightscout-reports","title":"N\u200bightscout Reports","text":"You can access the Reports tab from within your NS settings (the three horizontal lines in the upper right corner of your NS site). There are several types of reports which may be useful to you and/or your health care provider. The report types are listed in tabs at the top of the Reports section, and you can also select the range of dates you'd like any report to cover.
"},{"location":"data/nightscout/#day-to-day-report","title":"Day to Day report","text":"The day-to-day report will show a detailed overlay of boluses, basal rates, carbohydrates, CGM, and treatment notes. If you select the optional check boxes, you can also see information such as the insulin distribution pie graphs shown on the right of the figure below.
"},{"location":"data/nightscout/#daily-stats-report","title":"Daily Stats report","text":"The daily stats report is a pie chart showing the daily breakdown of your low, in-range, and high time in target, as well as other statistical analysis of your BG trends.
"},{"location":"data/nightscout/#distribution-report","title":"Distribution report","text":"The distribution report is a combination of all the individual daily stats reports all into one pie chart for the date range selected.
"},{"location":"data/nightscout/#glucose-percentile-report","title":"Glucose Percentile report","text":"The glucose percentile report will help you see just how consistent your blood glucose is at various times of the day. The average blood glucose is shown as a dark black line in the center of the colored bars. The wider the colored areas spread out from the center black line, the more scattered and variable your blood glucose values at that time have tended to be.
For the graph above, for example, the blood glucose control around 4-6 am is very consistent. The most variable time appears to be near lunchtime. The time of day, when low blood glucose seems to be the most problematic, is between 4-6 pm, which happens to be this person's exercise time. Based on the data in this graph, the person may benefit from setting a higher target about 2 hours before exercise time in order to help with the pattern of low blood glucose that occurs most often during that time.
"},{"location":"data/nightscout/#treatments","title":"Treatments","text":"The treatments report will show a listing of all the temporary basal rates set by Loop, as well as boluses, carbohydrates, site changes, and any other \"treatments\" which have been entered into NS.
"},{"location":"data/overview/","title":"Loop Data","text":"In order to assess your Loop's performance and more specifically your settings, having access to your historical data trends is invaluable. Seeing what your Loop was predicting, what actions were being taken, how blood glucose was reacting, and how insulin needs have been changing is critical when trying to assess any settings changes.
"},{"location":"data/overview/#problem-with-traditional-methods","title":"Problem with traditional methods","text":"As a person with diabetes, you\u2019re probably carrying around a lot of different devices that are holding a lot of different data; a blood glucose meter for your finger sticks, a pump for insulin delivery, a continuous glucose monitor for real-time glucose measurements, phone app for tracking meals, etc. When you go to your endocrinology office, you probably start the process by dropping many of those devices at the front desk to be individually downloaded and then having to pack all of them away 20 minutes later.
Then your clinic staff has the less-than-efficient process of trying to overlay all those different devices into some sort of cohesive strategy for how your diabetes may need some tweaks. Because of Loop use, a clinic currently has to look at separate reports from Medtronic pump, Contour Next Link BG meter, Dexcom CGM/Clarity, and our iPhone Health app, as a typical example. There are also a couple of issues with Loop when using this separate downloads method:
-
Medtronic\u2019s pump gets so clogged up by the numerous temporary basal rate records being recorded that the clinic can only pull about 7 days of data from the pump at most.
-
Clinic basically lacks an overlay of basal rate/bolus actions with blood glucose from Dexcom, which really limits their ability to recommend any settings changes.
-
DIY solutions could fill some of the gaps as far as overall management, but the problem was with the actual gathering of the reports. It\u2019s a little hard for the clinic to bring up reports live time, and sometimes remembering to print them out ahead of time just doesn't happen.
-
Manual notes the various tools are not the most streamlined experience (especially for young kids and teens to manage). This inefficiency can lead to people neglecting to note times of site failure or compression lows, for example. If the health care provider wants additional information about a particular blood glucose event(s), chances are good that there would be no notations to help explain the event without effort. If there was a note, it may be in Nightscout, Dexcom, or in text messages between caregivers...not really centralized and integrated.
The good news is that the community has some very useful tools to help address the shortcomings in the traditional data aggregation for Loop users. There are several tools available to help you and/or your endocrinologist evaluate your Loop data. Specifically, this section covers:
- Health app
- Nightscout
- Tidepool
- Perceptus
Disclosure
The original author of this page, Katie DiSimone \ud83d\udc9a, was employed by Tidepool when this page was initially created. She used Tidepool before becoming an employee. She was not prompted to say anything in particular with regards to this information; it is written based on her own evaluation of Tidepool's services. Specific questions related to Tidepool should be directed to support@tidepool.org.
"},{"location":"data/tidepool/#tidepools-mission","title":"Tidepool's mission","text":"In Tidepool's own words, their commitment to diabetes data and accessibility is pretty clear:
\u200b
Tidepool's mission is to make diabetes data more accessible, actionable and meaningful for people with diabetes, their care teams and researchers.
We believe that connected data leads to better decision-making. Tidepool's free software liberates data from diabetes devices, and places it in context of the real world. Tidepool is designed to help you discover insights and bring context to your diabetes management. And, to help make your data more actionable, we allow you to share your data with anyone you choose: caregivers, clinicians, endocrinologists, friends, researchers \u2013 anyone.
Tidepool has announced that they are embarking on a project to build and support an FDA-regulated version of Loop, to be available in the iOS App Store, intended to work with commercially available insulin pumps and CGMs.
"},{"location":"data/tidepool/#tidepool-data","title":"Tidepool\u00a0 Data","text":"Once you start uploading data into your Tidepool account, you'll see some of the benefits pretty quickly. Tidepool's data presentation is easy on the eyes. Lots of calm colors, logical data layout, quick to access important information...basically a breeze to get your needed data and get on with your analysis.
To use Tidepool's services, the process is pretty standard:
- Create a free Tidepool account to become the warehouse for your diabetes data.
- Upload device data to your account.
Children with type 1 diabetes can have their own data account, and parents can have a caregiver account. Users can also choose to share data with their endocrinology clinic. You can also choose to share your diabetes data anonymously for the support of type 1 diabetes researchers if desired.
"},{"location":"data/tidepool/#how-do-you-upload-data-into-tidepool-account","title":"How do you upload data into Tidepool account?","text":"There are two methods for uploading different data to your Tidepool account:
-
Tidepool Uploader (program on your computer) - Tidepool Uploader supports a large variety of current diabetes devices in the market; Animas, T-slim, Medtronic, Omnipod, Dexcom, and more. The full list of devices is here. To upload the data from these devices, you download the free Tidepool Uploader program, connect your device to the computer according to the instructions provided in the Uploader, and your data is then securely uploaded and stored in your Tidepool account. Most users will perform uploads on a semi-regular basis; weekly, monthly, or prior to each clinic visit.
-
Tidepool Mobile (app on your phone) - Tidepool Mobile is available for both Android and iOS/iPhone users. Tidepool Mobile is a companion app to Tidepool that lets you add notes about meals, exercise, or anything else. See your CGM, pump, and meter data alongside any notes you add. Track your favorite meals and your regular exercise, and learn from what happens.
In summary, the Uploader is for uploading the numeric data from your various diabetes devices, and the Mobile app is for adding contextual information and notes-on-the-go to make your numeric data more useful.
"},{"location":"data/tidepool/#does-tidepool-uploader-work-for-loop-users","title":"Does Tidepool Uploader work for Loop users?","text":"There has been a lack of ability to get Loop data into Tidepool using the Tidepool Uploader. The only pumps compatible with Loop and supported in Tidepool Uploader are the x23/x54 pumps, so users of the x15 and x22 pumps cannot upload their pump data. Furthermore, the Medtronic pump still suffers from the lack of internal storage sufficient to store more than 3-7 days' worth of Looping data.
However, there has been a very exciting development! Tidepool's Mobile iOS app has been updated to upload Apple's HealthKit data into Tidepool. Since Loop uses HealthKit to store all of your insulin, carbohydrate, and CGM data, this means Tidepool can now upload the full suite of available Loop information. So, there's no need to use Tidepool Uploader for Loop users...their Tidepool Mobile app can take care of the needed connections. Read below for lots of details about this important development.
"},{"location":"data/tidepool/#how-can-i-share-my-tidepool-data","title":"How can I share my Tidepool data?","text":"Sharing the data is simple. You can click on your account\u2019s Share option and enter in the email addresses for those that you want to share with. Those people will need a Tidepool account. If they don\u2019t have one currently, they will follow easy prompts for an account setup after they\u2019ve received your share invitation. Clinics using Tidepool will have a Tidepool account email that you can add to your account, enabling the clinic to easily view your data. You can also remove access for anyone with a simple click.
"},{"location":"data/tidepool/#viewing-your-tidepool-data","title":"Viewing your Tidepool data","text":"There are two distinct viewing options for your Tidepool data, and they are not identical. You can either view your data:
- Using a desktop computer using the Chrome browser and logging into your Tidepool account, or
-
Using your Tidepool Mobile app AND associated notes.
Supported Desktop Browser
Currently, only the Chrome desktop browser is compatible with Tidepool. You will not be able to use Safari or other browsers to view the data, nor will you be able to use the Chrome application on a phone. If you want to see your data on a computer, you need to use the Chrome browser.
Viewing Data on your Phone
If you want to see your data on your phone, you must use the Tidepool Mobile app and use an associated note to view the data surrounding the note's time frame. Mobile data viewing is not a live stream, but rather available as \"bookmarked\" time periods anchored by notes.
You will see your Loop's temporary basals, CGM readings, boluses, notes, carbohydrates, and various metrics about your data distribution. If you separately load your blood glucose meter or any other supported device to Tidepool, those will also overlay.
"},{"location":"data/tidepool/#what-loop-data-will-you-see-in-the-tidepool-mobile-app","title":"What Loop data will you see in the Tidepool Mobile app?","text":"The Tidepool Mobile app is not a live-viewing app for looping data. For people coming from Nightscout, this may be a bit confusing but realize the intended purpose of the Tidepool Mobile app isn\u2019t live-viewing. It is the place where you can (1) upload/sync HealthKit data, (2) easily add/edit/delete notes to the data set, and (3) search for notes and view Loop data surrounding that note.
In fact, you will basically see NO data in the Tidepool Mobile app unless you have Tidepool data uploaded and notes are added. Once you add a note, you are basically placing a bookmark on the data set. You will be able to click on the note and see 7 hours of old data before the note, and then the note will continue to collect 7 hours of data to display after the note. So, technically, you\u2019ll be able to refresh the app\u2019s view to see current data for approximately 7 hours after a note is placed.
For example, here\u2019s a sample note+data set below from my Tidepool Mobile app. Over the last couple of hours, my daughter noticed that she was staying above target (unusual for her on Loop with the meal she had) for quite a while. She had given a couple of small corrections (see the two 1-unit correction boluses?) without result. She started her secondary troubleshooting\u2026if it\u2019s not the food, maybe it\u2019s the infusion site? She realized it has been 4.5 days since changing her site. She changed the site and logged a note using the Tidepool Mobile app. That note appears in the Tidepool Mobile application, on my phone. It also shows up on her Tidepool's data, for her endocrinologist to see too, and we can refresh the view to see how blood glucose trends for the next 7 hours after the site change.
"},{"location":"data/tidepool/#tidepools-mobile-app-for-iosiphone","title":"Tidepool's Mobile app for iOS/iPhone","text":"As part of Tidepool Loop development, Tidepool updated its Tidepool Mobile iOS app to integrate with Loop user's data stored in Apple's HealthKit database. This update also supports the Jaeb Center observational study of Loop.
"},{"location":"data/tidepool/#how-does-the-loop-data-upload-work","title":"How does the Loop data upload work?","text":"The Tidepool Mobile iOS app will sync Loop's diabetes-related HealthKit data into your Tidepool account. The app will continuously upload that data so long as the Tidepool Mobile app is open, even if it is only open in the background. That data will then be viewable in two places: on the Tidepool Mobile app itself when using notes, or in your Tidepool account using the desktop Chrome browser (note: the two options are not identical viewing platforms, see discussion above). This means you will not have to plug any of your devices into a USB cable in order to upload your information to Tidepool.
"},{"location":"data/tidepool/#is-the-tidepool-mobile-app-a-replacement-for-nightscout","title":"Is the Tidepool Mobile app a replacement for Nightscout?","text":"Nope. This was not designed nor intended to be a replacement for your Nightscout site. They complement each other, rather than compete. This new upload of Loop data will allow you and your clinics to have a powerful tool to analyze Loop data through an easy-to-use, shared hub.
"},{"location":"data/tidepool/#are-there-any-known-issues-with-loop-and-tidepool-mobile-app","title":"Are there any known issues with Loop and Tidepool Mobile app?","text":"The settings for Loop do not show up in Device Settings in Tidepool Web. This is because this info can't be uploaded from Apple Health. If you used Tidepool Web for your Medtronic or Omnipod before, you will still see the pump settings from before you started looping.
You mean, what OTHER cool thing can the Tidepool Mobile application do besides automatically uploading the data from Loop? Tidepool Mobile also provides an easily searchable log of meal boluses. If you are still learning new meals in Loop\u2026how much to bolus, how long of a carb absorption\u2026these notes are searchable and super easy to add. Learning how to bolus for that Tofu Breakfast Burrito? Simply record a note of how you bolused for it. If you want to, come back afterwards and leave yourself some suggestions for the next time to try. This searchable information can also help teens learn and exercise independent skills. If they aren\u2019t sure of how to bolus for a meal, this could give them easy tips from past success without necessarily having to stop and ask a parent. As much as a parent might scoff at the idea of a kid looking up a meal, if the alternative is asking a parent\u2026that might be all the motivation it takes. How about co-parenting? Want to leave a note that another parent or caregiver can look up? \"How were the last Chicken McNuggets bolused?\" or \"When was the last site change?\" can easily be tracked and retrieved with notes.
For an easy example, search for the word burrito (doesn\u2019t have to be a hashtag), and any notes with the word \u201cburrito\u201d will be available for review, as well as any added comments.
Hormones can also be easily tracked with notes. What day-of-the-month and how did I change the basals? Looking to find patterns in those female hormones? This could be a really slick tracking tool to easily log periods of insulin resistance and what part of the cycle they are occurring at.
"},{"location":"data/tidepool/#is-the-tidepool-mobile-app-only-for-loop-users","title":"Is the Tidepool Mobile app only for Loop users?","text":"The Tidepool Mobile application uploads diabetes-related HealthKit data regardless of the source. Loop users store their data in HealthKit, so this is a nice fit. Other diabetes devices (e.g., OneDrop BG meter) and apps (e.g., Spike and Dexcom) also store their data in HealthKit. Some people even manually enter their diabetes data into HealthKit. For all those uses, the Tidepool Mobile application will upload the HealthKit data. OpenAPS does not store its data in HealthKit, so this will not upload OpenAPS-related data. However, if OpenAPS users are using Spike app, some integration of information through Nightscout is possible. Currently, OpenAPS SMBs, boluses, and carbohydrates can be uploaded to HealthKit for OpenAPS+Spike+Nightscout+iPhone users. Spike does plan to add temporary basal integration to HealthKit at some time in the future.
"},{"location":"data/tidepool/#how-can-you-get-the-tidepool-mobile-app","title":"How can you get the Tidepool Mobile app?","text":"Tidepool Mobile is available in the iOS App Store on your Loop device.
"},{"location":"how-to/bolus/","title":"Extended or Combo Bolus with Loop","text":"The majority of meals have most of their blood glucose impact within 2-3 hours after eating. Complex carbohydrates are slowed by their fat and protein content and can lead to extended time of blood glucose impacts. Many people with type 1 are familiar with the late blood glucose spikes from Chinese food, pasta, pizza or burritos. This extended blood glucose impact can be tricky to properly bolus for in traditional one-bolus insulin delivery.
In traditional multiple daily injection therapy, these complex meals may require additional insulin boluses to help control long, slow absorption meals. The trick is to try to time the second bolus at a time when blood glucose is starting to rise and it can be tricky to estimate how much to give to control the tail end of those slow meals.
If you're using traditional pump therapy, one technique is to use an extended bolus or dual wave bolus for your complex meal's insulin. In this method, a user give a portion of the insulin up front as an initial bolus and the remainder of the insulin is delivered as a psuedo high temp basal for a user-set duration of time. For example, you may give 5 units of bolus up-front and then \"extend\" 4 units of insulin over the next 60 minutes. In this way, you are providing a more complex bolus to help match the timing of the meal's blood glucose impacts; avoiding early low blood glucose and addressing later high blood glucose.
The transition to Loop use may be confusing at first for these meals since you cannot use an extended/combo bolus and simultaneously have Loop set temporary basals automatically. The good news is that Loop has a bolus calculator that has the ability to emulate an extended bolus situation...and it's implemented with the pizza icon (or any custom carbohydrate absorption that is set longer than 4 hours).
"},{"location":"how-to/bolus/#pizza-icon","title":"Pizza Icon","text":""},{"location":"how-to/bolus/#example-long-absorption-time-bolus","title":"Example long absorption time bolus","text":"For an example of Loop's bolus adjustments using carbohydrate absorption time, let's take a look at an example meal. This is an example of a long absorption complex carbohydrate meal. This is a mushroom (arborio) risotto dish with heavy cream and cheese ingredients. While some white rice can be fairly quick acting, after several times eating this dish, the family has noticed that the meal tends to have a longer duration of impact on blood glucose. Using a \"taco\" icon (3 hours absorption) was causing slight low blood glucose soon after eating the meal. Therefore, they have been using the pizza icon to enter the meal's carbohydrate absorption time.
The initial meal entry was 70g at a \"pizza\" icon aborption time (4 hours). Based on carbohydrate ratio of 8 g/U, the initial bowl of risotto at 60g should have been a bolus of 7.5 units. Loop recommended 5.3 units, or about 70% of the total bolus that would be needed to cover the total carbohydrates. Loop recommended the lower upfront bolus because a full bolus would have overwhelmed the slow absorption of carbohydrates, and the likelihood would be a low blood glucose shortly after eating.
As the meal was being absorped, Loop was tracking the carbohydrates still remaining to be absorped and expecting that blood glucose values would be rising soon (knowing that there was still insulin needing to be delivered once safely passed the near-meal low blood glucose potential). Loop would have provided the additional insulin via high temporary basals after seeing blood glucose impacts which indicated the potential for a low blood glucose had passed.
The Loop user then had a second, smaller bowl of risotto about 90 minutes later, and entered 30g at 4 hours absorption again. Notice this bowl of risotto had a bolus recommendation much different than the original bowl. The second bowl had a recommendation of 5 units, much greater amount of bolus relative to the amount of carbs entered than the first bowl had received, and more than the carbohydrate ratio alone would provide (3.75 units). Why the \"extra\" 1.25 units? Because Loop was including some extra bolus amount to cover what it predicted could safely be provided from the amount not given in the original bowl's bolus (the original bolus was approximately 2.2 units short of carbohydrate ratio alone recommendation). If the user had not had the second bowl, Loop would have been providing high temporary basals as soon as blood glucose had exceeded the correction range. And in fact, you can see that Loop still provided the remaining insulin via high temporary basals as blood glucose rose after the second bowl, in effect making up the small remaining difference.
"},{"location":"how-to/bolus/#low-carbhigh-fat-or-keto-diets","title":"Low carb/High Fat or Keto diets","text":"The example meal above, while relatively high carbohydrate, also helps illustrate how Loop can be used to bolus effectively for low carb/high fat (LC/HF) or Keto diets. Those diets tend to have low glycemic index food with a relatively long blood glucose impact. Said another way, they don't spike blood sugar as much, but often need additional insulin after the meal was eaten to account for the slow conversion of protein to glucose.
To account for those dietary differences, there are two useful strategies:
- convert some of the protein and fat to \"equivalent carbohydrates\" and
- extend the duration of those carbohydrates using a pizza icon or even longer, depending on the person/food.
Most LC/HF or Keto users will convert a portion of their fat and protein content into an equivent carbohydrate content. So while an example meal might only have 5g of carbohydrates based on nutritional labels, they may convert 25% of the 20g fat and 50% of the 20g protein grams into an additional 15g of \"equivalent carbs\" for bolusing purposes. The percentages that people use to convert fat and protein will usually be a bit of trial-and-error, but there are some published articles (here, here, here, and here) that may be helpful starting points, if you are interested.
"},{"location":"how-to/cgm/","title":"CGM data","text":"Obviously, without CGM data your Loop will not function. With the stability that Looping brings, many people are looking for ways to minimize their Loop downtime due to avoidable factors. Factors that make CGMs a potential problem include sensor failure, transmitter signal loss, poor calibrations, and the warmup period.
"},{"location":"how-to/cgm/#sensor-warmup","title":"Sensor warmup","text":"One of the obvious downtimes is during new sensor warmup. Unfortunately, the only way to avoid the 2-hour warmup (in the case of Dexcom's system) is to overlap your sensor sessions and this just isn't feasible for most people.
- Insurance doesn't provide adequate supplies to overlap sensor supplies; you would need two active transmitters and enough sensors to allow for overlap of use.
- Sometimes sensors fail before a session was due to expire, making a planned overlap impossible.
- Some people just don't want to wear two sensors at once.
Since warmups are an important part of stabilizing CGM values and decreasing sensor noise, they are just going to continue to be a no-Loop time period for the foreseeable future. Most people simply cancel a currently running temporary basal, or let it run its course, when they start a new sensor session. For the two hours of warmup, the Loop will revert to the basal schedule programmed into the pump.
Some people have opted to keep looping during sensor warmups by manually entering blood glucose checks through the Apple Health app for Loop to read. While technically Loop can use those values, the entries would have to be sufficiently frequent to be meaningful. If a person is willing to test 8-12 times in a two hour span, then looping through sensor warmup could be feasible through this method.
"},{"location":"how-to/cgm/#sensor-noise","title":"Sensor noise","text":"Depending on which CGM device you use, you may observe a sweet spot of data accuracy in the middle of your sensor's session but have sensor \"noise\" at the beginning and end of the session. Generally speaking, Loop does fairly well despite these times of sensor noise, although having a reasonable maximum basal rate does help mitigate the risk due to a noisy sensor.
The figure below shows an example of a new G6 sensor for approximately its first 21 hours of use.
As you can see on the left, the red dot was a finger stick blood glucose check and it nicely aligned with the CGM value when the session first started to provide live data. However, between that first value and approximately noon the next day, there was a considerable amount of jumpiness to the data (as well as some suspected compression lows or extreme sensor jumpiness in the first 6 hours). The CGM noise immediately decreased after the first calibration was provided at the noon-time meal bolus. (Personally, we give the G6 one calibration point about 8-12 hours after a session starts because we have noticed that it helps settle down the noise quickly. This is not a part of the standard Dexcom protocols.)
We similarly see an increase in sensor noise at the end of a sensor's useful life. The figure below is a G6 sensor at 9.5 days, but your end of sensor symptoms may occur differently and at longer/shorter days. Much of the specifics regarding useful sensor life seems to depend on the user's body chemistry and sensor insertion.
There is a marked increase in sensor noise and scatter, as well as several periods of sensor error that lead to lost CGM data. We opted to pull this sensor just shy of the 10 days due to this noise. However, as you can see, the Loop was still doing an aequate job controling overall blood glucose fluctuations despite the erratic CGM data.
"},{"location":"how-to/cgm/#compression-lows","title":"Compression lows","text":"A frequent question from people before starting Loop is \"How does Loop deal with compression lows?\" If you aren't familiar with compression lows, they are false low blood glucose alarms caused by sustained pressure on the sensor area. In effect, the phenomenom is much like resting on an arm for too long and causing it to fall asleep from poor blood flow.
The figure above showing an example of a compression low. CGM data shows blood glucose dropping low, but finger checks on a meter would confirm that the CGM data is falsely low. Often, Dexcom G5 and G6 will stop providing CGM values for awhile when their algorithm detects a suspected compression low. Once the person rolls off the sensor area and blood starts flowing well again, the CGM values come back online to a more reasonable tracking again.
Generally speaking, Loop will deal with compression lows just fine. Loop will suspend for a short time during the low blood glucose values, and then recover with temporary high temporary basals to make up for the active insulin it had missed as result of the compression low. No special actions need to be taken for compression lows.
"},{"location":"how-to/cgm/#calibration-jumps","title":"Calibration jumps","text":"It is also worth noting that Loop will not calculate blood glucose momentum in instances where CGM data is not continuous (i.e., must have at least three continuous CGM readings to draw the best-fit straight line trend). It also will not calculate blood glucose momentum when the last three CGM readings contain any calibration points, as those may not be representative of true blood glucose momentum trends. Therefore, calibration jumps or missing data will not adversely affect the insulin dosing based on blood glucose momentum.
"},{"location":"how-to/cgm/#loss-of-cgm-data","title":"Loss of CGM data","text":"Transmitter failure, sensor failure, or other equipment failures can lead to a failure in CGM data. In those cases, Loop will revert back to the pump settings after any currently running temporary basal finishes or is manually canceled by the user.
"},{"location":"how-to/cgm/#new-transmitter-or-reset-transmitter","title":"New Transmitter or Reset Transmitter","text":"-
If you begin a new transmitter, don't forget to update your transmitter ID in your Loop settings.
-
If you are going to reset a transmitter, don't forget to remove the transmitter ID from Loop settings during the reset and re-pairing of the transmitter. Do not add the transmitter ID back into Loop settings until you've completed the reset process and re-paired with the Dexcom app successfully.
For Medtronic pump users, you will have times where you need to disconnect from your insulin infusion site. Showers, some sports activities, swimming, and medical exams are some of the common reasons when disconnections are unavoidable.
The simplest way to deal with periods of disconnect is to suspend the Medtronic pump so that Loop knows no insulin was being delivered while you were disconnected. This will let Loop know that there is missing basal insulin that may need to be considered as part of the blood glucose prediction. Typically when disconnected from insulin delivery and negative insulin on board has developed, blood glucose will start to rise so Loop will expect to need to make up the missed insulin from disconnections.
Some users are prone to forget to resume insulin delivery after suspending for disconnects. One option is to use Siri by simply saying \"Hey Siri, set a timer for 30 minutes\" and that can serve as a reminder alarm to resume the pump after the shower is over. Other users may want to consider canceling any running temporary basal, setting a 30 minute zero temporary basal, and then setting Loop to open loop. This will keep IOB fairly accurate while at the same time providing an automatic restart of normal basals if the user forgets to resume delivery. The user will then have to close loop again when ready to resume looping. Find what option works for you (or your child) the best; there are a number of ways to deal with disconnects.
For disconnects where the disconnected time coincides with exercise that may increase insulin sensitivity (like swimming), a good strategy after reconnecting can be to set a higher temporary override target (aka workout mode) for some time after reconnecting and resuming insulin delivery. This will help Loop be a little gentler with high temporary basals during the time while the user is still experiencing increased insulin sensitivity after exercise. The length of time to set that override target can be a bit of trial-and-error as it will depend on the person and type of exercise.
"},{"location":"how-to/endo/","title":"Endocrinologist and Loop","text":"There is a wide range of comfort within the health care profession for DIY projects and type 1 diabetes. Loop is no different; some health care providers are comfortable and some are distrusting. As with all things health care related, honesty with your health care provider is paramount so that the best decisions can be made.
Since using Loop often requires people to secure access to pump supplies that they may not have otherwise have, often times the first conversation about Looping with an endocrinologist will be when trying to get prescriptions for Medtronic or Omnipod supplies. The best approach is to start with objective information. Find out if the clinic has other patients already Looping and whether or not they have concerns from what they have seen. Provide links to the LoopDocs site and this site so that your endocrinologist or health care provider can read about the system's algorithm and operations.
Most often the important aspects that need discussion are:
- Loop will revert to normal pump use, within a maximum of 30 minutes, in the event of Loop failure. What happens during failure of the system is the most common concern your health care provider will likely have when initially learning about the system.
- Loop depends on your basal rate, ISF, and carbohydrate ratio settings, just like traditional pump therapy. The conversations will not need to change regarding the evaluation and adequacy of those settings.
- Duration of insulin action is automatically set to 6-hours for the rapid-acting insulin models built into Loop, matching published data for those insulins. The settings in the pump for duration of insulin action are not used by Loop.
- Reasonable values for the suspend threshold and maximum basal rate should be discussed. These two settings are not part of normal pump therapy and should be discussed with your endocrinologist. Typically to start Looping, setting the maximum basal rate not much above your highest scheduled basal rate is an excellent way to get used to Loop and safely double-check that your settings (ISF, carbohydrate ratio, and basal schedule) are reasonably accurate. As you gain experience and confirm your settings, raising the maximum basal rate can help Loop become more effective to correct rising blood glucose more quickly.
- Develop a plan for how you will be providing Loop data to your health care provider in advance. Ask what information they would like to review so that you can have reports ready at each appointment. The usual downlaod of a Medtronic pump or Omnipod PDM will not be sufficient for many clinics once you start Loop. Instead, you may wish to discuss Nightscout or Tidepool data reports to provide the information those downloads would have otherwise provided.
- Develop a plan for how you will deal with pump failure. Many health care professionals will cite the age and out-of-warranty aspects as concerns for Loop users. Having a clear plan, in advance, for how to deal with possible pump failure will help with that concern. The two most obvious steps would be to have a backup pump available and an active prescription for long-acting insulin. A gentle reminder to the clinic that even in-warranty pumps can fail also helps frame the conversation regarding the need to always be prepared, looping or not.
Hands-down, exercise is the most commonly discussed topic of long-term Loop users. Exercise is highly specific to an individual and its impact on blood sugar is not always equal. For example, many people find anaerobic exercise such as weight lifting will raise blood glucose, and aerobic exercise such as running will lower blood glucose. The degree to which impacts are observed also depends on the insulin and carbohydrates on board going into the exercise. Blood glucose response in competition can be quite different than a regular training session due to the added adrenaline. Exercising with T1D does require extra preparation, but with preparation, things should go much easier.
"},{"location":"how-to/exercise/#exercise-basics","title":"Exercise Basics","text":"A 2013 JDRF survey found that 36 percent of T1D patients, caregivers and healthcare professionals view exercise as a challenge and would like to learn more about exercising safely. The JDRF PEAK (T!D Performance in Exercise and Knowledge) education curriculum explores the environmental, dietary, physiological and psychosocial elements that impact physical activity with T1D (source). As part of JDRF's PEAK resources, they have published a helpful article about exercise and T1D for athletes. Below is an excerpt from that article. There are a couple key considerations (highlighted in orange) that be different in Loop use. Those considerations are discussed further below.
With great exercise, comes great responsibility
It goes both ways: exercise affects glucose control; glucose control affects exercise. Consistent physical activity produces consistent insulin sensitivity, which helps the body process glucose more evenly (avoid spikes), maybe even improving HbA1c.
Some considerations:
- Risk for hypoglycemia can be reduced by avoiding too much insulin on board (IOB) during and after exercise.
- Continuous glucose monitors (CGMs) or regular blood glucose testing provides critical information to preemptively minimize blood-sugar highs and lows.
- Plan ahead! Reduce insulin and add carbs for aerobic (longer duration, lower intensity) exercise, which causes blood sugar to drop, or increase for anaerobic (short duration, high intensity), which may cause spikes.
- After most exercise, insulin sensitivity is higher for 24 hours. Carbohydrates needed to replace liver glycogen levels must be balanced with decreased insulin levels.
Intensity. Speed. Environment. Duration\u2026 all affect the body\u2019s reaction
There are three types of exercise:
- Aerobic
- Aerobic exercise is usually continuous, light-to moderate intensity exercise and includes activities like running, walking, long-distance swimming, biking or tennis. Aerobic exercise typically tends to lower blood glucose levels.
- Anaerobic
- Anaerobic exercise is generally shorter in duration, of maximal to super max intensity and includes activities like sprinting, gymnastic, ice hockey, or weight training. Anaerobic exercise typically tends to raise blood glucose levels.
- Mixed
- Mixed exercise is combination of the aerobic and anaerobic activity such as basketball or soccer. Managing blood glucose levels with mixed exercise is difficult, but using a tool like a continuous glucose monitor can help greatly.
Glucose levels during sports affect performance in many ways: strength, stamina, speed, agility, flexibility, safety and mental sharpness.
- Insulin resistance, which is when the body requires more insulin to process glucose, is a problem. Exercise combats this by increasing insulin sensitivity and glucose uptake by muscles. Muscles are better able to absorb glucose when contracted.
- Exercise thwarts postprandial (after-meal) hyperglycemia by slowing carbohydrate absorption, increasing glucose utilization and accelerating insulin action.
- Practice vs game day \u2013 each can have dramatically different effects on your T1D. Adrenaline and stress both raise blood glucose levels and are typically present on the day of your game or event.
Balancing Act: Finding your blood glucose zone and peak fitness zone
Caution: Delayed Onset Hypoglycemia can occur typically 6 to 12 hours (sometimes 24 hours) after exhaustive exercise, due to replenishment of muscle glycogen stores and enhances insulin sensitivity. It is possible to prevent this by:
- keeping records of your fitness regimen and the effects it has on your body
- using a CGM or frequent blood testing to monitor your body\u2019s response to activity
- decreasing basal insulin or meal boluses after activity
- consuming slow-acting carbohydrates after activity
Similar recording and monitoring tactics can be used to prevent hyperglycemia (blood-sugar highs). Some other options include:
- bolusing (at a half dose) 30 to 60 minutes prior to activity to offset rise of blood glucose, delayed bolusing or adding post-workout cool-down (depending on conditions of rise)
- incorporating relaxation, breathing, visualization tools to address pre-workout emotional stress
- limiting pump disconnection time
- hydrating
- administering rapid-acting insulin (only under certain conditions)
With all of that background laid out, your Loop can help with some of the techniques described above. Highlighted in orange above are parts of those recommendations and considerations that we should explore a little more for Loop in particular:
Risk for hypoglycemia can be reduced by avoiding too much insulin on board (IOB) during and after exercise. -- In Loop use, we can accomplish this through setting a workout target far enough ahead of the exercise to help decrease basal insulin on board going into exercise. Typically this means setting a workout target at least 1, if not 2, hours prior to exercise. Setting a workout target just as you begin a workout will not allow sufficient time to decrease IOB ahead of an aerobic workout. Loop workout targets are usually 50 mg/dL higher than your usual correction target, although that is a guide. Practice and observation of your typical exercise needs will help refine your optimal settings. And while not Loop specific, minimize heavy carb meals in the 3 hours before intense exercise that would require large boluses. If eating during that time, consider partial boluses and slower, lower carbohydrate snacks. If bolusing while workout targets are enacted, Loop will automatically recommend a smaller bolus than you would get with your normal correction target.
After most exercise, insulin sensitivity is higher for up to 24 hours. Carbohydrates needed to replace liver glycogen levels must be balanced with decreased insulin levels.-- For Loop use, this period of increased insulin sensitivity may require continuing or setting a workout target for some period of time after exercise is finished. This will help manage increased sensitivity by providing smaller boluses and less basal delivery to prevent hypoglycemia that could result from unadjusted settings under traditional therapy. The length of time to run a continued post-workout target will depend on the duration and intensity of exercise.
"},{"location":"how-to/high-bg/","title":"Stuck on High BG","text":"Safety consideration
If the continued delivery of corrective insulin fails to bring down high blood glucose, remember to follow your doctor's recommendations for ketone testing, hydration, and/or more frequent blood glucose monitoring. Check for equipment failure (infected pump site? blocked cannula?). Try fresh insulin. Contact your doctor immediately if you are experiencing sustained high blood glucose despite continued insulin delivery and corrections.
\"How do I deal with high blood glucose that seems to be stuck?\" When you and your Loop are disagreeing about the effectiveness of high temporary basals to bring down blood glucoose, take a deep breath. Unfortunately, most of the time our disagreement is rooted in the impatience that the pump's insulin works so dang slow. We get tired of waiting another 4-6 hours until blood glucose will be back in range after a correction. So before we start overriding Loop's predictions with manual actions, it may be useful to ask yourself some questions first to help guide potential corrective actions:
- Is this temporary or has this been going on for a long time?
- Is this consistently only after meals?
- Have you open loop tested your settings?
- Are you sick or stressed?
- Are you hormonal?
- Are you taking medication?
- Could this be caused by site failure or bad insulin?
Based on the answers to those questions, you may want to choose on of the potential solutions for delivering extra insulin to help:
- Setting lower temporary blood glucose targets
- Providing the \"pending\" insulin upfront via correction bolus
- Adjusting meal entries
- Fake carbohydrates
- Priming in insulin
- Open loop and correct manually
- Adjust settings
If you to believe the high blood glucose is a short-term issue, due to acute stress or anaerobic exercise for example, then using temporary lower override targets will help Loop be more aggressive at providing additional insulin. The pre-meal target would be a quick, easy implementation of this tool.
This technique, and the delivery pending insulin as a bolus (described below), are good to help control the urge to \"rage\" bolusing which often yields an overcorrection and subsequent low blood glucose. These two techniques used in combination are often successful for most short-term situations or meals that are just taking a long time to \"burn out\".
"},{"location":"how-to/high-bg/#pending-insulin-as-bolus","title":"Pending insulin as bolus","text":"To help correct a high blood glucose, Loop will apply a high temporary basal set to deliver over 30 minutes time. If you are stuck on a high blood glucose and want to speed up the correction, you can try to \"hurry\" along the correction by opening Loop's bolus tool. Within the information screen, you will see a \"pending insulin\" amount. That amount represents the amount of corrective insulin Loop plans on delivering with the currently running 30 minute temporary basal. You can choose to manually enter in that amount on the bolus deliver line as an upfront correction bolus, rather than waiting for it to be delivered over the next 30 minutes. This generally represents a pretty easy, safe way to help reasonably speed up a correction.
The safety consideration for this is that you do not want to unintentionally stack insulin by using this technique repeatedly in a short period of time. Wait a reasonable amount of time, such as an hour or two, to give insulin a chance to start to affect blood glucose. If the correction still hasn't helped, you may want to consider another technique to help bring down blood glucose.
"},{"location":"how-to/high-bg/#adjusting-meal-entries","title":"Adjusting meal entries","text":"Carbohydrate, fat, and protein estimations for bolusing is frequently a guessing game...and we've all guessed wrong before. If your blood glucose is remaining high due to a food guess gone wrong, you can definitely go back in time and adjust your meal's carbohydrate entry to more accurately reflect what you suspect it should have been. By adjusting the original entry, you'll help Loop know that the blood glucose effects it needs to control are greater (total carb entry increase) and/or longer (carbohydrate absorption time extension). Often you'll find Loop will offer an additional bolus immediately after you save your adjusted carbohydrate entry and/or the temporary basal adjustments will be more aggressive.
"},{"location":"how-to/high-bg/#fake-carbohydrates","title":"Fake Carbohydrates","text":"Some people use \"fake\" carbohydrate entries to push Loop to be more aggressive with correcting a high blood glucose. This can be effective for hyperglycemia due to short term causes, but is not the recommended way of repeatedly dealing with high blood glucose. If you find yourself regularly using fake carbohydrates often, you may want to consider whether one of your underlying settings needs adjusting.
"},{"location":"how-to/high-bg/#prime-insulin","title":"Prime insulin","text":"If your Preferred Data Source is left on the default \"Event History\", Loop will not count insulin delivered from \"prime\" commands in the Medtronic pump. This means that any insulin you deliver through your infusion set while using the prime delivery will not count towards active insulin or insulin on board (IOB).
If you think that you are being affected by a short term influence that will last for at least 4-6 hours, a conservative amount of primed insulin may help control high blood sugar more aggressively than Loop would have otherwise provided during the stressful event.
If you choose to deliver insulin via prime command, remember that your active insulin or IOB will not be accurate for 6 hours after the delivery. Caution should be taken using this method because if the short term stress suddenly ends and blood glucose begins to drop again, the drop may be more pronounced than Loop predicts due to the untracked primed insulin.
Note: For Omnipod users, Loop does not allow you to prime insulin in.
"},{"location":"how-to/high-bg/#open-loop-and-correct-manually","title":"Open loop and correct manually","text":"Don't forget that you can always go old-school and open your Loop until the cause of the high blood glucose is sorted out and you can get back to normal operations. Site failures are an especially good example of when open looping plus manual correction may be a less frustrating way of dealing with Loop during times when prediction just isn't meeting reality. Give IOB a chance to get back to reality by letting open loop run for 4-6 hours and you'll come back after site failure with a more functional Loop prediction.
"},{"location":"how-to/high-bg/#adjust-settings","title":"Adjust settings","text":"Finally, if your underlying cause doesn't appear to be short term (no site failures, no stressful event, no illness) or food related (no recent meals, happens even overnight), then you may need to adjust your underlying settings to help Loop get a more accurate prediction curve working for you.
Long duration \"stuck\" high blood glucose most often means you'll need to increase your underlying basal rate(s). If Loop's high temporary basals aren't making meaningful progress towards your correction range within 4-6 hours, and you don't suspect food, stress, or site failure, then a basal increase is likely a good place to start. Many people start with a 10% basal adjustment to begin with and wait to see the effectiveness before making any additional adjustments.
What kind of situations are most likely to need this adjustment? Monthly hormonal cycles and steroids are good examples of when underlying basal rate adjustments for a few days could help tremendously until things return to normal.
Consult your health care provider if you are at all uncomfortable self-adjusting basal rates.
"},{"location":"how-to/high-bg/#maximum-basal-rate","title":"Maximum basal rate","text":"It is worth mentioning one adjustment that will not be helpful in these situations. Increasing your maximum basal rate is very unlikely to resolve a stuck high blood glucose. Stuck high blood glucose is the result of Loop predicting that the insulin on board is sufficient to cover the needed correction. The issue is almost never that Loop is being limited by a maximum basal rate, but rather that Loop is predicting that the insulin it has been delivering for awhile is adequate. In other words, Loop doesn't even think that it needs to reach that maximum basal rate. The better adjustment would be to increase your scheduled basal rates, and not the maximum basal rate listed in Delivery Limits.
"},{"location":"how-to/injections/","title":"Test Your Settings","text":"Open Looping Fine-Tuning Settings Book settings
"},{"location":"how-to/iob/","title":"Using Morning IOB","text":"One of the easiest habits to help check your settings is to simply check-in on your morning blood glucose and active insulin (IOB). Check if you are (1) above, below, or within target range and (2) carrying significant positive or negative IOB in the morning before you start walking around (if you are prone to dawn phenomenon) or eating. Since mornings are free from food influence, these are almost like waking up from a mini-basal testing every overnight.
This Looped group post started the conversation:
I've been having more lows recently than I would like. Any help here would be really really appreciated.
I took a screenshot of an example situation...LOOP is predicting that I'll eventually rise up to 124 and is therefore giving me my full regular basal rate EVEN THOUGH im currently way below my desired goal range. I have a better knowledge of my diabetes than LOOP does (sorry if that is heresy around here) and I would not like LOOP to be giving me basal right now OR ELSE I'LL GO LOW.
QUESTION: Which parameter must I change on my loop settings in order to essentially tell LOOP to relax on the basal rates??
To illustrate how to use morning IOB and blood glucose to check your basal rates, we will walk through a series of screenshots from a recent Loop experience. When the Looper woke up, she was below her 95 mg/dL target, but relatively steady. At about 8:48am, Nightscout showed negative IOB of -1.33 units, blood glucose of 90 mg/dL and slightly rising but still below a target of 95 mg/dL. (Note: this situation is similar to what the original Facebook poster above was describLooped2](ing.)
Looking back on the night, blood glucose was pretty much below target the whole night and her scheduled basal delivery was turning off/on in an alternating pattern (the blue pattern area). Blood glucose wasn't concerning nor were alarms going off. However, this combination of data is a great indicator that basal rates need to be decreased. The tendency could be to let Loop just keep plodding along like this, but it can lead to certain less than desirable stress points on the algorithm.
Firstly, if the user were to bolus while carrying a lot of negative iob, they would get more of a bolus than otherwise needed...yes, Loop will likely suspend again as soon the meal began because blood glucose spike would be well contained with the extra bolus, but the system would just have a harder time that it should with well tuned settings.
Secondly, if the user were to go above their correction range, they would begin to get high temporary basals for what is an inaccurate amount of negative iob (because basals really didn't need to be this high). And high temporary basals at this point in time and with these settings would be too aggressive. (If the maximum basal is set really high, the problem compounds with bad underlying settings. This is why it is a good idea to keep your maximum basal relatively low when you first start Loop and are testing your settings.)
Useful Summary
You probably need to lower your overnight basal rates if you wake up:
- Carrying negative IOB, and
- Blood glucose is still below or just below target, and
- Overnight basals were turning off/on.
So, what will you see after you adjust your basal lower? You will see some of the negative IOB disappear and, as a result, the predicted blood glucose curve will not so such an aggressive prediction for a blood glucose rebound.
In this example, the user's basal rates were lowered across the board. All of the basal rates went down about 0.2 U/hr. It works out to about a 15% decrease, and is pretty standard adjustment for this user due to her monthly hormone changes.
The adjustment to basal rates caused Loop to reevaluate its math. After adjustment, Loop now has a negative IOB of -0.55 units vs the previous value of -1.33 units. This is a more reasonable given the situation. The prediction line with a -0.55 units IOB was not predicting nearly as aggressive of a \"rebound\" blood glucose rise.
Why did IOB and prediction change?
The visualization below might be easier. The red line is how loop knew things to be before basal rates were adjusted lower. The purple line is how Loop viewed basal schedule after the adjustment lower.
The green arrows highlight parts of the graph that are recalculated by Loop when the basal schedule was adjusted lower. Instead of Loop thinking those were NEEDED basals (aka, conforming to the old basal schedule), now Loop perceives those as \"extra basal\" insulin deliveries. Now those insulin deliveries are sitting ABOVE my scheduled basal dotted line in Nightscout. They are actually instances of positive IOB and therefore Loop is now correctly getting closer to realizing that perhaps all of that extra wasn't needed.
The negative/positive IOB plus a quick glance at overnight Loop actions/blood glucose relative to targets is a quick easy check on overnight basals.
If one was really exacting, you could adjust basals until a number closer to 0 IOB. In practical use though, getting roughly closer is usually helpful enough and smaller adjustments could be made later if still needed. The graph below shows the results several hours after the basals were decreased. As you can see, looking pretty decent.
The confirmation that adjustments were on-track would also be provided by looking at morning blood glucose and IOB the following morning. As you can see below, the user was at target blood glucose and carrying a very small amount of IOB. Also, there are far fewer instances of basals alternating off/on. All good signs that the basal adjustments were reasonable.
Note: This entire scenario can also be applied in situations where (1) blood glucose has been steady but above target and (2) Loop is holding significant positive IOB, and (3) Loop has been high temping fairly consistently. By increasing the scheduled basals, Loop will recalculate that the positive IOB was actually necessary (not extra) insulin delivery, and the positive IOB will decrease. This will help the prediction curve to more accurately realize that not all the insulin deliveries were being utilized as \"corrections\" but rather as needed background basal.
"},{"location":"how-to/low-treat/","title":"Low Treatments","text":"Low blood glucose will still inevitably happen at some point, even with Loop use. A difficult carb count, illness, equipment malfunction, exercise...you name it, eventually something will help to cause a low blood glucose.
"},{"location":"how-to/low-treat/#how-to-treat","title":"How to Treat","text":"Overall, most people find that they can treat low blood glucose with fewer rescue carbohydrates than they previously used to. Since typically Loop will have been suspending your basal insulin before a low has been reached, you will typically have less IOB to overcome with carbohydrates. In other words, Loop takes some of the momentum out of an upcoming blood glucose dip...which also makes it easier to pull up from declining blood glucose.
If you notice that your old way of treating low blood glucose is leading to rebound blood glucose greater than you expect/desire, eventually consider decreasing the amount of carbohydrates you use to treat predicted/upcoming low blood glucose.
"},{"location":"how-to/low-treat/#logging-low-treatments","title":"Logging Low Treatments","text":"Another common question for new Loop users is whether to enter the carbohydrates used to treat a low blood sugar. The answer is that there is not a great agreement across all Loop users on how to deal with low treatment entries. The decision is dependent, often, on factors not just about the Loop algorithm; factors such as extra effort, use of OpenAPS's autotune (which needs all carbohydrate inputs to do a better estimation), or desire to have more thorough clinical records.
For the most part, you do not have to enter low treatments. Not logging low treatments will work out pretty well if you are not overtreating the low and you have your ISF pretty well determined. In the event you overtreat lows and your ISF value is too low, your Loop may be more likely to overtreat rebound situations.
"},{"location":"how-to/low-treat/#roller-coaster-after-low-treatment","title":"Roller Coaster after Low Treatment","text":"If you are roller coastering blood glucose after treating lows (going low, quick rise from the treatment carbs, then drop again from Loop's high temps during the rise, and then repeating pattern), here are some tips:
- While treating the low, try setting a temporary workout target of an hour to help keep Loop from aggressively treating a rebound.
- Consider lowering your maximum basal rate in Loop (within the Delivery Limits setting). This will limit Loop's ability to aggressively treat the rebound rise and allow you to work on getting your settings adjusted, if needed.
- Consider raising your ISF value (for example, if your ISF is set to 50, consider raising value to 55) to help Loop understand that a smaller magnitude of insulin adjustment is needed. In other words, tell Loop that insulin is packing a bigger punch than you'd previously thought.
If your ISF is set to a value that is too low compared to what it really should be, one of the most common symptoms you'll see is a roller coaster of blood glucose where the temporary basals are cycling between zero\u00a0and strong high temporary rates.\u00a0Here are some example graphs from Looped group.\u00a0These are examples where too low\u00a0of ISF is more than likely a large factor in the roller coaster\u00a0(doesn't mean it is the only culprit, and is more difficult to ferret out when food is involved like the second graph). \u00a0But, lightning bolt high temporary basals followed by very quick blood glucose drops and zero temps is usually too low of ISF value...raise the ISF value (e.g., go from 50 to 55) to help Loop know that each unit of insulin is actually having more impact than you'd previously thought.
"},{"location":"how-to/overrides/","title":"Override Targets","text":"Your Loop has two override targets that you can set; Pre-meal and Workout. When activated, these targets will \"override\" or take the place of your usual correction range and therefore will affect Loop's decisions about insulin delivery.
You can set the value for these override targets from the Correction Range area of Loop settings. Generally speaking, pre-meal targets are set to value(s) lower than your correction range and workout targets are set to value(s) greater than your correction range. Loop will indicate that an override target is active by showing a darker blue bar on the blood glucose chart showing the duration and loation of the target.
"},{"location":"how-to/overrides/#pre-meal-targets","title":"Pre-Meal Targets","text":"The pre-meal target, when activated by pressing on the icon, will stay active for:
- one hour, or
- until carbs are entered, or
- until pre-meal target is manually cancelled by pressing the icon again.
Pre-meal target will automatically end when one of those three things happens, whichever comes first will end the pre-meal target.
The original intent of the pre-meal target was to help give a gentle, easy pre-bolus effect to help lessen a post-meal blood glucose spike. By lowering your Loop's target going into a meal, you can get a \"pre-bolus\" effect because Loop will aim for that lower target and thus tend to set a little more aggressive insulin delivery while the target is active. Loop will recognize the \"extra\" insulin that has been delivered while pre-meal is active, and Loop will smartly take that into account when it is finally time to recommend a bolus at meal entry time.
Pre-meal target won't replace the utility of a true pre-bolus in most situations, but it can be useful in many situations beyond simply helping meals. This lower target can help Loop be more aggressive:
- During stressful events where adrenaline might be causing increased blood glucose
- During exercise that may tend to increase blood glucose
- During illness that is causing high blood glucose
The workout target, when activated, will give a duration choice of 1 hour, 2 hours, or indefinitely. The workout target, set greater than your usual correction range, is useful to help in situations where Loop needs to be less aggressive with insulin delivery. The most common use of the workout target is to set it in advance and during exercise to help minimize IOB going into a workout. While the name is a hint to the most common timing of this override's use, there are other times where it can be helpful such as:
- Helping to recover from low blood glucose treatment
- Helping after workouts when still experiencing increased insulin sensitivity
- Helping during illness which is tending to lower blood glucose or increase insulin sensitivity
You can read more about the timing and use of workut targets before/during/after exercise in the Exercise page of these tips.
"},{"location":"how-to/site-fail/","title":"Infusion Site Failures","text":"Some of the most frustrating and dangerous times are the result of insulin infusion site failures. When using Loop, site failures most likely will lead to your active insulin to be incorrect. Insulin that the pump thought it had delivered to your body is instead blocked by a kinked cannula or leaking out on your skin surface. Loop will have an incorrectly higher active insulin value since it thought that insulin had been delivered, but in reality it was not. Therefore your Loop will be less effective in closed loop mode to bring down any resulting high blood glucose from the site failure.
To deal with insulin site failures, there is no way to 100% recover from the issue immediately. Generally, we have no idea how much insulin has failed to deliver nor when the failure happened, so simply replacing that exact amount of insulin will not be possible. There are steps to help transition back to successful closed looping.
Take care to not rage bolus in this situation and give the new site a chance to start absorbing insulin. In cases where basal insulin delivery has failed, it takes a full two hours for resumed basals to come up to full effect for blood glucose impacts. There will be a lot of patience and attention during the 4-6 hours after a failed site is replaced, unfortunately.
"},{"location":"how-to/site-fail/#open-loop","title":"Open Loop","text":"Open loop mode is a first line of defense during this situation. By open looping for a duration of 6 hours, you will give your Loop an opportunity to have accurate active insulin (i.e., IOB) accounted for again since that amount of time is the length of your insulin activity duration. If you are very impatient, waiting out at least 4 of the 6 hours will get most of your active insulin accounting back into range. As you can see from the figure below, at 4 hours there is only less than about 10% of the insulin tail remaining, so the bulk of active insulin will be correctly accounted for by then.
"},{"location":"how-to/site-fail/#prime-insulin","title":"Prime Insulin","text":"If you are more certain about how much insulin is missing from your Loop's tracking of active insulin, you can choose to deliver insulin using the \"prime\" delivery function in your Medtronic pump while having your tubing connected to the new, working infusion site. Assuming you are using a Preferred Data Source set to Event History (most people are since this is the default setting), then prime insulin is not counted towards active insulin. In this way, you can deliver insulin to make up for some of the insulin that Loop had thought had been delivered (but wasn't).
"},{"location":"how-to/teens-kids/","title":"Test Your Settings","text":"Open Looping Fine-Tuning Settings Book settings
"},{"location":"how-to/think-like-loop/","title":"Think Like a Loop","text":"\u201cWhy isn\u2019t Loop giving me more?\u201d That is a common question from new Loopers.
\"Why is Loop suspending right now?\" That is another common question.
For these questions, you'll go really far in Looping if you can remember these two things:
- Go to therapy with your Loop
- Everything depends on the predicted BG curve
I'll challenge you to \"pretend you're at the marriage therapist\" as you learn how to use Loop as a partner.
Why are you in disagreement with your partner? What would a therapist tell you? Usually, therapy is because of miscommunications with each other. You aren't on the same expectations or priorities. The same applies to Loop. For example, you'd be at the therapist's chair telling her \"Loop just isn't treating my high BG as fast as I want. This is so frustrating. I correct and Loop just suspends.\" That right there is the definition of miscommunications and different priorities. Why?
\u27a1\ufe0f Loop thinks you have enough insulin to be at target later.
\u27a1\ufe0f You think you need more right now.
So...look at the objectives and info you both have at hand.
Loop has the info from the settings you\u2019ve given it, how you\u2019ve described your meals, and how you\u2019ve given insulin. Loop does not know if you are stressed. Loop does not know if you are sick. Loop is not a rage bolus machine and is predicting 6 hours out to safely bring you down without you needing to prevent a low while it does that. Loop is considerate partner. It respects your input and acts gentle and even-keeled.
If you are frustrated with Loop\u2019s actions now...that means you told it the wrong info previously (or are lacking some patience that Loop has). Most often either your meal entry was wrong or your basals or...something else you\u2019ve told Loop has ended up being incorrect.
Fixing that miscommunication will allow Loop to do a better job.
If you misjudged a meal, go back and edit that meal entry.
If you misjudged your basal/ISF/carb ratio needs, correct those (testing to validate is good if you haven\u2019t done that yet)
If you aren\u2019t patient, you can override Loop...but do expect that is not likely to solve the miscommunication you and Loop are having. That\u2019s the relationship equivalent to just talking louder as opposed to talking smarter.
Loop will try to save you when you rage bolus, which may just make you even more frustrated. If you can stop and think \"What have I told Loop that might not be accurate right now?\" that will be a good step to a meaningful improvement in outcomes.
"},{"location":"how-to/think-like-loop/#tools-when-you-are-in-disagreement","title":"Tools when you are in disagreement","text":"Sure, even with the best therapist, you may still have disagreements. They happen. There are several options for recovering from times like that. The tool you choose may be situationally dependent. Seeing a huge quick BG climb after a meal may be better helped by a carb entry edit as opposed to added patience, but that will depend on your management style.
- Add some patience to your practice. A little patience to give the already added insulin time to work can help with Loop.
- Edit your meal's carb entry. If you find your meal going crazy different than expected (did they forget you said \"diet\" coke?), you can always go back and edit your original carb guess to be more accurate based on what you're seeing. You can decrease the carbs or increase the carbs. You can change the absorption time. These edits can be done by tapping on the carb chart on Loop's main screen.
- Use overrides to tell Loop about overall insulin needs changing. Hormones...oh lordy the hormones are horrible. That's a good time to use an override to let Loop know that overall you need more than usual. Or if you're about to hike the Himalayas, use an override to let Loop know you will need far less insulin than an average day.
- Open Loop. This is a great tool. If you just had things go really wrong for a bit (bad cannula kink anyone?) then you can open the loop until things settle down and you get squared away again. Give your manual injection, Afreeza, or whatever your desire is, and wait until you've had 4-6 hours of regular basals going before closing the loop again.
- Give pending insulin. If you are really feeling the itch to rage bolus, you can meet Loop at the halfway mark. If you click on the bolus tool while Loop is high temping you, you will see a \"pending insulin\" value. That is the amount of insulin Loop is planning on giving you through temp basals over the next 30 minutes. If you give that pending insulin total as a bolus now, you will get the entire amount working faster AND Loop agrees you are safe to use it. In this way, Loop won't automatically suspend insulin when you give pending insulin...because your predicted BG curve had said it was needed anyways. Win-win...you got a mini-rage bolus and Loop won't be working against you. Good therapy and a good way to avoid post-rage lows.
As you start to use Loop you will probably find yourself wondering at some point \"Why is it suspending insulin right now?\" or \"Why isn't it giving me increased basals right now?\" You might find yourself reverting to a bunch of old-school habits like looking at your carbs on board or insulin on board and trying to calculate a bunch of numbers. Stop there...that's so 1990s. You're a looper now and there is actually an easier way to assess things.
Thinking like a Loop involves just three things
\u27a1\ufe0f Predicted BG curve \u27a1\ufe0f Correction Range \u27a1\ufe0f Suspend Threshold
All of Loop's decisions are based on what your predicted BG curve is doing with respect to your correction range and suspend threshold. That's it. All the time. Everytime. It always comes down to that predicted BG curve.
That predicted BG curve has four contributions to its shape:
- Carbs (the ones you've told it about)
- Insulin (the ones you've given through your pump or recorded in the Health app)
- Blood glucose momentum (how fast your BG has been changing recently)
- Retrospective correction (how accurate/inaccurate Loop has been recently in predicting your BGs)
Loop models the inputs from those four contributions to form the final shape of the predicted BG curve. Once that curve is generated, Loop looks at where the entire curve will be for the next 6 hours relative to your suspend threshold and correction range, and take one of four actions:
Action 1: Set a 0 u/hr basal rate (aka suspend basals)
Loop will do this anytime your predicted BG curve has any portion that goes below your suspend threshold.
And the remaining three actions all assume NONE of your predicted BG curve is below suspend threshold, otherwise you'd be in Action 1.
Action 2: Set scheduled basal rate
Loop will set your scheduled basal rate when:
- All the predicted BG curve is within the correction range, including the very last value which is called your \"eventual BG\", or when
- If the eventual BG is above your correction range in 6 hours, but there's a dip below the correction range earlier in the curve.
Action 3: Increased basal rate
If the eventual BG and all the predicted BG curve is above your correction range, Loop will give you a high (increased) temp basal.
Action 4: Decreased basal rate
If your eventual BG in 6 hours is below the correction range, you'll get a decreased basal.
"},{"location":"how-to/think-like-loop/#what-would-loop-do","title":"What would Loop do?","text":"Test your new found skills...let's play a game where you try to guess the action...
What would Loop do?
It\u2019s 8pm and your predicted BG curve looks like the following. What do you expect Loop to recommend/enact at 8pm? A. Zero (suspend) temp basal B. Scheduled basal from your settings C. High temp basal D. Lower temp basal (between zero and scheduled)
What would your answer be?\u00a0\u00a0Before you give your final answer...consider this next twist.\u00a0 Would you give the same answer to this graph as you gave the graph above?\u00a0 If not, what would the answer be for this graph and why? (ignore the timestamp mismatch
This second case has a dramatic drop happening.\u00a0 Suspend threshold is still at 60 mg/dL, the correction range is still 90-110, the lowest value on the predicted BG chart is 75 mg/dL, and the eventual BG is 171 mg/dL for this example.\u00a0 So, pretty similar to the first example except for this precipitous drop going on right now.
Let me tell you some of the common pitfalls we all can easily slip into when trying to answer these \"Why is Loop giving me this basal?\" questions.
Wondering about IOB, COB, or DIA in order to answer...those aren't a factor in answering the question as they have already been used to make the predicted BG curve.\u00a0 In other words, they are accounted for already in the information presented.\u00a0 All you need to answer this question is provided by the predicted BG curve, your suspend threshold, and your correction range. Thinking about this as a human...humans tend to say \"Well, I'm on a rise/fall right now so...[insert Loop action based on that]\"\u00a0Loop isn't looking at the past BG movement alone, instead, it's looking at the\u00a0prediction curve ahead and applying its rules based on that.\u00a0 Any drop or rise going on will have been added to the predicted curve through the BG momentum and retrospective correction components of the algorithm...so again they're already incorporated into the predicted curve.
Restating for emphasis:\u00a0 All you need to answer this question is provided by the predicted BG curve, your suspend threshold, and your correction range.
The answer is B:\u00a0 Loop will give your scheduled basal in both situations shown above.\u00a0 When your predicted BG curve (1) drops for a time below the correction range but (2) all of the curve is still above suspend threshold, and Eventual BG is (3) above range or within range...Loop will give your scheduled basal.
The logic is a bit of a wait-and-see.\u00a0 Scheduled basal will maintain the delivery of insulin.\u00a0 Your settings haven't told it this is an \"oh my gosh...stop the insulin!\" moment (you're predicted to still stay above suspend threshold), but we also don't want to give high temps yet (to correct the eventual BG) because we'd like to safely make it through the part that is below correction range coming up.
If BGs were to drop (enough) or keep dropping (enough), your predicted BG curve would likely slip to your suspend threshold and then Loop would suspend.\u00a0(One important take away is to not set your suspend threshold so low that it no longer acts as a safety in these situations.) If BGs were to rise enough such that the whole predicted curve comes back into or above the correction range completely, you'd then get high temp basals to correct for that eventual BG that is above the correction range.
"},{"location":"settings/adjust/","title":"Adjusting Your Settings","text":"There will be times when you need to adjust your settings in order to keep your blood glucose within your correction range. The degree to which you adjust your settings will depend on a variety of personal decisions about how active you engage with your diabetes, how comfortable your endocrinologist may feel with patient-adjusted settings and the underlying reason for the change.
"},{"location":"settings/adjust/#troubleshooting-before-a-change","title":"Troubleshooting before a change","text":"The most common reason that people decide it may be time to adjust their settings is when previously great blood glucose control suddenly becomes elusive. What worked yesterday is no longer working today. Because of course, as soon as you test and dial-in all those settings, diabetes will throw you a curve ball and change your insulin needs. That's the way it works. It's not just YDMV (your diabetes may vary), it's actually YDWV (your diabetes will vary).
Before considering a change to your settings, the usual laundry list of possible other causes should be explored.
- Bad infusion site (kinked cannula, blood in cannula, infected site?)
- Bad insulin (cloudy? insulin exposed to heat/freezing?)
- Unexpected food impacts (that diet coke wasn't diet, kid sneaking food?)
- CGM damaged (poor calibration, damaged sensor site?)
Some signs that the issue is actually the result of equipment failure include the failure of blood glucose to decrease despite repeated bolus corrections. Often during equipment failure troubleshooting, you've blamed the food (and adjusted a carbohydrate count), given corrections, checked for hidden food wrappers, and maybe even open-looped rage bolused. Breathe deep, it is super frustrating to have equipment failure. One of the benefits of Loop can be that equipment failures can become a little easier to identify over time because blood glucose control becomes more predictable and reliable.
"},{"location":"settings/adjust/#assess-the-duration","title":"Assess the duration","text":"Once you have eliminated equipment failures as a source of unexpected blood glucose control problems, it is time to determine if this is a short-term or long-term issue. This assessment tends to be a skill that people will get better with practice.
Before considering a change to your settings, you should assess if you expect the issue to resolve on its own soon or if the issue might be caused by a more long(er) term issue that should be addressed.
"},{"location":"settings/adjust/#short-term-factors","title":"Short-term factors","text":"There's a long list of short-term factors that may temporarily affect how well Loop can automatically keep your blood glucose at your correction goal. Of course, the definition of \"temporarily\" is also subjective. Some people grow impatient with blood glucose excursions within hours and other people don't consider the need to correct excursions for days. That difference in tolerance will affect a user's assessment of how quickly they will consider adjusting settings. For these discussions, short-term is referring to less than a day.
For the most part, it is not a good idea to adjust settings for very short-term underlying causes such as:
- One or two meals that behave differently than expected
- Stress from a final exam
- Anticipation of holiday party buffet table
- Sprinting in the airport to catch a plane
- \"Baggage claim\" lows from pressure changes causing unexpected insulin delivery
- Weather
Those are very short term factors and if you change your underlying settings too quickly from just one or two observed blood glucose excursions, you'll likely end up needing to undo the changes just as quickly. For these short-term factors, the best tools involve using temporary target overrides and a measure of patience. In some medium-duration cases (factors that last for at least several hours), people also use \"fake\" carbohydrate entries to mimic the extra insulin needed during temporary stress on the system which causes high blood glucose.
"},{"location":"settings/adjust/#long-term-factors","title":"Long-term factors","text":"The list of long-term factors is just as long, but the most common are probably:
- Growth hormones
- Menstrual cycle hormones
- Illness
- Medications
- Travel
- Beginning new consistent exercise routine (or suddenly becoming more sedentary)
For long(er) term factors, you may benefit from adjusting your underlying settings to reflect the insulin-need changes that have happened as a result of the factors. By adjusting your underlying settings, you will help Loop to better predict your insulin needs and therefore you'll be more likely to achieve blood glucose in correction range.
"},{"location":"settings/adjust/#assess-the-data","title":"Assess the data","text":"Once you have assessed that there's a need for a settings adjustment, the next step is to decide which setting(s) to change and how to adjust them.
Whether you undertake the adjustments by yourself and/or with your health care provider's input is at your own discretion. Ideally, you would have your health care provider's input and support to discuss the observations and concerns you have and develop an adjusted setting profile based on those discussions. Many endocrinologists are still becoming familiar with Loop and its algorithm, so providing them links to the various sites would be a good starting point for discussion. This can help the endocrinologist understand where Loop is similar/different than traditional pump therapy. If you and your health care provider are comfortable with your self-adjustments and Loop's algorithm, then proceed ahead.
To begin the process of settings adjustment, visualizing your data trends is important. You need to be able to piece together what was eaten, how it was bolused, what temporary basals were applied, and how your blood glucose was reacting. The Data section of LoopTips covers the various places you can find your diabetes-related data to help you during adjustments.
"},{"location":"settings/overview/","title":"Overview","text":"In the context of the following discussion, the term \"settings\" refers to information that the user has typically entered into their pump. These settings include:
- basal rate schedule
- insulin sensitivity factor (ISF)
- carbohydrate to insulin ratio (CR)
- glucose target range
- duration of insulin activity (DIA)
Typically, most people with type 1 diabetes consult with their endocrinologist to develop and revise their settings. The endocrinologist usually reviews the patient's blood glucose and treatment history at each appointment to assess whether settings need to be adjusted. There are several great books available about the topic (Think Like A Pancreas and similar)
"},{"location":"settings/overview/#why-settings-matter","title":"Why settings matter","text":"Loop is primarily a set of math equations called an algorithm. The recommendations and actions that Loop takes are based on that algorithm, and the settings you enter in your Loop app are the basis for that math.
Let's start by thinking about basal rates. A well-adjusted basal schedule is designed to keep your blood glucose steady throughout the day, everything else being equal. If you were to not eat, not exercise, and basically keep a mellow lifestyle...basals should keep your blood glucose steady.
That is how Loop's math starts, and it's an important point to remember as you use and learn your Loop app. Loop's math is based on the assumption that the basal schedule you have provided in your settings are capable of keeping your blood glucose steady in the absence of other stressors. So as your blood glucose goes higher than your correction range for an unusual short-term influence like stress or unannounced carbs, you've been accoustomed to delivering a \"correction bolus\" to get back to range. Or if blood glucose goes below your correction range, you may need to eat recovery carbohydrates.
With all the excitement about automated insulin delivery, some people mistakenly assume that the user's settings don't matter anymore....that everything is automated. However, settings do still matter as they provide the basis for Loop's math. Diabetes is not a static math equation. Loop does not adjust your settings for you, that responsibility still falls to the Loop user when needed.
Let's give some examples to help. When you calculate a meal bolus, your math probably looks similar to this:
But many times you are probably adjusting that math based on any number of \"factors\" that you are aware could be affecting your blood glucose. Diatribe put together this great list of factors that affect blood glucose, and therefore could also affect how you'd adjust your insulin delivery from what they otherwise would normally be.
Some of those factors you are already used to making allowances for, such as:
- Pizza boluses...You probably are accustomed to needing an extended bolus and/or higher temporary basal rate to help with the late blood glucose spike common after the initial bolus starts to wear off. That pizza takes a long time to digest thanks to those slow-acting fats.
- Medications...How about that time that you had to take steroids for an achy joint? Those steroids probably resulted in a dramatic overall increase in insulin needs for several days, and you had to set aggressive temporary basal increases or other settings adjustments.
- Hormones...monthly hormone cycles have many women adjusting their basal rates and/or insulin sensitivity schedules for a week or so each month.
- Exercise...always a tough one because some exercise may make your blood glucose rise (lifting weights) and other exercise may make your blood glucose drop (running). Through repetition, most people have learned how to adjust for exercise by using temporary basal rates or timing of meals/boluses.
The degree to which Loop can deal with those stresses/factors is largely dependent on the duration of the \"factor\" itself and how well we can communicate to Loop that those factors are at play. For things we know are coming, we can tell Loop in advance, and Loop has ways of dealing with them mathematically. In the pizza example, Loop lets us mark the meal as long, slow digestion, and therefore better react to a potential late blood glucose rise. For exercise, we can use temporary override blood glucose targets before/after exercise to help Loop administer less insulin for awhile.
If, however, your insulin needs are impacted by something more long term such as hormone cycles, medications, or illness then sometimes your Loop may not be able to consistently hold you at your desired blood glucose. You may end up holding steady at a higher/lower blood glucose than you'd prefer. This is because Loop's math is still using the inputs that were from before you got sick/stressed/hormone'd (new word). Then the decision becomes \"Do you adjust your settings now or just wait it out until the stressor goes away?\"
"},{"location":"settings/overview/#when-do-you-need-to-change-settings","title":"When do you need to change settings?","text":"If the inputs to those math equations (i.e., your settings) are not accurate, your results for blood glucose control may also not be optimal.
Your basic insulin settings are what keeps your blood glucose steady without stressors present. For much of the time, the stressors in type 1 are short-lived and temporary (we just ride them out)...carbohydrate count is a little off, math test in algebra class, public speaking engagement, etc. Loop does well helping with the short-term stressors. They don't represent a prolonged need to change settings and the blood glucose impacts are short-lived as Loop corrects for them.
However, the longer term stressors are often helped by settings adjustments. Examples of longer term stressors are medications, illness, hormones, travel, or altitude. By letting Loop know that the status quo has changed (more/less insulin is needed overall) will help BGs get and stay in range. If you have told Loop that your insulin need is 1 unit/hour, but in reality, you're needing 1.3 units/hour because you've traveled to a colder place and will be sedentary for days...you will likely find yourself steady but higher than your target range while Looping. The high temporary basals that Loop will be providing in an attempt to bring you back to target will simply be going to fill the lack of adequate basal scheduled. If you update your basal schedule to reflect 1.3 units/hour of basal needs, Loop would be better positioned to provide needed insulin delivery to keep you in the target range.
"},{"location":"settings/overview/#automated-settings-adjustments","title":"Automated settings adjustments","text":"A common sentiment among new Loop users goes something like \"Why doesn't Loop see that I'm stuck on a high blood glucose? Why doesn't it bring me back to my target blood glucose?\"
What the user is actually implying is that the user thinks Loop should be doing some sort of retrospective analysis on the user's blood glucose and data to make a \"better\" decision. The user wants Loop to learn from recent data and adjust.
There is a short-term retrospective analysis built into Loop which will apply a weighted-correction based on the past 60 minutes of blood glucose changes. While this does help some, larger-scale \"learning\" is not currently a part of Loop's algorithm.
Perhaps in subsequent versions of Loop, auto-adjustment of settings or machine learning could be incorporated. Until then, you will need to tell Loop if your underlying settings need updating or make temporary adjustments for short term issues. These pages in LoopTips are written to help you identify when settings may need adjustment or when other techniques may be useful in given situations.
"},{"location":"settings/settings/","title":"Initial Settings","text":"Here's how 90% of Loopers start:
- Switch from their existing pump over to a Loop-compatible pump
- Within hours, they begin closed loop use of Loop
- When blood glucose isn't as smooth as hoped for, Looper begins to try to adjust settings while still in closed loop
The problem with this sequence is that adjusting settings on the fly while in closed loop can be difficult as a new user. As you become more experienced there are some great tips for adjusting settings while in closed loop, like Loop and Learn: Basal Testing in Closed Loop, but as a new Looper, you should start with the basics. Here are some recommendations to help make the transition to closed loop easier. The first place to start is with getting the best initial settings as you can.
"},{"location":"settings/settings/#try-open-looping-first","title":"Try Open Looping first","text":"If I could offer one bit of really useful advice, it would be to open loop for at least a few days. Adjust to the user interface and buttons you'll be pressing. This breaking-in phase will also let you get used to a new pump if you've never used Omnipod before or if you are new to using a Medtronic pump. If there are differences between how your body absorbs insulin from your old pump vs the new pump, open looping will be a good way to notice those before adding the complexity of automated corrections from Loop.
Open Loop is like Manual Pump
While open looping and entering meals with the pizza icon, Loop will recommend less upfront bolus with the intent to make up the remainder of needed insulin via automated dosing later. In those situations, you may have to manually dose the split portion of the bolus later, similar to normal pump use for slow digesting meals like pizza and quesadillas, since Loop will not be automatically dosing.
If you want extra help, you can read some excellent books about pump settings and what they mean. For example, \"Think Like a Pancreas\" by Gary Scheiner is a great reference for understanding some of the guiding principles in pump therapy. \u00a0Let me summarize the important parts about pump settings:
- Basal rates should keep your blood glucose steady in the absence of other influences (such as food, medications, etc).
- Boluses should return your blood glucose to target after a meal.
- Insulin Sensitivity Factor (ISF) is the amount one unit of insulin drops your blood glucose without other influences.
- Carb Ratio (CR) is the meal entry in grams that requires one unit of insulin to cover
Basal, ISF and CR
It's the way it was defined in the beginning, but it can be confusing.
If you need more insulin
- Basal rates should increase (higher number)
- ISF and CR should be stronger (lower number)
- They are in the denominator of the fraction
It is absolutely possible to have two wrong settings look like right settings when they balance out. The problem is that those wrong settings won't balance out in all situations...and Looping will not be smooth in those situations. Remember - in closed loop mode, Loop will adjust with every new glucose reading - typically every 5 minutes. What worked on a manual pump when adjusting infrequently, might not be suitable for Loop.
Example:
- Too low a value for basal rates can be offset by regular eating of meals with too strong (lower value) carb ratio
- If you stop eating though, you'll start going high because that extra insulin from the meal boluses won't be there to offset the lack in basal rates
Taking the time to validate your settings by truly testing them is really good practice before closing loop.
"},{"location":"settings/settings/#insulin-duration","title":"Insulin Duration","text":"The majority of pumpers used DIA at about 3 hours on traditional pump therapy.\u00a0There's a reason for that. \u00a0In traditional pumps, a simple linear model is used to estimate the decay in effectiveness of the insulin already delivered. And it is true that after 3 or 4 hours, insulin is not affecting glucose as much as initially. But the true biological lifetime of insulin has a long tail. Insulin from more than 5 hours ago is still making a small decrease in your glucose. Loop takes that into account using a realistic model for insulin effectiveness.
In your traditional pump, the DIA, using a linear model, was used to give a correction because blood glucose is stuck high or going low...in other words, DIA was used as a rough approximation to correct off-target blood glucose. It doesn't have to be rocket science then since we're making an approximation because some other numbers (carbohydrate count, basal rates, ISF, etc) weren't behaving the way we were expecting either and therefore leading to an off-target blood glucose.
With the release of Loop 3, all insulin types use a realistic model, so you will not be selecting a DIA. These models help prevent insulin stacking as you begin Loop. You can read more about insulin models and how they help determine the active insulin over time in the LoopDocs: Algorithm pages.
With Loop 3, you will select the Insulin Type you use, such as Novalog, Humalog, Apidra, Fiasp or Lyumjev. Each is associated with the appropriate exponential model for onset and peak action times.
"},{"location":"settings/settings/#basal-rates","title":"Basal Rates","text":"Basal rates are the keystone of your therapy settings. Establishing a quality basal schedule is setting yourself up for successful looping. Personally, we find testing basal rates can be pretty painless and doesn't require days of fasting. Instead, look for easy opportunities.\u00a0If you are willing to open-loop test, that is going to give the most accurate information in the quickest way.
It's a pretty easy test. Turn off your loop or open loop. Don't eat food, don't do crazy exercise, don't sit in a hot tub. Just have a relaxing average time period and see if your basal rates are able to hold you roughly steady. It doesn't matter if you are at target or not; the idea is to simply have zero extra active insulin from any boluses or corrections and watch what happens during those hours. Typically you would want to see about 3-4 hours of blood glucose without the influence of food boluses or activity. You can even try this tip: Loop and Learn: Basal Testing in Closed Loop
Many people, especially little kids, are not enthusiastic about fasting basal testing...so look for opportunities to make it less cumbersome if you have to eat during the test. For example, use a low carb meal/snack that you know how to bolus for (or doesn't need a bolus at all) and generally needs no corrections.\u00a0For us, that's two extra large scrambled eggs (or three small ones) with cheese\u00a0bolused at 8g.\u00a0If she eats that meal, the blood glucose response is slow and measured. Three hours after that meal, the bolus and food effects are really muted and we can start watching to see if blood glucose stayed pretty steady. For example, here's some of a recent open-loop basal testing we did and confirmed that blood glucose could stay pretty steady without the involvement of temporary basal help from looping.
"},{"location":"settings/settings/#insulin-sensitivity-factor","title":"Insulin Sensitivity Factor","text":"Insulin sensitivity factor (ISF) is the next logical setting to test.\u00a0If you've just done the basal test and gotten steady blood glucose with an open loop, try taking a glucose tab or two. Wait for your blood glucose to be steady at the higher value, and give a safe correction that you think will get you close to target. Watch the resulting blood glucose drop over the next 3-4 hours. You should see blood glucose come to a steady level again. How much did the blood glucose drop? How many units of insulin did you use? Divide the two numbers and you will have your ISF. \u00a0If your BG dropped 15 mg/dL\u00a0with half unit of insulin, your ISF is approximately 30 mg/dL per unit.
Testing ISF is an important step in successful closed Loop use. ISF is used in every Loop calculation and plays an important part in Loop's prediction curve. Most users find that they have to adjust ISF; most often increasing the value. If you are experiencing roller coastering blood glucose, as described below, that is often a sign that ISF needs to be weaker (higher value).
Before you read the next part, let's clear up some definitions and terms:
Insulin sensitivity factor described just that...how sensitive your blood glucose is to the effect of insulin. If your blood glucose doesn't drop much when you put in insulin and you have to use a lot, your insulin sensitivity is \"weak\". If your blood glucose drops quickly and you don't need much insulin to correct a high blood glucose, your insulin sensitivity is \"strong\".
The terminology gets confusing sometimes because some people think of the terms \"weak\" and \"strong\" in terms of how Loop's actions will behave. If you make your ISF value greater (i.e., changing it from 50 to 55), you are weakening the magnitude of Loop's adjustments because you are telling Loop that your sensitivity to insulin is stronger...you don't need as much to correct a high blood glucose and suspensions will have a tendency to rebound more.
In order to avoid the confusion that surrounds the words \"weak\" and \"strong\", instead we will describe the actual adjustment to the ISF value...the number itself.
If your ISF is set to a value that is too low compared to what it really should be, one of the most common symptoms you'll see is a roller coaster of blood glucose where the insulin adjustments are cycling between zero\u00a0basal and strong high temporary rates (or automatic boluses).\u00a0Here are some example graphs from Looped group.\u00a0These are examples where too low\u00a0of ISF is more than likely a large factor in the roller coaster\u00a0(doesn't mean it is the only culprit, and is more difficult to ferret out when food is involved like the second graph). Lightning bolt high temporary basals followed by very quick blood glucose drops and zero temps is usually too low of ISF value...raise the ISF value (e.g., go from 50 to 55) to help Loop know that each unit of insulin is actually having more impact than you'd previously thought. Making changes of about 10% for a setting is reasonable to try; just make sure you are changing it in the correct direction - remember: a lower number = more insulin; a higher number = less insulin.
"},{"location":"settings/settings/#carb-ratios","title":"Carb Ratios","text":"Now that you have basals and ISF set, here's where it gets really tempting to close loop and move on. And, truthfully, it's not that hard to test carb ratios on a closed loop vs an open loop if you've solidly tested all these other factors.
A good carb ratio will bring your blood glucose, for an average meal, back to the starting point of the meal within about 3 hours or so.
A bad carb ratio will leave you higher or lower than the starting point of the meal.
For example, this graph shows too aggressive of a carb ratio (value too low). \u00a0Three hours after the meal, there's nearly 0.50 units of active insulin, blood glucose is well below where the meal started, and definitely low treatments needed.
If you are finding that a correct carb ratio is yielding good blood glucose 3 hours later, but you aren't happy with the peak blood glucose during the meal, then it may be time to explore increasing or adding prebolusing time to your meal or implementing \"eating soon\" targets an hour before meals to help control the post-meal blood glucose spike.\u00a0Artificially strengthening carb ratios to help control post-meal blood glucose spike will likely yield lows 2-3 hours after a meal.
"}]} \ No newline at end of file +{"config":{"lang":["en"],"separator":"[\\s\\-]+","pipeline":["stopWordFilter"]},"docs":[{"location":"","title":"Introduction","text":"Welcome to LoopTips! The purpose of these pages is to help you make the most of your DIY closed-looping experience. In particular, these pages support users of the Loop app. These documents have been a long time in the making and as always, your feedback is welcome. If you have comments, please feel free to provide input at the Looped Facebook Group or in Loop's Zulipchat.
LoopTips is a companion to LoopDocs. LoopTips does not cover the build or installation of Loop app. If you have problems with building your Loop app or troubleshooting the technical aspects of Loop operations, please head over to LoopDocs for help. The LoopTips pages are specifically related to understanding how your loop is recommending the actions being taken, and how you can improve your blood glucose outcomes on Loop. Some of the pages that originated in LoopTips are now included in LoopDocs. Don't be surprised when you find links between the sites. (The color bar is deliberately a different shade to help you know where you are.)
"},{"location":"#what-is-a-closed-loop","title":"What is a closed loop?","text":"The U.S. Food and Drug Administration (FDA) has a general definition of an \"artificial pancreas device system\" (aka closed loop system) on their website, quoted below.
\"Most Artificial Pancreas Device Systems consists of three types of devices already familiar to many people with diabetes: a continuous glucose monitoring system (CGM) and an insulin infusion pump. A blood glucose device (such as a glucose meter) is used to calibrate the CGM.
A computer-controlled algorithm connects the CGM and insulin infusion pump to allow continuous communication between the two devices. Sometimes an artificial pancreas device system is referred to as a \"closed-loop\" system, an \"automated insulin delivery\" system, or an \"autonomous system for glycemic control.\"
An Artificial Pancreas Device System will not only monitor glucose levels in the body but also automatically adjust the delivery of insulin to reduce high blood glucose levels (hyperglycemia) and minimize the incidence of low blood glucose (hypoglycemia) with little or no input from the patient.\"
"},{"location":"#what-is-loop","title":"What is Loop?","text":"The Loop app is a do-it-yourself closed loop algorithm and user interface, developed through the work of community volunteers. You can read about the history of Loop development in LoopDocs. Loop predicts future blood glucose based on basals, carbohydrate intake, insulin deliveries, and current CGM readings. These blood glucose predictions provide Loop with the information needed to recommend a temporary basal rate to attain a targeted glucose range in the future. The system can either operate as an \u201copen loop\u201d by making recommendations to the user for their approval before enacting or as a \u201cclosed loop\u201d by automatically setting the recommended temporary basal rate.
As exciting as this sounds...we should first get the caveats out of the way. There are some limitations on what kind of pumps/equipment are required. This system doesn't work with most pumps and requires you to do some work. The pumps that are supported and the directions for building are thoroughly covered in LoopDocs.
"},{"location":"#disclaimer","title":"Disclaimer","text":"While it may seem obvious, please consult with your health care professional regarding your diabetes management. The suggestions and discussion in LoopTips are not a one-size-fits-all nor intended to replace the input from your doctor. You take full responsibility for building and running this system and do so at your own risk.
"},{"location":"translate/","title":"Translation","text":""},{"location":"translate/#google-translate-links","title":"Google Translate Links","text":"Click on a language from the list below to turn on Google automatic translation.
- Every LoopTips page gets automatically translated to the selected language as do links to other websites from LoopTips
- The Google Translate Tool will appear at the top of each page
- LoopTips: Google Translate Tool Instructions
- Google: Google Translate Help Link
Automatic Translation
These links connect this site to the Google Translation service.
- No human has reviewed the translated information for accuracy
- Please use the translation with care
- Not available in all regions
- Some links might not work as expected
- Any \"code\" not protected by blocks may not appear correctly - be sure to click on Original to make sure you a viewing code properly
\u0639\u0631\u0628\u064a
\u0411\u044a\u043b\u0433\u0430\u0440\u0441\u043a\u0438
\u010ce\u0161tina
Deutsch
Dansk
\u0395\u03bb\u03bb\u03b7\u03bd\u03b9\u03ba\u03ac
Espa\u00f1ol
\u65e5\u672c
Suomi
Fran\u00e7ais
\u05e2\u05d1\u05e8\u05d9\u05ea
Hrvatski
Italiano
\ud55c\uad6d\uc5b4
Norsk
Nederlands
Polski
Portugu\u00eas
Rom\u00e2n\u0103
\u0420\u0443\u0441\u0441\u043a\u0438\u0439
Sloven\u010dina
Svenska
Turkish
\u4e2d\u6587\uff08\u7b80\u4f53)
\u4e2d\u6587\uff08\u7e41\u9ad4)
"},{"location":"translate/#google-translate-tool-instructions","title":"Google Translate Tool Instructions","text":"Once Google Translate has been turned on, clicking on a link above shows a \"Google Translate: Can't translate this page error\".
- To modify the language for a single page, use the Google Translate tool
- To modify the language for the entire site, see Change Language
The graphic below shows the Google Translate Tool when maximized (default) for a browser and mobile display. The tool can be minimized by tapping on the up/down carets at the right of the tool. This is very useful if the tool obstructs part of the original screen. Additional options can be selected with the three-dot menu as shown in the graphic.
"},{"location":"translate/#change-language","title":"Change Language","text":"To modify the language choice for the whole site, copy the line below and paste it into the URL and then choose the desired language from the list
https://loopkit.github.io/looptips/translate\nOR
Use the Google Translation three-dot menu and select Go to Original URL while on the Translation page.
Perceptus is a website and set of free tools developed by Kenneth Stack and Gustavo Munoz, two active members of the DIY community. Perceptus provides two tools particularly helpful for Loop users; Dash and GlucoDyn. As a side note, Gustavo has an excellent Spanish Tedx Talk about building his own DIY looping algorithm. Kenneth is also the brains behind NSApple watch app.
"},{"location":"data/glucodyn/#dash","title":"Dash","text":"For Loop users with Nightscout integration running, Dash can provide an alternate view and data analysis for the Loop data stored in your Nightscout's mLab database. (Note: mLab used to be called MongoLab, still the same intended reference)
To set up your Dash integration, you will need to:
-
Signup for a Perceptus account here
-
Provide your mLab database information
Assuming you have a basic Heroku account to host your Nightscout website, you can access your mLab information as shown in the screenshots below. (Users who manually setup their own mLab databases as part of Azure accounts will need to access their mLab database manually using their login and account information.)
Login to your Heroku account to start the process, and then that will take you to your mLab information.
Your mLab database name and API Key can be found on the same page, as shown below. Copy and paste these bits of information into your User Settings in Dash.
The format for the MongoLab API URL string is: https://api.mlab.com/api/1/databases/your-database-name/collections/entries where you are replacing the your-database-name part with your info from the first circled area shown below.
Dash provides multiple views of your mlab information, and the data is updated live as your Nightscout data updates too. While there are some constraints (cannot set your own low/high range limits), the data views are very useful for identifying difficult times of the day or difficult days of the week.
"},{"location":"data/glucodyn/#glucodyn","title":"GlucoDyn","text":"GlucoDyn was created to educate people with T1D and their caregivers about blood glucose dynamics. Each day people with T1D make decisions about what and when to eat, and how much insulin to use to cover carbohydrates effect. It can be difficult to visualize what is happening throughout the day as carbohydrate absorption raises blood glucose and insulin reduces it \u2013 particularly when multiple events interact with one another.
GlucoDyn uses calculations similar to an insulin pump\u2019s bolus wizard to create graphs of blood glucose versus time, based on your inputs of carb and insulin events. The user can enter carbohydrates, boluses, and temporary basals, and GlucoDyn calculates the effect on blood glucose over time. GlucoDyn also allows the user to \u201cslide\u201d the events around once they have been entered \u2013 allowing you to see what would have happened to your blood glucose curve if you had bolused earlier or eaten a different amount of carbohydrates \u201con the fly\u201d \u2013 the curves change as you move the sliders.
Seeing the curves and understanding the effects of bolus timing, carbohydrate absorption rate glycemic index (GI) value, temporary basal timing and rates, etc. can help users with their overall understanding of T1D and its management.
Originally, Loop used the same carbohydrate absorption and insulin models as Glucodyn uses. Those have since been updated in Loop with dynamic carbohydrate absorption and exponential curves for insulin. Even with those changes, the GlucoDyn tool can still provide users an excellent tool to visualize how small changes in bolusing can affect trends in blood glucose. For example, it\u2019s quite interesting to see the effect of pre-bolusing. By delaying the carb intake by say 20 minutes, you can visualize the reduction of the maximum blood glucose. But, you have to be careful \u2013 before the carbohydrates have time to absorb insulin is working, and the minimum blood glucose will also drop. These are the kinds of tradeoffs that are critical to the management of T1D, and thankfully are automated by Loop's smart bolusing based on carbohydrate absorption estimates.
"},{"location":"data/health/","title":"Health App","text":""},{"location":"data/health/#overview-of-healthkit-and-health","title":"Overview of HealthKit and Health","text":"There's often confusion over the terminology surrounding Apple's health data storage so let's start there.
- HealthKit -- Apple's iOS database for storing health-related data. App developers can choose to use HealthKit to store information from their apps such as heart rate, blood glucose, insulin use, and a long list of other parameters
- Health App -- Apple allows iOS users to manage HealthKit permissions and view stored HealthKit data through the use of Apple's Health app (standard on iPhones and iPods, but not available on iPads).
When you first install Loop onto your iPhone, you will be asked to setup Loop's access to the HealthKit database. Loop uses the iPhone's HealthKit to read and write blood glucose, insulin, and carbohydrate data depending on how you initially setup this section.
At a minimum, you will need these HealthKit settings:
Loop
- insulin data: read and write
- carbohydrate data: read and write
- blood glucose data: read and write
Dexcom App
- blood glucose data: write (if you have this disabled, Loop will still function but will not display blood glucose values older than 3 hours old)
You can always check your HealthKit settings by opening the Health App, clicking on Sources at the bottom bar, and then clicking on the app's name you are interested in, for example, Loop.
Potential conflicts can arise when third-party apps are granted access to HealthKit permissions that may interfere with Loop's specified data permissions. For example, carbohydrate data is stored in Nightscout and Spike for some users...you wouldn't want to enable Spike app to write duplicate carbohydrate entries that Loop would be reading. Therefore, you should disable other apps from writing carbohydrate data to HealthKit so that Loop does not read those other entries unintentionally. Also, good practice, because carbohydrate entries in HealthKit that were created by non-Loop apps will not be able to customize carbohydrate absorption times nor be edited later if needed.
"},{"location":"data/health/#your-use-of-health-app","title":"Your use of Health App","text":"Summaries of your carbohydrates, insulin, and blood glucose results can be found by clicking on the Health Data at the bottom bar, and then selecting either the large Nutrition box (for carbohydrates) or smaller Results line (for insulin deliveries and blood glucose results).
If you toggle on the \"add to favorites\" slider for the individual data categories (insulin, blood glucose, carbohydrates), the data from those categories will be added to your Today view for easy quick reference and access.
The summary data for the categories can help you follow monthly trends, help identify periods of insulin sensitivity/resistance, evaluate total daily insulin use, breakdown of basal rate vs bolus insulin, and carbohydrate consumption. You can sort your data trends by day, week, month, or year views and scroll back through time in each of those data trends. You can even quickly use these data for endocrinology appointment discussions...as they provide the endocrinologist with a very quick and useful set of data points directly from your Loop.
If you drag two fingers separately like you are spreading them apart, you can get averages for the data set your fingers are covering, as shown below.
"},{"location":"data/nightscout/","title":"Nightscout","text":"Nightscout (NS) is a cornerstone of the DIY type 1 community. It is an excellent tool to remotely view Loop's actions and access insulin and blood glucose data. It allows for easy remote monitoring of Loop activities, troubleshooting of Loop errors, history of CGM, carbs, boluses, & temporary basal rates, and reports to analyze data trends and patterns. There is a Nightscout app in your iPhone App Store, or you can use a web browser to view the data. Setting up Nightscout for Loop is free and fairly quick. Once set up, the site can be accessed by anyone that you share your unique Nightscout URL with.
"},{"location":"data/nightscout/#nightscout-display","title":"Nightscout Display","text":"Nightscout is highly recommended for Loop users, especially those using Loop as caregivers. Nightscout displays are often the easiest way to troubleshoot Loop settings if you are having problems and seeking input from others. Below is some discussion about the general Nightscout display, as well as some Loop-specific display information.
"},{"location":"data/nightscout/#blood-glucose","title":"Blood Glucose","text":"Your blood glucose readings from your CGM are shown in green, yellow, or red in the main graph area of NS. You can adjust your high and low blood glucose targets in NS (when NS alerts will start for high/low BGs), but those will not affect your Loop performance. Loop only uses the blood glucose targets given in the Loop app settings. The main screen displays the time duration you have selected (in the example above, 12 hours of BG history). The very bottom of the screen shows the last 48 hours of blood glucose trends. You can scan backward by dragging the bottom timeline to the left if you want to review specific Loop actions or data in the last two days.
"},{"location":"data/nightscout/#carbohydrates","title":"Carbohydrates","text":"Carbohydrates are automatically uploaded to NS by the Loop app (when configured). The amount of carbohydrates on board (active carbs or COB) can be seen by clicking the Loop pill. The size of a white carbohydrate dot on the graph is proportional to the amount of carbohydrates entered...bigger meals get bigger dots. Loop does not read carbohydrates from NS (nor from the pump's bolus wizard) for use in looping calculations. Loop only uploads carbohydrates to NS that have been entered in the Loop app and recorded to HealthKit.
Insulin boluses are also automatically uploaded to NS by the Loop app. The bolus is shown as a filled-in blue lower half of the dot, and the specific amount of the bolus is also shown. There may be a slight delay between when the carbohydrates for a meal are uploaded to NS to when the meal bolus appears. This is because depending on the pump model and size of the bolus, the bolus may take a few minutes to actually be given and then Loop has to perform a pump read to verify the insulin delivery (i.e., the bolus was given). As shown in the example above, the meal carbohydrates have been uploaded, but the bolus delivery is still in progress. Insulin on board (active insulin or IOB) is also shown in the Loop pill.
Within your NS profile, you will enter in the scheduled basal rates from your Loop settings. This is not automatically updated when you update the Loop app; you will have to manually update your basal profile in your NS site if you change your basal profile in Loop. If you don't update the NS basal profile, it doesn't cause any problems in Loop function...you simply won't have an accurate VISUAL representation of high/low temporary basal rates within NS. The dashed blue line represents the scheduled basal profile (as entered in your NS profile). The solid blue lines indicate the actual basal amounts set for a given time...so as Loop sets temporary basal rates higher or lower than your scheduled basal rate, the solid blue line will jump above or below your dashed blue line. There is a slight delay (up to 2-5 minutes) in the rendering of basal changes on the blue lines, so if you want the accurate current basal rate, you should check the Loop pill.
"},{"location":"data/nightscout/#predicted-blood-glucose","title":"Predicted Blood Glucose","text":"The purple line to the right of the blood glucose readings is Loop's predicted blood glucose curve. Watching the behavior of that purple line can help you understand why Loop is making decisions regarding high or low temporary basal rates. You can read more on that topic in the Algorithm section of LoopDocs.
"},{"location":"data/nightscout/#loop-pill","title":"L\u200boop pill","text":"The Loop pill is the little display box that when hovered over or clicked, will provide additional information about recent Loop activities and status. The information included is the last time Loop ran, the temporary basal rate set, IOB, and COB. Looking at the Loop pill is a quick method for assessing if your Loop is currently active, as well. The small symbols to the right of the word Loop have a specific meaning. The small horizontal lightning bolt, shown in this example, is an active loop that recently enacted a temporary basal rate.
The SAGE, BAGE, and CAGE pills are for sensor age, pump battery age, and cannula age. These optional pills can track the time since your CGM sensor, pump battery, and pump site were last changed. You can set up custom alerts to remind you when it is time to change the devices, or simply use the visuals to keep track of your particular timing for site/sensor changes.
"},{"location":"data/nightscout/#nightscout-reports","title":"N\u200bightscout Reports","text":"You can access the Reports tab from within your NS settings (the three horizontal lines in the upper right corner of your NS site). There are several types of reports which may be useful to you and/or your health care provider. The report types are listed in tabs at the top of the Reports section, and you can also select the range of dates you'd like any report to cover.
"},{"location":"data/nightscout/#day-to-day-report","title":"Day to Day report","text":"The day-to-day report will show a detailed overlay of boluses, basal rates, carbohydrates, CGM, and treatment notes. If you select the optional check boxes, you can also see information such as the insulin distribution pie graphs shown on the right of the figure below.
"},{"location":"data/nightscout/#daily-stats-report","title":"Daily Stats report","text":"The daily stats report is a pie chart showing the daily breakdown of your low, in-range, and high time in target, as well as other statistical analysis of your BG trends.
"},{"location":"data/nightscout/#distribution-report","title":"Distribution report","text":"The distribution report is a combination of all the individual daily stats reports all into one pie chart for the date range selected.
"},{"location":"data/nightscout/#glucose-percentile-report","title":"Glucose Percentile report","text":"The glucose percentile report will help you see just how consistent your blood glucose is at various times of the day. The average blood glucose is shown as a dark black line in the center of the colored bars. The wider the colored areas spread out from the center black line, the more scattered and variable your blood glucose values at that time have tended to be.
For the graph above, for example, the blood glucose control around 4-6 am is very consistent. The most variable time appears to be near lunchtime. The time of day, when low blood glucose seems to be the most problematic, is between 4-6 pm, which happens to be this person's exercise time. Based on the data in this graph, the person may benefit from setting a higher target about 2 hours before exercise time in order to help with the pattern of low blood glucose that occurs most often during that time.
"},{"location":"data/nightscout/#treatments","title":"Treatments","text":"The treatments report will show a listing of all the temporary basal rates set by Loop, as well as boluses, carbohydrates, site changes, and any other \"treatments\" which have been entered into NS.
"},{"location":"data/overview/","title":"Loop Data","text":"In order to assess your Loop's performance and more specifically your settings, having access to your historical data trends is invaluable. Seeing what your Loop was predicting, what actions were being taken, how blood glucose was reacting, and how insulin needs have been changing is critical when trying to assess any settings changes.
"},{"location":"data/overview/#problem-with-traditional-methods","title":"Problem with traditional methods","text":"As a person with diabetes, you\u2019re probably carrying around a lot of different devices that are holding a lot of different data; a blood glucose meter for your finger sticks, a pump for insulin delivery, a continuous glucose monitor for real-time glucose measurements, phone app for tracking meals, etc. When you go to your endocrinology office, you probably start the process by dropping many of those devices at the front desk to be individually downloaded and then having to pack all of them away 20 minutes later.
Then your clinic staff has the less-than-efficient process of trying to overlay all those different devices into some sort of cohesive strategy for how your diabetes may need some tweaks. Because of Loop use, a clinic currently has to look at separate reports from Medtronic pump, Contour Next Link BG meter, Dexcom CGM/Clarity, and our iPhone Health app, as a typical example. There are also a couple of issues with Loop when using this separate downloads method:
-
Medtronic\u2019s pump gets so clogged up by the numerous temporary basal rate records being recorded that the clinic can only pull about 7 days of data from the pump at most.
-
Clinic basically lacks an overlay of basal rate/bolus actions with blood glucose from Dexcom, which really limits their ability to recommend any settings changes.
-
DIY solutions could fill some of the gaps as far as overall management, but the problem was with the actual gathering of the reports. It\u2019s a little hard for the clinic to bring up reports live time, and sometimes remembering to print them out ahead of time just doesn't happen.
-
Manual notes the various tools are not the most streamlined experience (especially for young kids and teens to manage). This inefficiency can lead to people neglecting to note times of site failure or compression lows, for example. If the health care provider wants additional information about a particular blood glucose event(s), chances are good that there would be no notations to help explain the event without effort. If there was a note, it may be in Nightscout, Dexcom, or in text messages between caregivers...not really centralized and integrated.
The good news is that the community has some very useful tools to help address the shortcomings in the traditional data aggregation for Loop users. There are several tools available to help you and/or your endocrinologist evaluate your Loop data. Specifically, this section covers:
- Health app
- Nightscout
- Tidepool
- Perceptus
Disclosure
The original author of this page, Katie DiSimone \ud83d\udc9a, was employed by Tidepool when this page was initially created. She used Tidepool before becoming an employee. She was not prompted to say anything in particular with regards to this information; it is written based on her own evaluation of Tidepool's services. Specific questions related to Tidepool should be directed to support@tidepool.org.
"},{"location":"data/tidepool/#tidepools-mission","title":"Tidepool's mission","text":"In Tidepool's own words, their commitment to diabetes data and accessibility is pretty clear:
\u200b
Tidepool's mission is to make diabetes data more accessible, actionable and meaningful for people with diabetes, their care teams and researchers.
We believe that connected data leads to better decision-making. Tidepool's free software liberates data from diabetes devices, and places it in context of the real world. Tidepool is designed to help you discover insights and bring context to your diabetes management. And, to help make your data more actionable, we allow you to share your data with anyone you choose: caregivers, clinicians, endocrinologists, friends, researchers \u2013 anyone.
Tidepool has announced that they are embarking on a project to build and support an FDA-regulated version of Loop, to be available in the iOS App Store, intended to work with commercially available insulin pumps and CGMs.
"},{"location":"data/tidepool/#tidepool-data","title":"Tidepool\u00a0 Data","text":"Once you start uploading data into your Tidepool account, you'll see some of the benefits pretty quickly. Tidepool's data presentation is easy on the eyes. Lots of calm colors, logical data layout, quick to access important information...basically a breeze to get your needed data and get on with your analysis.
To use Tidepool's services, the process is pretty standard:
- Create a free Tidepool account to become the warehouse for your diabetes data.
- Upload device data to your account.
Children with type 1 diabetes can have their own data account, and parents can have a caregiver account. Users can also choose to share data with their endocrinology clinic. You can also choose to share your diabetes data anonymously for the support of type 1 diabetes researchers if desired.
"},{"location":"data/tidepool/#how-do-you-upload-data-into-tidepool-account","title":"How do you upload data into Tidepool account?","text":"There are two methods for uploading different data to your Tidepool account:
-
Tidepool Uploader (program on your computer) - Tidepool Uploader supports a large variety of current diabetes devices in the market; Animas, T-slim, Medtronic, Omnipod, Dexcom, and more. The full list of devices is here. To upload the data from these devices, you download the free Tidepool Uploader program, connect your device to the computer according to the instructions provided in the Uploader, and your data is then securely uploaded and stored in your Tidepool account. Most users will perform uploads on a semi-regular basis; weekly, monthly, or prior to each clinic visit.
-
Tidepool Mobile (app on your phone) - Tidepool Mobile is available for both Android and iOS/iPhone users. Tidepool Mobile is a companion app to Tidepool that lets you add notes about meals, exercise, or anything else. See your CGM, pump, and meter data alongside any notes you add. Track your favorite meals and your regular exercise, and learn from what happens.
In summary, the Uploader is for uploading the numeric data from your various diabetes devices, and the Mobile app is for adding contextual information and notes-on-the-go to make your numeric data more useful.
"},{"location":"data/tidepool/#does-tidepool-uploader-work-for-loop-users","title":"Does Tidepool Uploader work for Loop users?","text":"There has been a lack of ability to get Loop data into Tidepool using the Tidepool Uploader. The only pumps compatible with Loop and supported in Tidepool Uploader are the x23/x54 pumps, so users of the x15 and x22 pumps cannot upload their pump data. Furthermore, the Medtronic pump still suffers from the lack of internal storage sufficient to store more than 3-7 days' worth of Looping data.
However, there has been a very exciting development! Tidepool's Mobile iOS app has been updated to upload Apple's HealthKit data into Tidepool. Since Loop uses HealthKit to store all of your insulin, carbohydrate, and CGM data, this means Tidepool can now upload the full suite of available Loop information. So, there's no need to use Tidepool Uploader for Loop users...their Tidepool Mobile app can take care of the needed connections. Read below for lots of details about this important development.
"},{"location":"data/tidepool/#how-can-i-share-my-tidepool-data","title":"How can I share my Tidepool data?","text":"Sharing the data is simple. You can click on your account\u2019s Share option and enter in the email addresses for those that you want to share with. Those people will need a Tidepool account. If they don\u2019t have one currently, they will follow easy prompts for an account setup after they\u2019ve received your share invitation. Clinics using Tidepool will have a Tidepool account email that you can add to your account, enabling the clinic to easily view your data. You can also remove access for anyone with a simple click.
"},{"location":"data/tidepool/#viewing-your-tidepool-data","title":"Viewing your Tidepool data","text":"There are two distinct viewing options for your Tidepool data, and they are not identical. You can either view your data:
- Using a desktop computer using the Chrome browser and logging into your Tidepool account, or
-
Using your Tidepool Mobile app AND associated notes.
Supported Desktop Browser
Currently, only the Chrome desktop browser is compatible with Tidepool. You will not be able to use Safari or other browsers to view the data, nor will you be able to use the Chrome application on a phone. If you want to see your data on a computer, you need to use the Chrome browser.
Viewing Data on your Phone
If you want to see your data on your phone, you must use the Tidepool Mobile app and use an associated note to view the data surrounding the note's time frame. Mobile data viewing is not a live stream, but rather available as \"bookmarked\" time periods anchored by notes.
You will see your Loop's temporary basals, CGM readings, boluses, notes, carbohydrates, and various metrics about your data distribution. If you separately load your blood glucose meter or any other supported device to Tidepool, those will also overlay.
"},{"location":"data/tidepool/#what-loop-data-will-you-see-in-the-tidepool-mobile-app","title":"What Loop data will you see in the Tidepool Mobile app?","text":"The Tidepool Mobile app is not a live-viewing app for looping data. For people coming from Nightscout, this may be a bit confusing but realize the intended purpose of the Tidepool Mobile app isn\u2019t live-viewing. It is the place where you can (1) upload/sync HealthKit data, (2) easily add/edit/delete notes to the data set, and (3) search for notes and view Loop data surrounding that note.
In fact, you will basically see NO data in the Tidepool Mobile app unless you have Tidepool data uploaded and notes are added. Once you add a note, you are basically placing a bookmark on the data set. You will be able to click on the note and see 7 hours of old data before the note, and then the note will continue to collect 7 hours of data to display after the note. So, technically, you\u2019ll be able to refresh the app\u2019s view to see current data for approximately 7 hours after a note is placed.
For example, here\u2019s a sample note+data set below from my Tidepool Mobile app. Over the last couple of hours, my daughter noticed that she was staying above target (unusual for her on Loop with the meal she had) for quite a while. She had given a couple of small corrections (see the two 1-unit correction boluses?) without result. She started her secondary troubleshooting\u2026if it\u2019s not the food, maybe it\u2019s the infusion site? She realized it has been 4.5 days since changing her site. She changed the site and logged a note using the Tidepool Mobile app. That note appears in the Tidepool Mobile application, on my phone. It also shows up on her Tidepool's data, for her endocrinologist to see too, and we can refresh the view to see how blood glucose trends for the next 7 hours after the site change.
"},{"location":"data/tidepool/#tidepools-mobile-app-for-iosiphone","title":"Tidepool's Mobile app for iOS/iPhone","text":"As part of Tidepool Loop development, Tidepool updated its Tidepool Mobile iOS app to integrate with Loop user's data stored in Apple's HealthKit database. This update also supports the Jaeb Center observational study of Loop.
"},{"location":"data/tidepool/#how-does-the-loop-data-upload-work","title":"How does the Loop data upload work?","text":"The Tidepool Mobile iOS app will sync Loop's diabetes-related HealthKit data into your Tidepool account. The app will continuously upload that data so long as the Tidepool Mobile app is open, even if it is only open in the background. That data will then be viewable in two places: on the Tidepool Mobile app itself when using notes, or in your Tidepool account using the desktop Chrome browser (note: the two options are not identical viewing platforms, see discussion above). This means you will not have to plug any of your devices into a USB cable in order to upload your information to Tidepool.
"},{"location":"data/tidepool/#is-the-tidepool-mobile-app-a-replacement-for-nightscout","title":"Is the Tidepool Mobile app a replacement for Nightscout?","text":"Nope. This was not designed nor intended to be a replacement for your Nightscout site. They complement each other, rather than compete. This new upload of Loop data will allow you and your clinics to have a powerful tool to analyze Loop data through an easy-to-use, shared hub.
"},{"location":"data/tidepool/#are-there-any-known-issues-with-loop-and-tidepool-mobile-app","title":"Are there any known issues with Loop and Tidepool Mobile app?","text":"The settings for Loop do not show up in Device Settings in Tidepool Web. This is because this info can't be uploaded from Apple Health. If you used Tidepool Web for your Medtronic or Omnipod before, you will still see the pump settings from before you started looping.
You mean, what OTHER cool thing can the Tidepool Mobile application do besides automatically uploading the data from Loop? Tidepool Mobile also provides an easily searchable log of meal boluses. If you are still learning new meals in Loop\u2026how much to bolus, how long of a carb absorption\u2026these notes are searchable and super easy to add. Learning how to bolus for that Tofu Breakfast Burrito? Simply record a note of how you bolused for it. If you want to, come back afterwards and leave yourself some suggestions for the next time to try. This searchable information can also help teens learn and exercise independent skills. If they aren\u2019t sure of how to bolus for a meal, this could give them easy tips from past success without necessarily having to stop and ask a parent. As much as a parent might scoff at the idea of a kid looking up a meal, if the alternative is asking a parent\u2026that might be all the motivation it takes. How about co-parenting? Want to leave a note that another parent or caregiver can look up? \"How were the last Chicken McNuggets bolused?\" or \"When was the last site change?\" can easily be tracked and retrieved with notes.
For an easy example, search for the word burrito (doesn\u2019t have to be a hashtag), and any notes with the word \u201cburrito\u201d will be available for review, as well as any added comments.
Hormones can also be easily tracked with notes. What day-of-the-month and how did I change the basals? Looking to find patterns in those female hormones? This could be a really slick tracking tool to easily log periods of insulin resistance and what part of the cycle they are occurring at.
"},{"location":"data/tidepool/#is-the-tidepool-mobile-app-only-for-loop-users","title":"Is the Tidepool Mobile app only for Loop users?","text":"The Tidepool Mobile application uploads diabetes-related HealthKit data regardless of the source. Loop users store their data in HealthKit, so this is a nice fit. Other diabetes devices (e.g., OneDrop BG meter) and apps (e.g., Spike and Dexcom) also store their data in HealthKit. Some people even manually enter their diabetes data into HealthKit. For all those uses, the Tidepool Mobile application will upload the HealthKit data. OpenAPS does not store its data in HealthKit, so this will not upload OpenAPS-related data. However, if OpenAPS users are using Spike app, some integration of information through Nightscout is possible. Currently, OpenAPS SMBs, boluses, and carbohydrates can be uploaded to HealthKit for OpenAPS+Spike+Nightscout+iPhone users. Spike does plan to add temporary basal integration to HealthKit at some time in the future.
"},{"location":"data/tidepool/#how-can-you-get-the-tidepool-mobile-app","title":"How can you get the Tidepool Mobile app?","text":"Tidepool Mobile is available in the iOS App Store on your Loop device.
"},{"location":"how-to/bolus/","title":"Extended or Combo Bolus with Loop","text":"The majority of meals have most of their blood glucose impact within 2-3 hours after eating. Complex carbohydrates are slowed by their fat and protein content and can lead to extended time of blood glucose impacts. Many people with type 1 are familiar with the late blood glucose spikes from Chinese food, pasta, pizza or burritos. This extended blood glucose impact can be tricky to properly bolus for in traditional one-bolus insulin delivery.
In traditional multiple daily injection therapy, these complex meals may require additional insulin boluses to help control long, slow absorption meals. The trick is to try to time the second bolus at a time when blood glucose is starting to rise and it can be tricky to estimate how much to give to control the tail end of those slow meals.
If you're using traditional pump therapy, one technique is to use an extended bolus or dual wave bolus for your complex meal's insulin. In this method, a user give a portion of the insulin up front as an initial bolus and the remainder of the insulin is delivered as a psuedo high temp basal for a user-set duration of time. For example, you may give 5 units of bolus up-front and then \"extend\" 4 units of insulin over the next 60 minutes. In this way, you are providing a more complex bolus to help match the timing of the meal's blood glucose impacts; avoiding early low blood glucose and addressing later high blood glucose.
The transition to Loop use may be confusing at first for these meals since you cannot use an extended/combo bolus and simultaneously have Loop set temporary basals automatically. The good news is that Loop has a bolus calculator that has the ability to emulate an extended bolus situation...and it's implemented with the pizza icon (or any custom carbohydrate absorption that is set longer than 4 hours).
"},{"location":"how-to/bolus/#pizza-icon","title":"Pizza Icon","text":""},{"location":"how-to/bolus/#example-long-absorption-time-bolus","title":"Example long absorption time bolus","text":"For an example of Loop's bolus adjustments using carbohydrate absorption time, let's take a look at an example meal. This is an example of a long absorption complex carbohydrate meal. This is a mushroom (arborio) risotto dish with heavy cream and cheese ingredients. While some white rice can be fairly quick acting, after several times eating this dish, the family has noticed that the meal tends to have a longer duration of impact on blood glucose. Using a \"taco\" icon (3 hours absorption) was causing slight low blood glucose soon after eating the meal. Therefore, they have been using the pizza icon to enter the meal's carbohydrate absorption time.
The initial meal entry was 70g at a \"pizza\" icon aborption time (4 hours). Based on carbohydrate ratio of 8 g/U, the initial bowl of risotto at 60g should have been a bolus of 7.5 units. Loop recommended 5.3 units, or about 70% of the total bolus that would be needed to cover the total carbohydrates. Loop recommended the lower upfront bolus because a full bolus would have overwhelmed the slow absorption of carbohydrates, and the likelihood would be a low blood glucose shortly after eating.
As the meal was being absorped, Loop was tracking the carbohydrates still remaining to be absorped and expecting that blood glucose values would be rising soon (knowing that there was still insulin needing to be delivered once safely passed the near-meal low blood glucose potential). Loop would have provided the additional insulin via high temporary basals after seeing blood glucose impacts which indicated the potential for a low blood glucose had passed.
The Loop user then had a second, smaller bowl of risotto about 90 minutes later, and entered 30g at 4 hours absorption again. Notice this bowl of risotto had a bolus recommendation much different than the original bowl. The second bowl had a recommendation of 5 units, much greater amount of bolus relative to the amount of carbs entered than the first bowl had received, and more than the carbohydrate ratio alone would provide (3.75 units). Why the \"extra\" 1.25 units? Because Loop was including some extra bolus amount to cover what it predicted could safely be provided from the amount not given in the original bowl's bolus (the original bolus was approximately 2.2 units short of carbohydrate ratio alone recommendation). If the user had not had the second bowl, Loop would have been providing high temporary basals as soon as blood glucose had exceeded the correction range. And in fact, you can see that Loop still provided the remaining insulin via high temporary basals as blood glucose rose after the second bowl, in effect making up the small remaining difference.
"},{"location":"how-to/bolus/#low-carbhigh-fat-or-keto-diets","title":"Low carb/High Fat or Keto diets","text":"The example meal above, while relatively high carbohydrate, also helps illustrate how Loop can be used to bolus effectively for low carb/high fat (LC/HF) or Keto diets. Those diets tend to have low glycemic index food with a relatively long blood glucose impact. Said another way, they don't spike blood sugar as much, but often need additional insulin after the meal was eaten to account for the slow conversion of protein to glucose.
To account for those dietary differences, there are two useful strategies:
- convert some of the protein and fat to \"equivalent carbohydrates\" and
- extend the duration of those carbohydrates using a pizza icon or even longer, depending on the person/food.
Most LC/HF or Keto users will convert a portion of their fat and protein content into an equivent carbohydrate content. So while an example meal might only have 5g of carbohydrates based on nutritional labels, they may convert 25% of the 20g fat and 50% of the 20g protein grams into an additional 15g of \"equivalent carbs\" for bolusing purposes. The percentages that people use to convert fat and protein will usually be a bit of trial-and-error, but there are some published articles (here, here, here, and here) that may be helpful starting points, if you are interested.
"},{"location":"how-to/cgm/","title":"CGM data","text":"Obviously, without CGM data your Loop will not function. With the stability that Looping brings, many people are looking for ways to minimize their Loop downtime due to avoidable factors. Factors that make CGMs a potential problem include sensor failure, transmitter signal loss, poor calibrations, and the warmup period.
"},{"location":"how-to/cgm/#sensor-warmup","title":"Sensor warmup","text":"One of the obvious downtimes is during new sensor warmup. Unfortunately, the only way to avoid the 2-hour warmup (in the case of Dexcom's system) is to overlap your sensor sessions and this just isn't feasible for most people.
- Insurance doesn't provide adequate supplies to overlap sensor supplies; you would need two active transmitters and enough sensors to allow for overlap of use.
- Sometimes sensors fail before a session was due to expire, making a planned overlap impossible.
- Some people just don't want to wear two sensors at once.
Since warmups are an important part of stabilizing CGM values and decreasing sensor noise, they are just going to continue to be a no-Loop time period for the foreseeable future. Most people simply cancel a currently running temporary basal, or let it run its course, when they start a new sensor session. For the two hours of warmup, the Loop will revert to the basal schedule programmed into the pump.
Some people have opted to keep looping during sensor warmups by manually entering blood glucose checks through the Apple Health app for Loop to read. While technically Loop can use those values, the entries would have to be sufficiently frequent to be meaningful. If a person is willing to test 8-12 times in a two hour span, then looping through sensor warmup could be feasible through this method.
"},{"location":"how-to/cgm/#sensor-noise","title":"Sensor noise","text":"Depending on which CGM device you use, you may observe a sweet spot of data accuracy in the middle of your sensor's session but have sensor \"noise\" at the beginning and end of the session. Generally speaking, Loop does fairly well despite these times of sensor noise, although having a reasonable maximum basal rate does help mitigate the risk due to a noisy sensor.
The figure below shows an example of a new G6 sensor for approximately its first 21 hours of use.
As you can see on the left, the red dot was a finger stick blood glucose check and it nicely aligned with the CGM value when the session first started to provide live data. However, between that first value and approximately noon the next day, there was a considerable amount of jumpiness to the data (as well as some suspected compression lows or extreme sensor jumpiness in the first 6 hours). The CGM noise immediately decreased after the first calibration was provided at the noon-time meal bolus. (Personally, we give the G6 one calibration point about 8-12 hours after a session starts because we have noticed that it helps settle down the noise quickly. This is not a part of the standard Dexcom protocols.)
We similarly see an increase in sensor noise at the end of a sensor's useful life. The figure below is a G6 sensor at 9.5 days, but your end of sensor symptoms may occur differently and at longer/shorter days. Much of the specifics regarding useful sensor life seems to depend on the user's body chemistry and sensor insertion.
There is a marked increase in sensor noise and scatter, as well as several periods of sensor error that lead to lost CGM data. We opted to pull this sensor just shy of the 10 days due to this noise. However, as you can see, the Loop was still doing an aequate job controling overall blood glucose fluctuations despite the erratic CGM data.
"},{"location":"how-to/cgm/#compression-lows","title":"Compression lows","text":"A frequent question from people before starting Loop is \"How does Loop deal with compression lows?\" If you aren't familiar with compression lows, they are false low blood glucose alarms caused by sustained pressure on the sensor area. In effect, the phenomenom is much like resting on an arm for too long and causing it to fall asleep from poor blood flow.
The figure above showing an example of a compression low. CGM data shows blood glucose dropping low, but finger checks on a meter would confirm that the CGM data is falsely low. Often, Dexcom G5 and G6 will stop providing CGM values for awhile when their algorithm detects a suspected compression low. Once the person rolls off the sensor area and blood starts flowing well again, the CGM values come back online to a more reasonable tracking again.
Generally speaking, Loop will deal with compression lows just fine. Loop will suspend for a short time during the low blood glucose values, and then recover with temporary high temporary basals to make up for the active insulin it had missed as result of the compression low. No special actions need to be taken for compression lows.
"},{"location":"how-to/cgm/#calibration-jumps","title":"Calibration jumps","text":"It is also worth noting that Loop will not calculate blood glucose momentum in instances where CGM data is not continuous (i.e., must have at least three continuous CGM readings to draw the best-fit straight line trend). It also will not calculate blood glucose momentum when the last three CGM readings contain any calibration points, as those may not be representative of true blood glucose momentum trends. Therefore, calibration jumps or missing data will not adversely affect the insulin dosing based on blood glucose momentum.
"},{"location":"how-to/cgm/#loss-of-cgm-data","title":"Loss of CGM data","text":"Transmitter failure, sensor failure, or other equipment failures can lead to a failure in CGM data. In those cases, Loop will revert back to the pump settings after any currently running temporary basal finishes or is manually canceled by the user.
"},{"location":"how-to/cgm/#new-transmitter-or-reset-transmitter","title":"New Transmitter or Reset Transmitter","text":"-
If you begin a new transmitter, don't forget to update your transmitter ID in your Loop settings.
-
If you are going to reset a transmitter, don't forget to remove the transmitter ID from Loop settings during the reset and re-pairing of the transmitter. Do not add the transmitter ID back into Loop settings until you've completed the reset process and re-paired with the Dexcom app successfully.
For Medtronic pump users, you will have times where you need to disconnect from your insulin infusion site. Showers, some sports activities, swimming, and medical exams are some of the common reasons when disconnections are unavoidable.
The simplest way to deal with periods of disconnect is to suspend the Medtronic pump so that Loop knows no insulin was being delivered while you were disconnected. This will let Loop know that there is missing basal insulin that may need to be considered as part of the blood glucose prediction. Typically when disconnected from insulin delivery and negative insulin on board has developed, blood glucose will start to rise so Loop will expect to need to make up the missed insulin from disconnections.
Some users are prone to forget to resume insulin delivery after suspending for disconnects. One option is to use Siri by simply saying \"Hey Siri, set a timer for 30 minutes\" and that can serve as a reminder alarm to resume the pump after the shower is over. Other users may want to consider canceling any running temporary basal, setting a 30 minute zero temporary basal, and then setting Loop to open loop. This will keep IOB fairly accurate while at the same time providing an automatic restart of normal basals if the user forgets to resume delivery. The user will then have to close loop again when ready to resume looping. Find what option works for you (or your child) the best; there are a number of ways to deal with disconnects.
For disconnects where the disconnected time coincides with exercise that may increase insulin sensitivity (like swimming), a good strategy after reconnecting can be to set a higher temporary override target (aka workout mode) for some time after reconnecting and resuming insulin delivery. This will help Loop be a little gentler with high temporary basals during the time while the user is still experiencing increased insulin sensitivity after exercise. The length of time to set that override target can be a bit of trial-and-error as it will depend on the person and type of exercise.
"},{"location":"how-to/endo/","title":"Endocrinologist and Loop","text":"There is a wide range of comfort within the health care profession for DIY projects and type 1 diabetes. Loop is no different; some health care providers are comfortable and some are distrusting. As with all things health care related, honesty with your health care provider is paramount so that the best decisions can be made.
Since using Loop often requires people to secure access to pump supplies that they may not have otherwise have, often times the first conversation about Looping with an endocrinologist will be when trying to get prescriptions for Medtronic or Omnipod supplies. The best approach is to start with objective information. Find out if the clinic has other patients already Looping and whether or not they have concerns from what they have seen. Provide links to the LoopDocs site and this site so that your endocrinologist or health care provider can read about the system's algorithm and operations.
Most often the important aspects that need discussion are:
- Loop will revert to normal pump use, within a maximum of 30 minutes, in the event of Loop failure. What happens during failure of the system is the most common concern your health care provider will likely have when initially learning about the system.
- Loop depends on your basal rate, ISF, and carbohydrate ratio settings, just like traditional pump therapy. The conversations will not need to change regarding the evaluation and adequacy of those settings.
- Duration of insulin action is automatically set to 6-hours for the rapid-acting insulin models built into Loop, matching published data for those insulins. The settings in the pump for duration of insulin action are not used by Loop.
- Reasonable values for the suspend threshold and maximum basal rate should be discussed. These two settings are not part of normal pump therapy and should be discussed with your endocrinologist. Typically to start Looping, setting the maximum basal rate not much above your highest scheduled basal rate is an excellent way to get used to Loop and safely double-check that your settings (ISF, carbohydrate ratio, and basal schedule) are reasonably accurate. As you gain experience and confirm your settings, raising the maximum basal rate can help Loop become more effective to correct rising blood glucose more quickly.
- Develop a plan for how you will be providing Loop data to your health care provider in advance. Ask what information they would like to review so that you can have reports ready at each appointment. The usual downlaod of a Medtronic pump or Omnipod PDM will not be sufficient for many clinics once you start Loop. Instead, you may wish to discuss Nightscout or Tidepool data reports to provide the information those downloads would have otherwise provided.
- Develop a plan for how you will deal with pump failure. Many health care professionals will cite the age and out-of-warranty aspects as concerns for Loop users. Having a clear plan, in advance, for how to deal with possible pump failure will help with that concern. The two most obvious steps would be to have a backup pump available and an active prescription for long-acting insulin. A gentle reminder to the clinic that even in-warranty pumps can fail also helps frame the conversation regarding the need to always be prepared, looping or not.
Hands-down, exercise is the most commonly discussed topic of long-term Loop users. Exercise is highly specific to an individual and its impact on blood sugar is not always equal. For example, many people find anaerobic exercise such as weight lifting will raise blood glucose, and aerobic exercise such as running will lower blood glucose. The degree to which impacts are observed also depends on the insulin and carbohydrates on board going into the exercise. Blood glucose response in competition can be quite different than a regular training session due to the added adrenaline. Exercising with T1D does require extra preparation, but with preparation, things should go much easier.
"},{"location":"how-to/exercise/#exercise-basics","title":"Exercise Basics","text":"A 2013 JDRF survey found that 36 percent of T1D patients, caregivers and healthcare professionals view exercise as a challenge and would like to learn more about exercising safely. The JDRF PEAK (T!D Performance in Exercise and Knowledge) education curriculum explores the environmental, dietary, physiological and psychosocial elements that impact physical activity with T1D (source). As part of JDRF's PEAK resources, they have published a helpful article about exercise and T1D for athletes. Below is an excerpt from that article. There are a couple key considerations (highlighted in orange) that be different in Loop use. Those considerations are discussed further below.
With great exercise, comes great responsibility
It goes both ways: exercise affects glucose control; glucose control affects exercise. Consistent physical activity produces consistent insulin sensitivity, which helps the body process glucose more evenly (avoid spikes), maybe even improving HbA1c.
Some considerations:
- Risk for hypoglycemia can be reduced by avoiding too much insulin on board (IOB) during and after exercise.
- Continuous glucose monitors (CGMs) or regular blood glucose testing provides critical information to preemptively minimize blood-sugar highs and lows.
- Plan ahead! Reduce insulin and add carbs for aerobic (longer duration, lower intensity) exercise, which causes blood sugar to drop, or increase for anaerobic (short duration, high intensity), which may cause spikes.
- After most exercise, insulin sensitivity is higher for 24 hours. Carbohydrates needed to replace liver glycogen levels must be balanced with decreased insulin levels.
Intensity. Speed. Environment. Duration\u2026 all affect the body\u2019s reaction
There are three types of exercise:
- Aerobic
- Aerobic exercise is usually continuous, light-to moderate intensity exercise and includes activities like running, walking, long-distance swimming, biking or tennis. Aerobic exercise typically tends to lower blood glucose levels.
- Anaerobic
- Anaerobic exercise is generally shorter in duration, of maximal to super max intensity and includes activities like sprinting, gymnastic, ice hockey, or weight training. Anaerobic exercise typically tends to raise blood glucose levels.
- Mixed
- Mixed exercise is combination of the aerobic and anaerobic activity such as basketball or soccer. Managing blood glucose levels with mixed exercise is difficult, but using a tool like a continuous glucose monitor can help greatly.
Glucose levels during sports affect performance in many ways: strength, stamina, speed, agility, flexibility, safety and mental sharpness.
- Insulin resistance, which is when the body requires more insulin to process glucose, is a problem. Exercise combats this by increasing insulin sensitivity and glucose uptake by muscles. Muscles are better able to absorb glucose when contracted.
- Exercise thwarts postprandial (after-meal) hyperglycemia by slowing carbohydrate absorption, increasing glucose utilization and accelerating insulin action.
- Practice vs game day \u2013 each can have dramatically different effects on your T1D. Adrenaline and stress both raise blood glucose levels and are typically present on the day of your game or event.
Balancing Act: Finding your blood glucose zone and peak fitness zone
Caution: Delayed Onset Hypoglycemia can occur typically 6 to 12 hours (sometimes 24 hours) after exhaustive exercise, due to replenishment of muscle glycogen stores and enhances insulin sensitivity. It is possible to prevent this by:
- keeping records of your fitness regimen and the effects it has on your body
- using a CGM or frequent blood testing to monitor your body\u2019s response to activity
- decreasing basal insulin or meal boluses after activity
- consuming slow-acting carbohydrates after activity
Similar recording and monitoring tactics can be used to prevent hyperglycemia (blood-sugar highs). Some other options include:
- bolusing (at a half dose) 30 to 60 minutes prior to activity to offset rise of blood glucose, delayed bolusing or adding post-workout cool-down (depending on conditions of rise)
- incorporating relaxation, breathing, visualization tools to address pre-workout emotional stress
- limiting pump disconnection time
- hydrating
- administering rapid-acting insulin (only under certain conditions)
With all of that background laid out, your Loop can help with some of the techniques described above. Highlighted in orange above are parts of those recommendations and considerations that we should explore a little more for Loop in particular:
Risk for hypoglycemia can be reduced by avoiding too much insulin on board (IOB) during and after exercise. -- In Loop use, we can accomplish this through setting a workout target far enough ahead of the exercise to help decrease basal insulin on board going into exercise. Typically this means setting a workout target at least 1, if not 2, hours prior to exercise. Setting a workout target just as you begin a workout will not allow sufficient time to decrease IOB ahead of an aerobic workout. Loop workout targets are usually 50 mg/dL higher than your usual correction target, although that is a guide. Practice and observation of your typical exercise needs will help refine your optimal settings. And while not Loop specific, minimize heavy carb meals in the 3 hours before intense exercise that would require large boluses. If eating during that time, consider partial boluses and slower, lower carbohydrate snacks. If bolusing while workout targets are enacted, Loop will automatically recommend a smaller bolus than you would get with your normal correction target.
After most exercise, insulin sensitivity is higher for up to 24 hours. Carbohydrates needed to replace liver glycogen levels must be balanced with decreased insulin levels.-- For Loop use, this period of increased insulin sensitivity may require continuing or setting a workout target for some period of time after exercise is finished. This will help manage increased sensitivity by providing smaller boluses and less basal delivery to prevent hypoglycemia that could result from unadjusted settings under traditional therapy. The length of time to run a continued post-workout target will depend on the duration and intensity of exercise.
"},{"location":"how-to/high-bg/","title":"Stuck on High BG","text":"Safety consideration
If the continued delivery of corrective insulin fails to bring down high blood glucose, remember to follow your doctor's recommendations for ketone testing, hydration, and/or more frequent blood glucose monitoring. Check for equipment failure (infected pump site? blocked cannula?). Try fresh insulin. Contact your doctor immediately if you are experiencing sustained high blood glucose despite continued insulin delivery and corrections.
\"How do I deal with high blood glucose that seems to be stuck?\" When you and your Loop are disagreeing about the effectiveness of high temporary basals to bring down blood glucoose, take a deep breath. Unfortunately, most of the time our disagreement is rooted in the impatience that the pump's insulin works so dang slow. We get tired of waiting another 4-6 hours until blood glucose will be back in range after a correction. So before we start overriding Loop's predictions with manual actions, it may be useful to ask yourself some questions first to help guide potential corrective actions:
- Is this temporary or has this been going on for a long time?
- Is this consistently only after meals?
- Have you open loop tested your settings?
- Are you sick or stressed?
- Are you hormonal?
- Are you taking medication?
- Could this be caused by site failure or bad insulin?
Based on the answers to those questions, you may want to choose on of the potential solutions for delivering extra insulin to help:
- Setting lower temporary blood glucose targets
- Providing the \"pending\" insulin upfront via correction bolus
- Adjusting meal entries
- Fake carbohydrates
- Priming in insulin
- Open loop and correct manually
- Adjust settings
If you to believe the high blood glucose is a short-term issue, due to acute stress or anaerobic exercise for example, then using temporary lower override targets will help Loop be more aggressive at providing additional insulin. The pre-meal target would be a quick, easy implementation of this tool.
This technique, and the delivery pending insulin as a bolus (described below), are good to help control the urge to \"rage\" bolusing which often yields an overcorrection and subsequent low blood glucose. These two techniques used in combination are often successful for most short-term situations or meals that are just taking a long time to \"burn out\".
"},{"location":"how-to/high-bg/#pending-insulin-as-bolus","title":"Pending insulin as bolus","text":"To help correct a high blood glucose, Loop will apply a high temporary basal set to deliver over 30 minutes time. If you are stuck on a high blood glucose and want to speed up the correction, you can try to \"hurry\" along the correction by opening Loop's bolus tool. Within the information screen, you will see a \"pending insulin\" amount. That amount represents the amount of corrective insulin Loop plans on delivering with the currently running 30 minute temporary basal. You can choose to manually enter in that amount on the bolus deliver line as an upfront correction bolus, rather than waiting for it to be delivered over the next 30 minutes. This generally represents a pretty easy, safe way to help reasonably speed up a correction.
The safety consideration for this is that you do not want to unintentionally stack insulin by using this technique repeatedly in a short period of time. Wait a reasonable amount of time, such as an hour or two, to give insulin a chance to start to affect blood glucose. If the correction still hasn't helped, you may want to consider another technique to help bring down blood glucose.
"},{"location":"how-to/high-bg/#adjusting-meal-entries","title":"Adjusting meal entries","text":"Carbohydrate, fat, and protein estimations for bolusing is frequently a guessing game...and we've all guessed wrong before. If your blood glucose is remaining high due to a food guess gone wrong, you can definitely go back in time and adjust your meal's carbohydrate entry to more accurately reflect what you suspect it should have been. By adjusting the original entry, you'll help Loop know that the blood glucose effects it needs to control are greater (total carb entry increase) and/or longer (carbohydrate absorption time extension). Often you'll find Loop will offer an additional bolus immediately after you save your adjusted carbohydrate entry and/or the temporary basal adjustments will be more aggressive.
"},{"location":"how-to/high-bg/#fake-carbohydrates","title":"Fake Carbohydrates","text":"Some people use \"fake\" carbohydrate entries to push Loop to be more aggressive with correcting a high blood glucose. This can be effective for hyperglycemia due to short term causes, but is not the recommended way of repeatedly dealing with high blood glucose. If you find yourself regularly using fake carbohydrates often, you may want to consider whether one of your underlying settings needs adjusting.
"},{"location":"how-to/high-bg/#prime-insulin","title":"Prime insulin","text":"If your Preferred Data Source is left on the default \"Event History\", Loop will not count insulin delivered from \"prime\" commands in the Medtronic pump. This means that any insulin you deliver through your infusion set while using the prime delivery will not count towards active insulin or insulin on board (IOB).
If you think that you are being affected by a short term influence that will last for at least 4-6 hours, a conservative amount of primed insulin may help control high blood sugar more aggressively than Loop would have otherwise provided during the stressful event.
If you choose to deliver insulin via prime command, remember that your active insulin or IOB will not be accurate for 6 hours after the delivery. Caution should be taken using this method because if the short term stress suddenly ends and blood glucose begins to drop again, the drop may be more pronounced than Loop predicts due to the untracked primed insulin.
Note: For Omnipod users, Loop does not allow you to prime insulin in.
"},{"location":"how-to/high-bg/#open-loop-and-correct-manually","title":"Open loop and correct manually","text":"Don't forget that you can always go old-school and open your Loop until the cause of the high blood glucose is sorted out and you can get back to normal operations. Site failures are an especially good example of when open looping plus manual correction may be a less frustrating way of dealing with Loop during times when prediction just isn't meeting reality. Give IOB a chance to get back to reality by letting open loop run for 4-6 hours and you'll come back after site failure with a more functional Loop prediction.
"},{"location":"how-to/high-bg/#adjust-settings","title":"Adjust settings","text":"Finally, if your underlying cause doesn't appear to be short term (no site failures, no stressful event, no illness) or food related (no recent meals, happens even overnight), then you may need to adjust your underlying settings to help Loop get a more accurate prediction curve working for you.
Long duration \"stuck\" high blood glucose most often means you'll need to increase your underlying basal rate(s). If Loop's high temporary basals aren't making meaningful progress towards your correction range within 4-6 hours, and you don't suspect food, stress, or site failure, then a basal increase is likely a good place to start. Many people start with a 10% basal adjustment to begin with and wait to see the effectiveness before making any additional adjustments.
What kind of situations are most likely to need this adjustment? Monthly hormonal cycles and steroids are good examples of when underlying basal rate adjustments for a few days could help tremendously until things return to normal.
Consult your health care provider if you are at all uncomfortable self-adjusting basal rates.
"},{"location":"how-to/high-bg/#maximum-basal-rate","title":"Maximum basal rate","text":"It is worth mentioning one adjustment that will not be helpful in these situations. Increasing your maximum basal rate is very unlikely to resolve a stuck high blood glucose. Stuck high blood glucose is the result of Loop predicting that the insulin on board is sufficient to cover the needed correction. The issue is almost never that Loop is being limited by a maximum basal rate, but rather that Loop is predicting that the insulin it has been delivering for awhile is adequate. In other words, Loop doesn't even think that it needs to reach that maximum basal rate. The better adjustment would be to increase your scheduled basal rates, and not the maximum basal rate listed in Delivery Limits.
"},{"location":"how-to/injections/","title":"Test Your Settings","text":"Open Looping Fine-Tuning Settings Book settings
"},{"location":"how-to/iob/","title":"Using Morning IOB","text":"One of the easiest habits to help check your settings is to simply check-in on your morning blood glucose and active insulin (IOB). Check if you are (1) above, below, or within target range and (2) carrying significant positive or negative IOB in the morning before you start walking around (if you are prone to dawn phenomenon) or eating. Since mornings are free from food influence, these are almost like waking up from a mini-basal testing every overnight.
This Looped group post started the conversation:
I've been having more lows recently than I would like. Any help here would be really really appreciated.
I took a screenshot of an example situation...LOOP is predicting that I'll eventually rise up to 124 and is therefore giving me my full regular basal rate EVEN THOUGH im currently way below my desired goal range. I have a better knowledge of my diabetes than LOOP does (sorry if that is heresy around here) and I would not like LOOP to be giving me basal right now OR ELSE I'LL GO LOW.
QUESTION: Which parameter must I change on my loop settings in order to essentially tell LOOP to relax on the basal rates??
To illustrate how to use morning IOB and blood glucose to check your basal rates, we will walk through a series of screenshots from a recent Loop experience. When the Looper woke up, she was below her 95 mg/dL target, but relatively steady. At about 8:48am, Nightscout showed negative IOB of -1.33 units, blood glucose of 90 mg/dL and slightly rising but still below a target of 95 mg/dL. (Note: this situation is similar to what the original Facebook poster above was describLooped2](ing.)
Looking back on the night, blood glucose was pretty much below target the whole night and her scheduled basal delivery was turning off/on in an alternating pattern (the blue pattern area). Blood glucose wasn't concerning nor were alarms going off. However, this combination of data is a great indicator that basal rates need to be decreased. The tendency could be to let Loop just keep plodding along like this, but it can lead to certain less than desirable stress points on the algorithm.
Firstly, if the user were to bolus while carrying a lot of negative iob, they would get more of a bolus than otherwise needed...yes, Loop will likely suspend again as soon the meal began because blood glucose spike would be well contained with the extra bolus, but the system would just have a harder time that it should with well tuned settings.
Secondly, if the user were to go above their correction range, they would begin to get high temporary basals for what is an inaccurate amount of negative iob (because basals really didn't need to be this high). And high temporary basals at this point in time and with these settings would be too aggressive. (If the maximum basal is set really high, the problem compounds with bad underlying settings. This is why it is a good idea to keep your maximum basal relatively low when you first start Loop and are testing your settings.)
Useful Summary
You probably need to lower your overnight basal rates if you wake up:
- Carrying negative IOB, and
- Blood glucose is still below or just below target, and
- Overnight basals were turning off/on.
So, what will you see after you adjust your basal lower? You will see some of the negative IOB disappear and, as a result, the predicted blood glucose curve will not so such an aggressive prediction for a blood glucose rebound.
In this example, the user's basal rates were lowered across the board. All of the basal rates went down about 0.2 U/hr. It works out to about a 15% decrease, and is pretty standard adjustment for this user due to her monthly hormone changes.
The adjustment to basal rates caused Loop to reevaluate its math. After adjustment, Loop now has a negative IOB of -0.55 units vs the previous value of -1.33 units. This is a more reasonable given the situation. The prediction line with a -0.55 units IOB was not predicting nearly as aggressive of a \"rebound\" blood glucose rise.
Why did IOB and prediction change?
The visualization below might be easier. The red line is how loop knew things to be before basal rates were adjusted lower. The purple line is how Loop viewed basal schedule after the adjustment lower.
The green arrows highlight parts of the graph that are recalculated by Loop when the basal schedule was adjusted lower. Instead of Loop thinking those were NEEDED basals (aka, conforming to the old basal schedule), now Loop perceives those as \"extra basal\" insulin deliveries. Now those insulin deliveries are sitting ABOVE my scheduled basal dotted line in Nightscout. They are actually instances of positive IOB and therefore Loop is now correctly getting closer to realizing that perhaps all of that extra wasn't needed.
The negative/positive IOB plus a quick glance at overnight Loop actions/blood glucose relative to targets is a quick easy check on overnight basals.
If one was really exacting, you could adjust basals until a number closer to 0 IOB. In practical use though, getting roughly closer is usually helpful enough and smaller adjustments could be made later if still needed. The graph below shows the results several hours after the basals were decreased. As you can see, looking pretty decent.
The confirmation that adjustments were on-track would also be provided by looking at morning blood glucose and IOB the following morning. As you can see below, the user was at target blood glucose and carrying a very small amount of IOB. Also, there are far fewer instances of basals alternating off/on. All good signs that the basal adjustments were reasonable.
Note: This entire scenario can also be applied in situations where (1) blood glucose has been steady but above target and (2) Loop is holding significant positive IOB, and (3) Loop has been high temping fairly consistently. By increasing the scheduled basals, Loop will recalculate that the positive IOB was actually necessary (not extra) insulin delivery, and the positive IOB will decrease. This will help the prediction curve to more accurately realize that not all the insulin deliveries were being utilized as \"corrections\" but rather as needed background basal.
"},{"location":"how-to/low-treat/","title":"Low Treatments","text":"Low blood glucose will still inevitably happen at some point, even with Loop use. A difficult carb count, illness, equipment malfunction, exercise...you name it, eventually something will help to cause a low blood glucose.
"},{"location":"how-to/low-treat/#how-to-treat","title":"How to Treat","text":"Overall, most people find that they can treat low blood glucose with fewer rescue carbohydrates than they previously used to. Since typically Loop will have been suspending your basal insulin before a low has been reached, you will typically have less IOB to overcome with carbohydrates. In other words, Loop takes some of the momentum out of an upcoming blood glucose dip...which also makes it easier to pull up from declining blood glucose.
If you notice that your old way of treating low blood glucose is leading to rebound blood glucose greater than you expect/desire, eventually consider decreasing the amount of carbohydrates you use to treat predicted/upcoming low blood glucose.
"},{"location":"how-to/low-treat/#logging-low-treatments","title":"Logging Low Treatments","text":"Another common question for new Loop users is whether to enter the carbohydrates used to treat a low blood sugar. The answer is that there is not a great agreement across all Loop users on how to deal with low treatment entries. The decision is dependent, often, on factors not just about the Loop algorithm; factors such as extra effort, use of OpenAPS's autotune (which needs all carbohydrate inputs to do a better estimation), or desire to have more thorough clinical records.
For the most part, you do not have to enter low treatments. Not logging low treatments will work out pretty well if you are not overtreating the low and you have your ISF pretty well determined. In the event you overtreat lows and your ISF value is too low, your Loop may be more likely to overtreat rebound situations.
"},{"location":"how-to/low-treat/#roller-coaster-after-low-treatment","title":"Roller Coaster after Low Treatment","text":"If you are roller coastering blood glucose after treating lows (going low, quick rise from the treatment carbs, then drop again from Loop's high temps during the rise, and then repeating pattern), here are some tips:
- While treating the low, try setting a temporary workout target of an hour to help keep Loop from aggressively treating a rebound.
- Consider lowering your maximum basal rate in Loop (within the Delivery Limits setting). This will limit Loop's ability to aggressively treat the rebound rise and allow you to work on getting your settings adjusted, if needed.
- Consider raising your ISF value (for example, if your ISF is set to 50, consider raising value to 55) to help Loop understand that a smaller magnitude of insulin adjustment is needed. In other words, tell Loop that insulin is packing a bigger punch than you'd previously thought.
If your ISF is set to a value that is too low compared to what it really should be, one of the most common symptoms you'll see is a roller coaster of blood glucose where the temporary basals are cycling between zero\u00a0and strong high temporary rates.\u00a0Here are some example graphs from Looped group.\u00a0These are examples where too low\u00a0of ISF is more than likely a large factor in the roller coaster\u00a0(doesn't mean it is the only culprit, and is more difficult to ferret out when food is involved like the second graph). \u00a0But, lightning bolt high temporary basals followed by very quick blood glucose drops and zero temps is usually too low of ISF value...raise the ISF value (e.g., go from 50 to 55) to help Loop know that each unit of insulin is actually having more impact than you'd previously thought.
"},{"location":"how-to/overrides/","title":"Override Targets","text":"Your Loop has two override targets that you can set; Pre-meal and Workout. When activated, these targets will \"override\" or take the place of your usual correction range and therefore will affect Loop's decisions about insulin delivery.
You can set the value for these override targets from the Correction Range area of Loop settings. Generally speaking, pre-meal targets are set to value(s) lower than your correction range and workout targets are set to value(s) greater than your correction range. Loop will indicate that an override target is active by showing a darker blue bar on the blood glucose chart showing the duration and loation of the target.
"},{"location":"how-to/overrides/#pre-meal-targets","title":"Pre-Meal Targets","text":"The pre-meal target, when activated by pressing on the icon, will stay active for:
- one hour, or
- until carbs are entered, or
- until pre-meal target is manually cancelled by pressing the icon again.
Pre-meal target will automatically end when one of those three things happens, whichever comes first will end the pre-meal target.
The original intent of the pre-meal target was to help give a gentle, easy pre-bolus effect to help lessen a post-meal blood glucose spike. By lowering your Loop's target going into a meal, you can get a \"pre-bolus\" effect because Loop will aim for that lower target and thus tend to set a little more aggressive insulin delivery while the target is active. Loop will recognize the \"extra\" insulin that has been delivered while pre-meal is active, and Loop will smartly take that into account when it is finally time to recommend a bolus at meal entry time.
Pre-meal target won't replace the utility of a true pre-bolus in most situations, but it can be useful in many situations beyond simply helping meals. This lower target can help Loop be more aggressive:
- During stressful events where adrenaline might be causing increased blood glucose
- During exercise that may tend to increase blood glucose
- During illness that is causing high blood glucose
The workout target, when activated, will give a duration choice of 1 hour, 2 hours, or indefinitely. The workout target, set greater than your usual correction range, is useful to help in situations where Loop needs to be less aggressive with insulin delivery. The most common use of the workout target is to set it in advance and during exercise to help minimize IOB going into a workout. While the name is a hint to the most common timing of this override's use, there are other times where it can be helpful such as:
- Helping to recover from low blood glucose treatment
- Helping after workouts when still experiencing increased insulin sensitivity
- Helping during illness which is tending to lower blood glucose or increase insulin sensitivity
You can read more about the timing and use of workut targets before/during/after exercise in the Exercise page of these tips.
"},{"location":"how-to/site-fail/","title":"Infusion Site Failures","text":"Some of the most frustrating and dangerous times are the result of insulin infusion site failures. When using Loop, site failures most likely will lead to your active insulin to be incorrect. Insulin that the pump thought it had delivered to your body is instead blocked by a kinked cannula or leaking out on your skin surface. Loop will have an incorrectly higher active insulin value since it thought that insulin had been delivered, but in reality it was not. Therefore your Loop will be less effective in closed loop mode to bring down any resulting high blood glucose from the site failure.
To deal with insulin site failures, there is no way to 100% recover from the issue immediately. Generally, we have no idea how much insulin has failed to deliver nor when the failure happened, so simply replacing that exact amount of insulin will not be possible. There are steps to help transition back to successful closed looping.
Take care to not rage bolus in this situation and give the new site a chance to start absorbing insulin. In cases where basal insulin delivery has failed, it takes a full two hours for resumed basals to come up to full effect for blood glucose impacts. There will be a lot of patience and attention during the 4-6 hours after a failed site is replaced, unfortunately.
"},{"location":"how-to/site-fail/#open-loop","title":"Open Loop","text":"Open loop mode is a first line of defense during this situation. By open looping for a duration of 6 hours, you will give your Loop an opportunity to have accurate active insulin (i.e., IOB) accounted for again since that amount of time is the length of your insulin activity duration. If you are very impatient, waiting out at least 4 of the 6 hours will get most of your active insulin accounting back into range. As you can see from the figure below, at 4 hours there is only less than about 10% of the insulin tail remaining, so the bulk of active insulin will be correctly accounted for by then.
"},{"location":"how-to/site-fail/#prime-insulin","title":"Prime Insulin","text":"If you are more certain about how much insulin is missing from your Loop's tracking of active insulin, you can choose to deliver insulin using the \"prime\" delivery function in your Medtronic pump while having your tubing connected to the new, working infusion site. Assuming you are using a Preferred Data Source set to Event History (most people are since this is the default setting), then prime insulin is not counted towards active insulin. In this way, you can deliver insulin to make up for some of the insulin that Loop had thought had been delivered (but wasn't).
"},{"location":"how-to/teens-kids/","title":"Test Your Settings","text":"Open Looping Fine-Tuning Settings Book settings
"},{"location":"how-to/think-like-loop/","title":"Think Like a Loop","text":"\u201cWhy isn\u2019t Loop giving me more?\u201d That is a common question from new Loopers.
\"Why is Loop suspending right now?\" That is another common question.
For these questions, you'll go really far in Looping if you can remember these two things:
- Go to therapy with your Loop
- Everything depends on the predicted BG curve
I'll challenge you to \"pretend you're at the marriage therapist\" as you learn how to use Loop as a partner.
Why are you in disagreement with your partner? What would a therapist tell you? Usually, therapy is because of miscommunications with each other. You aren't on the same expectations or priorities. The same applies to Loop. For example, you'd be at the therapist's chair telling her \"Loop just isn't treating my high BG as fast as I want. This is so frustrating. I correct and Loop just suspends.\" That right there is the definition of miscommunications and different priorities. Why?
\u27a1\ufe0f Loop thinks you have enough insulin to be at target later.
\u27a1\ufe0f You think you need more right now.
So...look at the objectives and info you both have at hand.
Loop has the info from the settings you\u2019ve given it, how you\u2019ve described your meals, and how you\u2019ve given insulin. Loop does not know if you are stressed. Loop does not know if you are sick. Loop is not a rage bolus machine and is predicting 6 hours out to safely bring you down without you needing to prevent a low while it does that. Loop is considerate partner. It respects your input and acts gentle and even-keeled.
If you are frustrated with Loop\u2019s actions now...that means you told it the wrong info previously (or are lacking some patience that Loop has). Most often either your meal entry was wrong or your basals or...something else you\u2019ve told Loop has ended up being incorrect.
Fixing that miscommunication will allow Loop to do a better job.
If you misjudged a meal, go back and edit that meal entry.
If you misjudged your basal/ISF/carb ratio needs, correct those (testing to validate is good if you haven\u2019t done that yet)
If you aren\u2019t patient, you can override Loop...but do expect that is not likely to solve the miscommunication you and Loop are having. That\u2019s the relationship equivalent to just talking louder as opposed to talking smarter.
Loop will try to save you when you rage bolus, which may just make you even more frustrated. If you can stop and think \"What have I told Loop that might not be accurate right now?\" that will be a good step to a meaningful improvement in outcomes.
"},{"location":"how-to/think-like-loop/#tools-when-you-are-in-disagreement","title":"Tools when you are in disagreement","text":"Sure, even with the best therapist, you may still have disagreements. They happen. There are several options for recovering from times like that. The tool you choose may be situationally dependent. Seeing a huge quick BG climb after a meal may be better helped by a carb entry edit as opposed to added patience, but that will depend on your management style.
- Add some patience to your practice. A little patience to give the already added insulin time to work can help with Loop.
- Edit your meal's carb entry. If you find your meal going crazy different than expected (did they forget you said \"diet\" coke?), you can always go back and edit your original carb guess to be more accurate based on what you're seeing. You can decrease the carbs or increase the carbs. You can change the absorption time. These edits can be done by tapping on the carb chart on Loop's main screen.
- Use overrides to tell Loop about overall insulin needs changing. Hormones...oh lordy the hormones are horrible. That's a good time to use an override to let Loop know that overall you need more than usual. Or if you're about to hike the Himalayas, use an override to let Loop know you will need far less insulin than an average day.
- Open Loop. This is a great tool. If you just had things go really wrong for a bit (bad cannula kink anyone?) then you can open the loop until things settle down and you get squared away again. Give your manual injection, Afreeza, or whatever your desire is, and wait until you've had 4-6 hours of regular basals going before closing the loop again.
- Give pending insulin. If you are really feeling the itch to rage bolus, you can meet Loop at the halfway mark. If you click on the bolus tool while Loop is high temping you, you will see a \"pending insulin\" value. That is the amount of insulin Loop is planning on giving you through temp basals over the next 30 minutes. If you give that pending insulin total as a bolus now, you will get the entire amount working faster AND Loop agrees you are safe to use it. In this way, Loop won't automatically suspend insulin when you give pending insulin...because your predicted BG curve had said it was needed anyways. Win-win...you got a mini-rage bolus and Loop won't be working against you. Good therapy and a good way to avoid post-rage lows.
As you start to use Loop you will probably find yourself wondering at some point \"Why is it suspending insulin right now?\" or \"Why isn't it giving me increased basals right now?\" You might find yourself reverting to a bunch of old-school habits like looking at your carbs on board or insulin on board and trying to calculate a bunch of numbers. Stop there...that's so 1990s. You're a looper now and there is actually an easier way to assess things.
Thinking like a Loop involves just three things
\u27a1\ufe0f Predicted BG curve \u27a1\ufe0f Correction Range \u27a1\ufe0f Suspend Threshold
All of Loop's decisions are based on what your predicted BG curve is doing with respect to your correction range and suspend threshold. That's it. All the time. Everytime. It always comes down to that predicted BG curve.
That predicted BG curve has four contributions to its shape:
- Carbs (the ones you've told it about)
- Insulin (the ones you've given through your pump or recorded in the Health app)
- Blood glucose momentum (how fast your BG has been changing recently)
- Retrospective correction (how accurate/inaccurate Loop has been recently in predicting your BGs)
Loop models the inputs from those four contributions to form the final shape of the predicted BG curve. Once that curve is generated, Loop looks at where the entire curve will be for the next 6 hours relative to your suspend threshold and correction range, and take one of four actions:
Action 1: Set a 0 u/hr basal rate (aka suspend basals)
Loop will do this anytime your predicted BG curve has any portion that goes below your suspend threshold.
And the remaining three actions all assume NONE of your predicted BG curve is below suspend threshold, otherwise you'd be in Action 1.
Action 2: Set scheduled basal rate
Loop will set your scheduled basal rate when:
- All the predicted BG curve is within the correction range, including the very last value which is called your \"eventual BG\", or when
- If the eventual BG is above your correction range in 6 hours, but there's a dip below the correction range earlier in the curve.
Action 3: Increased basal rate
If the eventual BG and all the predicted BG curve is above your correction range, Loop will give you a high (increased) temp basal.
Action 4: Decreased basal rate
If your eventual BG in 6 hours is below the correction range, you'll get a decreased basal.
"},{"location":"how-to/think-like-loop/#what-would-loop-do","title":"What would Loop do?","text":"Test your new found skills...let's play a game where you try to guess the action...
What would Loop do?
It\u2019s 8pm and your predicted BG curve looks like the following. What do you expect Loop to recommend/enact at 8pm? A. Zero (suspend) temp basal B. Scheduled basal from your settings C. High temp basal D. Lower temp basal (between zero and scheduled)
What would your answer be?\u00a0\u00a0Before you give your final answer...consider this next twist.\u00a0 Would you give the same answer to this graph as you gave the graph above?\u00a0 If not, what would the answer be for this graph and why? (ignore the timestamp mismatch
This second case has a dramatic drop happening.\u00a0 Suspend threshold is still at 60 mg/dL, the correction range is still 90-110, the lowest value on the predicted BG chart is 75 mg/dL, and the eventual BG is 171 mg/dL for this example.\u00a0 So, pretty similar to the first example except for this precipitous drop going on right now.
Let me tell you some of the common pitfalls we all can easily slip into when trying to answer these \"Why is Loop giving me this basal?\" questions.
Wondering about IOB, COB, or DIA in order to answer...those aren't a factor in answering the question as they have already been used to make the predicted BG curve.\u00a0 In other words, they are accounted for already in the information presented.\u00a0 All you need to answer this question is provided by the predicted BG curve, your suspend threshold, and your correction range. Thinking about this as a human...humans tend to say \"Well, I'm on a rise/fall right now so...[insert Loop action based on that]\"\u00a0Loop isn't looking at the past BG movement alone, instead, it's looking at the\u00a0prediction curve ahead and applying its rules based on that.\u00a0 Any drop or rise going on will have been added to the predicted curve through the BG momentum and retrospective correction components of the algorithm...so again they're already incorporated into the predicted curve.
Restating for emphasis:\u00a0 All you need to answer this question is provided by the predicted BG curve, your suspend threshold, and your correction range.
The answer is B:\u00a0 Loop will give your scheduled basal in both situations shown above.\u00a0 When your predicted BG curve (1) drops for a time below the correction range but (2) all of the curve is still above suspend threshold, and Eventual BG is (3) above range or within range...Loop will give your scheduled basal.
The logic is a bit of a wait-and-see.\u00a0 Scheduled basal will maintain the delivery of insulin.\u00a0 Your settings haven't told it this is an \"oh my gosh...stop the insulin!\" moment (you're predicted to still stay above suspend threshold), but we also don't want to give high temps yet (to correct the eventual BG) because we'd like to safely make it through the part that is below correction range coming up.
If BGs were to drop (enough) or keep dropping (enough), your predicted BG curve would likely slip to your suspend threshold and then Loop would suspend.\u00a0(One important take away is to not set your suspend threshold so low that it no longer acts as a safety in these situations.) If BGs were to rise enough such that the whole predicted curve comes back into or above the correction range completely, you'd then get high temp basals to correct for that eventual BG that is above the correction range.
"},{"location":"settings/adjust/","title":"Adjusting Your Settings","text":"There will be times when you need to adjust your settings in order to keep your blood glucose within your correction range. The degree to which you adjust your settings will depend on a variety of personal decisions about how active you engage with your diabetes, how comfortable your endocrinologist may feel with patient-adjusted settings and the underlying reason for the change.
"},{"location":"settings/adjust/#troubleshooting-before-a-change","title":"Troubleshooting before a change","text":"The most common reason that people decide it may be time to adjust their settings is when previously great blood glucose control suddenly becomes elusive. What worked yesterday is no longer working today. Because of course, as soon as you test and dial-in all those settings, diabetes will throw you a curve ball and change your insulin needs. That's the way it works. It's not just YDMV (your diabetes may vary), it's actually YDWV (your diabetes will vary).
Before considering a change to your settings, the usual laundry list of possible other causes should be explored.
- Bad infusion site (kinked cannula, blood in cannula, infected site?)
- Bad insulin (cloudy? insulin exposed to heat/freezing?)
- Unexpected food impacts (that diet coke wasn't diet, kid sneaking food?)
- CGM damaged (poor calibration, damaged sensor site?)
Some signs that the issue is actually the result of equipment failure include the failure of blood glucose to decrease despite repeated bolus corrections. Often during equipment failure troubleshooting, you've blamed the food (and adjusted a carbohydrate count), given corrections, checked for hidden food wrappers, and maybe even open-looped rage bolused. Breathe deep, it is super frustrating to have equipment failure. One of the benefits of Loop can be that equipment failures can become a little easier to identify over time because blood glucose control becomes more predictable and reliable.
"},{"location":"settings/adjust/#assess-the-duration","title":"Assess the duration","text":"Once you have eliminated equipment failures as a source of unexpected blood glucose control problems, it is time to determine if this is a short-term or long-term issue. This assessment tends to be a skill that people will get better with practice.
Before considering a change to your settings, you should assess if you expect the issue to resolve on its own soon or if the issue might be caused by a more long(er) term issue that should be addressed.
"},{"location":"settings/adjust/#short-term-factors","title":"Short-term factors","text":"There's a long list of short-term factors that may temporarily affect how well Loop can automatically keep your blood glucose at your correction goal. Of course, the definition of \"temporarily\" is also subjective. Some people grow impatient with blood glucose excursions within hours and other people don't consider the need to correct excursions for days. That difference in tolerance will affect a user's assessment of how quickly they will consider adjusting settings. For these discussions, short-term is referring to less than a day.
For the most part, it is not a good idea to adjust settings for very short-term underlying causes such as:
- One or two meals that behave differently than expected
- Stress from a final exam
- Anticipation of holiday party buffet table
- Sprinting in the airport to catch a plane
- \"Baggage claim\" lows from pressure changes causing unexpected insulin delivery
- Weather
Those are very short term factors and if you change your underlying settings too quickly from just one or two observed blood glucose excursions, you'll likely end up needing to undo the changes just as quickly. For these short-term factors, the best tools involve using temporary target overrides and a measure of patience. In some medium-duration cases (factors that last for at least several hours), people also use \"fake\" carbohydrate entries to mimic the extra insulin needed during temporary stress on the system which causes high blood glucose.
"},{"location":"settings/adjust/#long-term-factors","title":"Long-term factors","text":"The list of long-term factors is just as long, but the most common are probably:
- Growth hormones
- Menstrual cycle hormones
- Illness
- Medications
- Travel
- Beginning new consistent exercise routine (or suddenly becoming more sedentary)
For long(er) term factors, you may benefit from adjusting your underlying settings to reflect the insulin-need changes that have happened as a result of the factors. By adjusting your underlying settings, you will help Loop to better predict your insulin needs and therefore you'll be more likely to achieve blood glucose in correction range.
"},{"location":"settings/adjust/#assess-the-data","title":"Assess the data","text":"Once you have assessed that there's a need for a settings adjustment, the next step is to decide which setting(s) to change and how to adjust them.
Whether you undertake the adjustments by yourself and/or with your health care provider's input is at your own discretion. Ideally, you would have your health care provider's input and support to discuss the observations and concerns you have and develop an adjusted setting profile based on those discussions. Many endocrinologists are still becoming familiar with Loop and its algorithm, so providing them links to the various sites would be a good starting point for discussion. This can help the endocrinologist understand where Loop is similar/different than traditional pump therapy. If you and your health care provider are comfortable with your self-adjustments and Loop's algorithm, then proceed ahead.
To begin the process of settings adjustment, visualizing your data trends is important. You need to be able to piece together what was eaten, how it was bolused, what temporary basals were applied, and how your blood glucose was reacting. The Data section of LoopTips covers the various places you can find your diabetes-related data to help you during adjustments.
"},{"location":"settings/overview/","title":"Overview","text":"In the context of the following discussion, the term \"settings\" refers to information that the user has typically entered into their pump. These settings include:
- basal rate schedule
- insulin sensitivity factor (ISF)
- carbohydrate to insulin ratio (CR)
- glucose target range
- duration of insulin activity (DIA)
Typically, most people with type 1 diabetes consult with their endocrinologist to develop and revise their settings. The endocrinologist usually reviews the patient's blood glucose and treatment history at each appointment to assess whether settings need to be adjusted. There are several great books available about the topic (Think Like A Pancreas and similar)
"},{"location":"settings/overview/#why-settings-matter","title":"Why settings matter","text":"Loop is primarily a set of math equations called an algorithm. The recommendations and actions that Loop takes are based on that algorithm, and the settings you enter in your Loop app are the basis for that math.
Let's start by thinking about basal rates. A well-adjusted basal schedule is designed to keep your blood glucose steady throughout the day, everything else being equal. If you were to not eat, not exercise, and basically keep a mellow lifestyle...basals should keep your blood glucose steady.
That is how Loop's math starts, and it's an important point to remember as you use and learn your Loop app. Loop's math is based on the assumption that the basal schedule you have provided in your settings are capable of keeping your blood glucose steady in the absence of other stressors. So as your blood glucose goes higher than your correction range for an unusual short-term influence like stress or unannounced carbs, you've been accoustomed to delivering a \"correction bolus\" to get back to range. Or if blood glucose goes below your correction range, you may need to eat recovery carbohydrates.
With all the excitement about automated insulin delivery, some people mistakenly assume that the user's settings don't matter anymore....that everything is automated. However, settings do still matter as they provide the basis for Loop's math. Diabetes is not a static math equation. Loop does not adjust your settings for you, that responsibility still falls to the Loop user when needed.
Let's give some examples to help. When you calculate a meal bolus, your math probably looks similar to this:
But many times you are probably adjusting that math based on any number of \"factors\" that you are aware could be affecting your blood glucose. Diatribe put together this great list of factors that affect blood glucose, and therefore could also affect how you'd adjust your insulin delivery from what they otherwise would normally be.
Some of those factors you are already used to making allowances for, such as:
- Pizza boluses...You probably are accustomed to needing an extended bolus and/or higher temporary basal rate to help with the late blood glucose spike common after the initial bolus starts to wear off. That pizza takes a long time to digest thanks to those slow-acting fats.
- Medications...How about that time that you had to take steroids for an achy joint? Those steroids probably resulted in a dramatic overall increase in insulin needs for several days, and you had to set aggressive temporary basal increases or other settings adjustments.
- Hormones...monthly hormone cycles have many women adjusting their basal rates and/or insulin sensitivity schedules for a week or so each month.
- Exercise...always a tough one because some exercise may make your blood glucose rise (lifting weights) and other exercise may make your blood glucose drop (running). Through repetition, most people have learned how to adjust for exercise by using temporary basal rates or timing of meals/boluses.
The degree to which Loop can deal with those stresses/factors is largely dependent on the duration of the \"factor\" itself and how well we can communicate to Loop that those factors are at play. For things we know are coming, we can tell Loop in advance, and Loop has ways of dealing with them mathematically. In the pizza example, Loop lets us mark the meal as long, slow digestion, and therefore better react to a potential late blood glucose rise. For exercise, we can use temporary override blood glucose targets before/after exercise to help Loop administer less insulin for awhile.
If, however, your insulin needs are impacted by something more long term such as hormone cycles, medications, or illness then sometimes your Loop may not be able to consistently hold you at your desired blood glucose. You may end up holding steady at a higher/lower blood glucose than you'd prefer. This is because Loop's math is still using the inputs that were from before you got sick/stressed/hormone'd (new word). Then the decision becomes \"Do you adjust your settings now or just wait it out until the stressor goes away?\"
"},{"location":"settings/overview/#when-do-you-need-to-change-settings","title":"When do you need to change settings?","text":"If the inputs to those math equations (i.e., your settings) are not accurate, your results for blood glucose control may also not be optimal.
Your basic insulin settings are what keeps your blood glucose steady without stressors present. For much of the time, the stressors in type 1 are short-lived and temporary (we just ride them out)...carbohydrate count is a little off, math test in algebra class, public speaking engagement, etc. Loop does well helping with the short-term stressors. They don't represent a prolonged need to change settings and the blood glucose impacts are short-lived as Loop corrects for them.
However, the longer term stressors are often helped by settings adjustments. Examples of longer term stressors are medications, illness, hormones, travel, or altitude. By letting Loop know that the status quo has changed (more/less insulin is needed overall) will help BGs get and stay in range. If you have told Loop that your insulin need is 1 unit/hour, but in reality, you're needing 1.3 units/hour because you've traveled to a colder place and will be sedentary for days...you will likely find yourself steady but higher than your target range while Looping. The high temporary basals that Loop will be providing in an attempt to bring you back to target will simply be going to fill the lack of adequate basal scheduled. If you update your basal schedule to reflect 1.3 units/hour of basal needs, Loop would be better positioned to provide needed insulin delivery to keep you in the target range.
"},{"location":"settings/overview/#automated-settings-adjustments","title":"Automated settings adjustments","text":"A common sentiment among new Loop users goes something like \"Why doesn't Loop see that I'm stuck on a high blood glucose? Why doesn't it bring me back to my target blood glucose?\"
What the user is actually implying is that the user thinks Loop should be doing some sort of retrospective analysis on the user's blood glucose and data to make a \"better\" decision. The user wants Loop to learn from recent data and adjust.
There is a short-term retrospective analysis built into Loop which will apply a weighted-correction based on the past 60 minutes of blood glucose changes. While this does help some, larger-scale \"learning\" is not currently a part of Loop's algorithm.
Perhaps in subsequent versions of Loop, auto-adjustment of settings or machine learning could be incorporated. Until then, you will need to tell Loop if your underlying settings need updating or make temporary adjustments for short term issues. These pages in LoopTips are written to help you identify when settings may need adjustment or when other techniques may be useful in given situations.
"},{"location":"settings/settings/","title":"Initial Settings","text":"Here's how 90% of Loopers start:
- Switch from their existing pump over to a Loop-compatible pump
- Within hours, they begin closed loop use of Loop
- When blood glucose isn't as smooth as hoped for, Looper begins to try to adjust settings while still in closed loop
The problem with this sequence is that adjusting settings on the fly while in closed loop can be difficult as a new user. As you become more experienced there are some great tips for adjusting settings while in closed loop, like Loop and Learn: Basal Testing in Closed Loop, but as a new Looper, you should start with the basics. Here are some recommendations to help make the transition to closed loop easier. The first place to start is with getting the best initial settings as you can.
"},{"location":"settings/settings/#try-open-looping-first","title":"Try Open Looping first","text":"If I could offer one bit of really useful advice, it would be to open loop for at least a few days. Adjust to the user interface and buttons you'll be pressing. This breaking-in phase will also let you get used to a new pump if you've never used Omnipod before or if you are new to using a Medtronic pump. If there are differences between how your body absorbs insulin from your old pump vs the new pump, open looping will be a good way to notice those before adding the complexity of automated corrections from Loop.
Open Loop is like Manual Pump
While open looping and entering meals with the pizza icon, Loop will recommend less upfront bolus with the intent to make up the remainder of needed insulin via automated dosing later. In those situations, you may have to manually dose the split portion of the bolus later, similar to normal pump use for slow digesting meals like pizza and quesadillas, since Loop will not be automatically dosing.
If you want extra help, you can read some excellent books about pump settings and what they mean. For example, \"Think Like a Pancreas\" by Gary Scheiner is a great reference for understanding some of the guiding principles in pump therapy. \u00a0Let me summarize the important parts about pump settings:
- Basal rates should keep your blood glucose steady in the absence of other influences (such as food, medications, etc).
- Boluses should return your blood glucose to target after a meal.
- Insulin Sensitivity Factor (ISF) is the amount one unit of insulin drops your blood glucose without other influences.
- Carb Ratio (CR) is the meal entry in grams that requires one unit of insulin to cover
Basal, ISF and CR
It's the way it was defined in the beginning, but it can be confusing.
If you need more insulin
- Basal rates should increase (higher number)
- ISF and CR should be stronger (lower number)
- They are in the denominator of the fraction
It is absolutely possible to have two wrong settings look like right settings when they balance out. The problem is that those wrong settings won't balance out in all situations...and Looping will not be smooth in those situations. Remember - in closed loop mode, Loop will adjust with every new glucose reading - typically every 5 minutes. What worked on a manual pump when adjusting infrequently, might not be suitable for Loop.
Example:
- Too low a value for basal rates can be offset by regular eating of meals with too strong (lower value) carb ratio
- If you stop eating though, you'll start going high because that extra insulin from the meal boluses won't be there to offset the lack in basal rates
Taking the time to validate your settings by truly testing them is really good practice before closing loop.
"},{"location":"settings/settings/#insulin-duration","title":"Insulin Duration","text":"The majority of pumpers used DIA at about 3 hours on traditional pump therapy.\u00a0There's a reason for that. \u00a0In traditional pumps, a simple linear model is used to estimate the decay in effectiveness of the insulin already delivered. And it is true that after 3 or 4 hours, insulin is not affecting glucose as much as initially. But the true biological lifetime of insulin has a long tail. Insulin from more than 5 hours ago is still making a small decrease in your glucose. Loop takes that into account using a realistic model for insulin effectiveness.
In your traditional pump, the DIA, using a linear model, was used to give a correction because blood glucose is stuck high or going low...in other words, DIA was used as a rough approximation to correct off-target blood glucose. It doesn't have to be rocket science then since we're making an approximation because some other numbers (carbohydrate count, basal rates, ISF, etc) weren't behaving the way we were expecting either and therefore leading to an off-target blood glucose.
With the release of Loop 3, all insulin types use a realistic model, so you will not be selecting a DIA. These models help prevent insulin stacking as you begin Loop. You can read more about insulin models and how they help determine the active insulin over time in the LoopDocs: Algorithm pages.
With Loop 3, you will select the Insulin Type you use, such as Novalog, Humalog, Apidra, Fiasp or Lyumjev. Each is associated with the appropriate exponential model for onset and peak action times.
"},{"location":"settings/settings/#basal-rates","title":"Basal Rates","text":"Basal rates are the keystone of your therapy settings. Establishing a quality basal schedule is setting yourself up for successful looping. Personally, we find testing basal rates can be pretty painless and doesn't require days of fasting. Instead, look for easy opportunities.\u00a0If you are willing to open-loop test, that is going to give the most accurate information in the quickest way.
It's a pretty easy test. Turn off your loop or open loop. Don't eat food, don't do crazy exercise, don't sit in a hot tub. Just have a relaxing average time period and see if your basal rates are able to hold you roughly steady. It doesn't matter if you are at target or not; the idea is to simply have zero extra active insulin from any boluses or corrections and watch what happens during those hours. Typically you would want to see about 3-4 hours of blood glucose without the influence of food boluses or activity. You can even try this tip: Loop and Learn: Basal Testing in Closed Loop
Many people, especially little kids, are not enthusiastic about fasting basal testing...so look for opportunities to make it less cumbersome if you have to eat during the test. For example, use a low carb meal/snack that you know how to bolus for (or doesn't need a bolus at all) and generally needs no corrections.\u00a0For us, that's two extra large scrambled eggs (or three small ones) with cheese\u00a0bolused at 8g.\u00a0If she eats that meal, the blood glucose response is slow and measured. Three hours after that meal, the bolus and food effects are really muted and we can start watching to see if blood glucose stayed pretty steady. For example, here's some of a recent open-loop basal testing we did and confirmed that blood glucose could stay pretty steady without the involvement of temporary basal help from looping.
"},{"location":"settings/settings/#insulin-sensitivity-factor","title":"Insulin Sensitivity Factor","text":"Insulin sensitivity factor (ISF) is the next logical setting to test.\u00a0If you've just done the basal test and gotten steady blood glucose with an open loop, try taking a glucose tab or two. Wait for your blood glucose to be steady at the higher value, and give a safe correction that you think will get you close to target. Watch the resulting blood glucose drop over the next 3-4 hours. You should see blood glucose come to a steady level again. How much did the blood glucose drop? How many units of insulin did you use? Divide the two numbers and you will have your ISF. \u00a0If your BG dropped 15 mg/dL\u00a0with half unit of insulin, your ISF is approximately 30 mg/dL per unit.
Testing ISF is an important step in successful closed Loop use. ISF is used in every Loop calculation and plays an important part in Loop's prediction curve. Most users find that they have to adjust ISF; most often increasing the value. If you are experiencing roller coastering blood glucose, as described below, that is often a sign that ISF needs to be weaker (higher value).
Before you read the next part, let's clear up some definitions and terms:
Insulin sensitivity factor described just that...how sensitive your blood glucose is to the effect of insulin. If your blood glucose doesn't drop much when you put in insulin and you have to use a lot, your insulin sensitivity is \"weak\". If your blood glucose drops quickly and you don't need much insulin to correct a high blood glucose, your insulin sensitivity is \"strong\".
The terminology gets confusing sometimes because some people think of the terms \"weak\" and \"strong\" in terms of how Loop's actions will behave. If you make your ISF value greater (i.e., changing it from 50 to 55), you are weakening the magnitude of Loop's adjustments because you are telling Loop that your sensitivity to insulin is stronger...you don't need as much to correct a high blood glucose and suspensions will have a tendency to rebound more.
In order to avoid the confusion that surrounds the words \"weak\" and \"strong\", instead we will describe the actual adjustment to the ISF value...the number itself.
If your ISF is set to a value that is too low compared to what it really should be, one of the most common symptoms you'll see is a roller coaster of blood glucose where the insulin adjustments are cycling between zero\u00a0basal and strong high temporary rates (or automatic boluses).\u00a0Here are some example graphs from Looped group.\u00a0These are examples where too low\u00a0of ISF is more than likely a large factor in the roller coaster\u00a0(doesn't mean it is the only culprit, and is more difficult to ferret out when food is involved like the second graph). Lightning bolt high temporary basals followed by very quick blood glucose drops and zero temps is usually too low of ISF value...raise the ISF value (e.g., go from 50 to 55) to help Loop know that each unit of insulin is actually having more impact than you'd previously thought. Making changes of about 10% for a setting is reasonable to try; just make sure you are changing it in the correct direction - remember: a lower number = more insulin; a higher number = less insulin.
"},{"location":"settings/settings/#carb-ratios","title":"Carb Ratios","text":"Now that you have basals and ISF set, here's where it gets really tempting to close loop and move on. And, truthfully, it's not that hard to test carb ratios on a closed loop vs an open loop if you've solidly tested all these other factors.
A good carb ratio will bring your blood glucose, for an average meal, back to the starting point of the meal within about 3 hours or so.
A bad carb ratio will leave you higher or lower than the starting point of the meal.
For example, this graph shows too aggressive of a carb ratio (value too low). \u00a0Three hours after the meal, there's nearly 0.50 units of active insulin, blood glucose is well below where the meal started, and definitely low treatments needed.
If you are finding that a correct carb ratio is yielding good blood glucose 3 hours later, but you aren't happy with the peak blood glucose during the meal, then it may be time to explore increasing or adding prebolusing time to your meal or implementing \"eating soon\" targets an hour before meals to help control the post-meal blood glucose spike.\u00a0Artificially strengthening carb ratios to help control post-meal blood glucose spike will likely yield lows 2-3 hours after a meal.
"}]} \ No newline at end of file diff --git a/sitemap.xml.gz b/sitemap.xml.gz index 1537000..9175ac3 100644 Binary files a/sitemap.xml.gz and b/sitemap.xml.gz differ