Association of physical activity metrics with glucose variability in people with type 1 diabetes: A cross-sectional study

This study aims to investigate the association of physical activity metrics with measures of glucose variability in people with type 1 diabetes. From August 2019 to January 2022, people with type 1 diabetes, attending clinics or participating in ongoing research at the Dasman Diabetes Institute in Kuwait, were invited to participate in the study. Physical activity was measured over a 7-day period using a wrist-worn accelerometer, and glucose variability data were measured by continuous glucose monitoring (CGM) of the same period. Three hundred and eleven participants were recruited (age 33 (10) years, BMI 27(5) kg/m2 and n = 311 (169 female and 142 male)). Overall physical activity levels were not associated with any measure of glucose variability. The intensity gradient, which measures the distribution of physical activity intensity, was negatively associated with mean glucose (-1.01(-0.28, -1.74) and p = 0.007), CONGA (-1.00(-0.28, -1.72) and p = 0.007), J-index (-11.7(-2.23, 21.2) and p = 0.016), HBGI (-2.73(-0.44, -5.02) and p = 0.020), GRADE (-2.27(-0.59, -3.95), p = 0.009) and GRADE - euglycaemia (-4.26(-0.46, -8.06) and p = 0-029) and the M-value (-4.41 (-0.05, -8.77) and p = 0.049). Overall physical activity remains important, but it may be worth recommending people with type 1 diabetes to spend proportionately more of their day doing moderate to higher intensity physical activity, although this remains to be confirmed in an appropriately designed trial. Physical activity is recommended to people with type 1 diabetes due to its broad health benefits. The relationship between physical activity and glucose variability is unclear. The current study shows that overall physical activity levels are not associated with measures of glucose variability, but spending proportionately more of their day doing moderate to higher intensity physical activity was associated with better glucose variability.