Relation Between Physical Activity, Sedentary Behavior And Chronic Disease Risk Factors Using Principal Component Analysis.

Abstract

INTRODUCTION: The overall health status of individual’s with chronic disease (CD) are positively and negatively affected by physical activity (PA) and sedentary behavior (SB), respectively. The purpose of this study was to examine the relation between PA, SB and selected indices of health in a diverse CD population using a principal component analysis (PCA). METHODS: Participants (n=237, 54.4% female, age (mean±SD) 62.2±11.1 yr) were recruited at induction to a community-based exercise program for CD. Primary CD included cardiovascular (n=101), respiratory (n=48), cancer (n=80), diabetes (n=34), arthritis (n=26) and unclassified (n=78). BMI and waist to hip ratio (WHR) were measured and calculated using standard procedures. Upper and lower body strength, flexibility and cardiorespiratory fitness were assessed using a hand-grip test, sit-to-stand test (STS), sit and reach test (SRT), and 6-min time trial (6MTT), respectively. PA and SB were recorded using an activPAL3 micro accelerometer. QoL was assessed using the EQ5D VAS and the PHQ8. Fasting serum levels of glucose, triglycerides, HDL-C, LDL-C and CRP were measured. Blood pressure (BP) was measured using a 24-hour ambulatory BP monitor. ActivPAL generated PA and SB variables were analyzed using PCA. General linear models were used to investigate the association between PA and SB and indices of health. RESULTS: PCA analysis of sedentary time, standing time, stepping time, LIPA, MVPA, step count, sedentary bout lengths and total number of sedentary bouts generated three distinct factor; i) prolonged sedentary behavior (PSB), ii) physical activity (PA), and iii) broken sedentary behavior (BSB). The three derived variables account for 86% of the total the variance in PA and SB. There was a significant main effect for PSB on LDL-C (F (1,189) = 9.06) and PHQ8 scores (F(1,162) = 6.82). There was a significant main effect for PA on BMI (F(1,99) = 14.48), WHR (F(1,99) = 5.77), STS (F(1,222) =77.08), 6 MTT (F(1,222) = 77.08), EQ5D VAS (F(1,162) = 14.13), triglycerides (F(1,188) = 4.95), CRP (F(1,155) = 4.28), and systolic BP (F(1,99) = 4.94). There was a significant main effect for BSB on HDL cholesterol (F(1,188) = 6.25). CONCLUSIONS: The PCA derived factors PSB, PA and BSB are associated with established disease risk factors in patients with CD