PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
June 22, 2018Diabetes

Predicting Exercise-Related Changes in Glucose in People with Type 1 Diabetes Using Linear Models and Incorporating Knowledge of Prior Exercise

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Does including past exercise history improve the prediction accuracy of glucose drop during exercise in adults with Type 1 Diabetes?

Population

20 adults with Type 1 Diabetes, mean age 35.2 years, 14 females, participating in a 4-arm, 4-day outpatient…

Comparison

Linear predictive model including past exercise… vs Linear predictive model excluding past exercise…

Design

Other

Follow-up

4 days

Authors

NTNichole S. TylerRRRavi ReddyJYJoseph El Youssef

Discussion

Loading...

Member takes

Overview

May aid real-time hypoglycemia prediction during exercise in T1D; leaves open prospective validation before clinical adoption.

Structured PICO

Does including past exercise history improve the prediction accuracy of glucose drop during exercise in adults with Type 1 Diabetes?

P
Population
20 adults with Type 1 Diabetes, mean age 35.2 years, 14 females, participating in a 4-arm, 4-day outpatient artificial pancreas study.
I
Intervention
Linear predictive model including past exercise history (historical model) to estimate glucose drop during exercise.
C
Comparator
Linear predictive model excluding past exercise history (naïve model).
O
Outcome
Prediction accuracy of glucose drop during exercise (measured by root-mean squared error [RMSE] and R2).surrogate

Linear models can accurately predict exercise-related glucose drops in Type 1 Diabetes, with modest improvements when incorporating prior exercise data.

Cite This Study

Tyler et al. (2018) studied this question.

synapsesocial.com/papers/6a70ac80b27f15817827786dhttps://doi.org/10.2337/db18-751-p
View Full Paper
Ask AI
Bookmark
Share