Abstract
Background/Aims Physical inactivity is highly prevalent in US adults and significantly increases mortality risk. We examined the impact of implementing a new primary care visit process to record patient exercise frequency and intensity (“Exercise as a Vital Sign” [EVS]) at the beginning of each visit.
Methods We conducted an observational, quasi-experimental cohort study of 696,267 patients and 1,188 primary care providers to examine the impact of EVS implemented between April 2010 to October 2011 in 4 of 15 regional medical centers. Patients without primary care physicians (PCPs) or with co-morbid conditions that precluded exercise were excluded. We assessed documentation of exercise in PCP progress notes; healthy lifestyle-related referrals (e.g., exercise programs, nutrition and weight loss consultation); and changes in weight or glycemic control in patients with overweight or type 2 diabetes.
Results EVS implementation was associated with greater exercise-related PCP progress note documentation (26.2% vs. 23.7% of visits, odds of visit documentation increased 1.12 [95% CI:1.11–1.13] adjusting for demographic differences and repeated measures); and a small but significant increase in lifestyle-related referrals (2.1% vs. 1.7%; aOR 1.14 [1.11–1.18]), particularly in obese patients (BMI ≥30 kg/m2; 4.0% vs. 3.2% visits, P <0.001) and patients with type 2 diabetes (3.5% vs. 2.6% visits, P <0.001) compared to visits without EVS. Among patients with at least two BMI measures (n = 633,864), patients in EVS medical centers had slight weight loss (−0.002 kg/m2) whereas control site patients had weight gain (+0.02 kg/m2). Difference in weight change between facilities favored EVS in linear models controlling for baseline differences (−0.03 kg/ m2 [95% CI: −0.04 to − 0.02 kg/m2]). Differences in A1c (n = 70,083) were similarly small but favored EVS facilities (change in A1c: −0.06% [95% CI: −0.08 to −0.05%]).
Conclusions Exercise data collection during initial patient intake resulted in increased PCP progress note documentation and lifestyle-related referrals. The population-level impact on weight and glycemic control was favorable but slight. We conclude that EVS is a necessary but insufficient first step towards improving the identification and treatment of physical inactivity in primary care.
