Background and Aims: Exercise uptake is low in people with type 1 diabetes (T1D), with hypoglycaemia fear cited as the main barrier. We establish factors that predict hypoglycaemia after exercise.

Materials and Methods: People with T1D exercising regularly (minimum 150min/week) recorded the type, timing and duration of exercise, and heart rate, for 80 days using a smartwatch. Participants used Dexcom G6 Continuous Glucose Monitoring throughout. Predictors of hypoglycaemia were analysed.

Results: 24 (8 men), mean (SD) age 35 (11) years, duration of diabetes 14 (7) years, BMI 26 (4.7) kg/m2 and HbA1c of 58 (12) mmol/mol recorded 1454 activities with complete CGM. During the 24 hours after exercise, increasing age and baseline HbA1c increased the risk of experiencing hypoglycaemia <70mg/dL (Age OR 1.028 (95% CI 1.019 - 1.038); HbA1c OR 1.027 (95% CI 1.016 - 1.037)), with similar results at other thresholds. Every additional minute of activity increased exposure to hypoglycaemia <70mg/dL (OR 1.007 (95% CI 1.004-1.010)), 54mg/dL (OR 1.011 (95% CI 1.008-1.014)) and 50mg/dL (OR 1.010 (1.007-1.014)). Exercising in the afternoon reduced the risk of hypoglycaemia at all thresholds (70mg/dL OR 0.766 (95% CI 0.602-0.975), 54mg/dL OR 0.575 (95% CI 0.403-0.820) 50mg/dL OR 0.503 (95% CI 0.330-0.766)) compared to the morning. Females had a reduced risk of hypoglycaemia <70mg/dL compared to males (OR 0.696 (95% CI 0.556-0.871)). Running (OR 2.898 (95% CI 1.958-4.290) and cycling (OR 2.989 (95%CI 2.154 - 4.148)) increased exposure to hypoglycaemia <70mg/dL compared to walking. In multivariate analysis, age (p=0.002), baseline HbA1c (p=0.004), sex (p<0.001), activity type (p<0.001), duration (p=0.009) and timing of exercise (p=0.004) were significant predictors of hypoglycaemia in the 24 hours after exercise.

Conclusion: To empower adults with T1D to avoid hypoglycaemia after exercise, education and support should focus on the timing, duration, and activity type, alongside support to optimise HbA1c.


S.K.Rilstone: None. N.Oliver: Advisory Panel; Dexcom, Inc., Medtronic, Roche Diabetes Care, Research Support; Dexcom, Inc., Medtronic. I.F.Godsland: None. N.E.Hill: Research Support; Dexcom, Inc.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at