Physical activity (PA) in T1D patients has been linked to improved cardiovascular and psychosocial health. Glycemic control has been reported as a benefit of PA, but it remains difficult to assess ’real-life’ PA in T1D pediatric patients.
In a novel study, we examined the relationship between weekly Fitbit-measured PA and blood glucose (BG) in 1patients over 6 months. BG was measured via home meters. Using generalized linear mixed effects regression we examined the association between weekly PA measures—steps and hours of moderate to vigorous PA (MVPA)—and weekly summaries of BG levels in, below, and above target range (70-140 mg/dL).
On average, patients took 56K steps/week and had a weekly BG measurement of 221 mg/dL. In regression models, as PA increased, patients tended to have significantly lower mean BG levels and more BG measurements in target range. Thus, an increase in 10K steps was associated with a 3.1 mg/dL drop in BG levels and a 6% increase in the odds of a target range BG (see Table). An additional hour of MVPA was associated with a 2.3 mg/dL drop in BG levels and a 4% increase in the odds of a target BG measurement. PA was also associated with being below the target range.
| Generalized linear mixed effects regression models examining relationship between weekly PA and BG levels1 (n=105) | ||||
| Average BG measurement | % BG measurement between 70-140 mg/dL | % BG measurement below 70 mg/dL | % BG measurement above 140 mg/dL | |
| β (95% CI)2 | OR (95% CI)3 | OR (95% CI)3 | OR (95% CI)3 | |
| Steps (per 10K) | -3.1 (-4.4, -1.8) | 1.06 (1.05, 1.07) | 1.06 (1.03, 1.08) | 0.94 (0.93, 0.95) |
| Hours of MVPA | -2.3 (-3.6, -1.0) | 1.04 (1.03, 1.05) | 1.03 (1.01, 1.05) | 0.96 (0.95, 0.97) |
| Generalized linear mixed effects regression models examining relationship between weekly PA and BG levels1 (n=105) | ||||
| Average BG measurement | % BG measurement between 70-140 mg/dL | % BG measurement below 70 mg/dL | % BG measurement above 140 mg/dL | |
| β (95% CI)2 | OR (95% CI)3 | OR (95% CI)3 | OR (95% CI)3 | |
| Steps (per 10K) | -3.1 (-4.4, -1.8) | 1.06 (1.05, 1.07) | 1.06 (1.03, 1.08) | 0.94 (0.93, 0.95) |
| Hours of MVPA | -2.3 (-3.6, -1.0) | 1.04 (1.03, 1.05) | 1.03 (1.01, 1.05) | 0.96 (0.95, 0.97) |
1 Adjusted for baseline HbA1c, baseline BMI (z-score), age group, sex, treatment group.
2 Beta is the fixed effects linear regression coefficient for the specified PA measurement.
3 OR is the odds ratio of the indicated event for each unit increase of PA.
This study provides evidence for the positive effects of PA on BG among pediatric T1D patients, providing a further impetus for promoting PA among this population.
T.L. Barnes: None. D.A. Watson: None. L.M. Gandrud: Consultant; Self; UnitedHealth Group.
