No abstract

Health benefits of meeting the 2008 Aerobic Physical Activity (PA) Guidelines for Americans (PAG) are well-known; although it is not known how well responses to the PA vital sign (PAVS) that identify individuals as meeting or not meeting PAG relate to patient health. PURPOSE: To assess construct validity of the PAVS by examining odds of being overweight or obese (i.e. BMI≥25.0) or having a high Charlson Comorbidity Index (CCI) according to responses to the PAVS as meeting or not meeting PAG. METHODS: Patients who completed a PAVS between November 2011 and November 2013 at one primary care clinic were included in the study. Body weight status and CCI scores were automatically calculated from data in electronic health records (EHRs). Chi-square analysis was used to test differences in being normal weight versus overweight, and scoring below versus above the 50 th percentile of the CCI, according to meeting or not meeting PAG. Odds ratios (OR) from multivariate logistic models were used to determine if patients who reported to the PAVS as not meeting PAG were more likely to be overweight or obese, or score higher on the CCI compared to patients who reported to the PAVS as meeting PAG. RESULTS: The number of patient visits eligible for analyses from men and women ≥18 years was 34,712. Most participants were female (56.3%), older (≥65 yrs, 43.7%), overweight , 30.9%) and had a moderate disease burden (Charlson Comorbidity Index 2-4, 32.7%). Patients that did not meet PAG according to the PAVS were more likely than normal weight patients to have a higher BMI OR=1.19, p=0.001; OR=1.39, p<0.0001; OR=2.42, p<0.0001; BMI≥40, OR=3.7, p<0.0001). Patients not meeting PAG were also significantly more likely to have a higher disease burden (above 50 th Charlson percentile, OR=1.8, p<0.0001). CONCLUSION: The strong association of the PAVS with patient BMI and moderately-strong association with disease burden found in this study supports construct validity of the PAVS because these two constructs are each known to be influenced by, and influence, PA levels. Future assessments of measurement properties of the PAVS should be done with objective and repeated measures of PA in the patient population.