Research Article

The Association between Leisure-Time Physical Activity and Risk of Undetected Prediabetes

Table 3

Odds ratios of prediabetes according to category of leisure-time physical activity, stratified by age, NHANES 2007–2012, ages 20–65 y.

CaseParticipant CrudeModel 1Model 2
OR95% CIOR95% CIOR95% CI

20–44 y
Total leisure-time physical activity (MET-hours/week)
None (reference)35220621.001.001.00
Low13010370.700.56, 0.860.710.57, 0.890.880.69, 1.11
Moderate1007890.710.56, 0.900.730.57, 0.930.900.70, 1.17
High12510820.630.51, 0.790.660.53, 0.830.780.61, 0.99
P-trend<0.001<0.0010.050
Vigorous leisure-time physical activity (MET-hours/week)
None (reference)52432271.001.001.00
Low606600.520.39, 0.680.530.40, 0.710.610.45, 0.83
Moderate645290.710.54, 0.940.710.54, 0.950.760.56, 1.03
High595540.610.46, 0.820.650.49, 0.870.720.53, 1.00
P-trend<0.001<0.0010.007
Moderate leisure-time physical activity (MET-hours/week)
None (reference)41626981.001.001.00
Low868040.660.51, 0.840.660.51, 0.850.900.69, 1.18
Moderate1047600.870.69, 1.100.910.72, 1.161.210.93, 1.56
High1017080.910.72, 1.150.920.73, 1.171.140.88, 1.48
P-trend0.1950.3050.171

45–65 y
Total leisure-time physical activity (MET-hours/week)
None (reference)71317311.001.001.00
Low2276630.740.62, 0.900.730.61, 0.880.880.72, 1.07
Moderate1584600.750.60, 0.930.760.61, 0.940.980.78, 1.23
High1093800.570.45, 0.730.600.47, 0.760.730.57, 0.95
P-trend<0.001<0.0010.047
Vigorous leisure-time physical activity (MET-hours/week)
None (reference)104026651.001.001.00
Low662210.660.49, 0.890.700.52, 0.950.860.62, 1.18
Moderate562040.590.43, 0.810.630.46, 0.870.790.57, 1.12
High451540.640.45, 0.920.670.47, 0.960.860.58, 1.26
P-trend<0.001<0.0010.135
Moderate leisure-time physical activity (MET-hours/week)
None (reference)78319371.001.001.00
Low1344050.730.58, 0.910.740.59, 0.930.910.72, 1.16
Moderate1514700.700.56, 0.860.690.55, 0.850.880.70, 1.10
High1394220.720.58, 0.900.710.56, 0.890.870.69, 1.11
P-trend<0.001<0.0010.150

OR, odds ratio; CI, confidence interval; MET, metabolic equivalent.
Model 1 adjusted for age and gender.
Model 2 adjusted for age, gender, race, BMI, educational level, smoking status, alcohol consumption, dietary pattern, daily total energy intake, hypertension, total work-related physical activity, and walking or bicycling for transportation.