Calibrating self‐reported MVPA strengthens relationship with metabolic syndrome

Background Inaccurate self‐report of physical activity can weaken the association between moderate‐to‐vigorous physical activity (MVPA) and health outcomes. We explored the use of regression calibration to adjust for reporting error in the relationship self‐reported MVPA and risk factors for metabolic syndrome. Methods Accelerometer‐measured MVPA and self‐reported MVPA data from years 2003–04 and 2005–06 of the National Health and Nutrition Examination Survey were used to develop a model to correct for measurement error in self‐reported MVPA ( calibration model ). Log‐binomial regression was used to examine the relationship between self‐reported MVPA and risk factors for, and overall risk of, metabolic syndrome ( risk model ). Risk factors for Metabolic Syndrome were based on the 2009 Joint Scientific Statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Risk factors included high blood pressure (systolic ≥130 mmHg, diastolic ≥85 mmHg, or taking medication to lower blood pressure), elevated blood triglycerides (≥150 mg/dL), low HDL‐cholesterol (men: <40 mg/dL; women: <50 mg/dL), elevated fasting plasma glucose (≥100 mg/dL, or taking medication to lower blood glucose), and high waist‐circumference (men: >102 cm; women: >88 cm). Metabolic syndrome was defined as having three or more of these risk factors. Estimates from the calibration and risk models were combined to obtain estimates of odds ratios corrected for measurement error in self‐reported MVPA. Results On average, respondents reported 300.5 (± 15.2) more weekly minutes of moderate‐to‐vigorous physical activity (MPVA) than was captured by accelerometer ( p <0.01). Over‐reporting was inversely related to body mass index, age, and female gender. Over‐reporting of MVPA was positively related to education level and Hispanic ethnicity (vs. non‐Hispanic White) ( p <0.01). Quintile of self‐reported MVPA was inversely related to high waist circumference (OR: 0.88; 95% CI: 0.82, 0.94), low HDL‐C (OR: 0.91; 95% CI: 0.87, 0.95), and high triglycerides (OR: 0.92; 95% CI: 0.86, 0.98), but not significantly related to high blood pressure, high fasting blood glucose, or overall risk of metabolic syndrome. In calibrated models, quintile of self‐reported MVPA was inversely related to high blood pressure (OR: 0.65; 95% CI: 0.57, 0.73), high waist circumference (OR: 0.37; 95% CI: 0.31, 0.43) low HDL‐C (OR: 0.47: 95% CI: 0.40; 0.54), high triglycerides (OR: 0.50; 95% CI: 0.44, 0.58), and overall risk of metabolic syndrome (OR: 0.75; 95% CI: 0.66, 0.85). Self‐reported MVPA was not significantly associated with high fasting blood glucose. Conclusion Adjusting for reporting error considerably strengthens the association between self‐reported MVPA and risk factors for, and overall risk of, metabolic syndrome. Support or Funding Information Funding for this study comes from the National Institutes of Health, National Cancer Institute (5 R25 CA057730‐24) 1 Coefficients from calibration model, and coefficients and odds ratios from risk model and corrected risk model of the relationship between quintile of self‐reported moderate‐to‐vigorous physical activity and risk factors for, and overall risk of, metabolic syndrome among us adults1Outcome Calibration model Risk model Corrected risk modelλ 1 SE(λ 1 ) Var(λ 1 ) β 1 SE(β 1 ) Var(β 1 ) OR 1 95% Cl β c SE(β c ) Var(β c ) OR c 95% ClHigh blood pressure 2 0.13 0.01 0.00 −0.06 0.03 0.00 0.95 0.89 1.01 −0.44 0.07 0.00 0.55 0.57 0.73 High waist circumference 3 0.13 0.01 0.00 −0.13 0.03 0.00 0.88 0.82 0.94 −1.00 0.08 0.01 0.37 0.31 0.43 LOW HDL‐C 4 0.13 0.01 0.00 −0.10 0.02 0.00 0.91 0.87 0.95 −0.76 0.08 0.01 0.47 0.40 0.54 Elevated blood Triglycerides 5 0.13 0.01 0.00 −0.09 0.03 0.00 0.92 0.86 0.98 −0.68 0.07 0.01 0.50 0.44 0.58 Elevated fasting plasma glucose 6 0.13 0.01 0.00 0.01 0.04 0.00 1.01 0.94 1.09 0.07 0.06 0.00 1.07 0.95 1.20 Metabolic syndrome 7 0.13 0.01 0.00 −0.07 0.03 0.00 0.93 0.88 0.99 −0.54 0.07 0.00 0.58 0.51 0.661 Statistical models were adjusted for body mass index (BMI), age, gender, education level, race/ethnicity, and annual household income 2 Defined as systolic blood pressure≥130 mmHg, diastolic blood pressure ≥85 mmHg, or taking medication to lower blood pressure 3 Defined as having a waist circumference >102 cm in men, or a waist circumference >88 cm in women 4 Defined as having plasma HDL‐cholesterol <40 mg/dL in men, or <50 mg/dL in women 5 Defined as having fasting plasma triglycerides ≥150mg/dL 6 Defined as having fasting plasma glucose ≥100 mg/dL 7 Defined as having three or more of the above risk factors