Contributions of changes in physical activity, sedentary time, diet and body weight to changes in cardiometabolic risk

Eivind Andersen,Alex Mcconnachie,Sally Wyke,Annie S Anderson,Judith G M Jelsma,Naveed Sattar,Nanette Mutrie,Jason M R Gill,Marit Sørensen,Glyn C Roberts,Willem Van Mechelen,Hidde P Van Der Ploeg,Femke Van Nassau,Hugo V Pereira,Marlene N Silva,Øystein B Røynesdal,Kate Hunt,Cindy M Gray

doi:10.1186/s12966-021-01237-1

Abstract

BackgroundIncreased physical activity (PA), reduced time spent sedentary (SED), healthier diet and reduced body weight may all have a positive impact on cardiometabolic risk. The relative importance of change in each of these variables on cardiometabolic risk, however, is unclear. We therefore sought to investigate the relative contributions of changes in PA, SED, diet and body weight on cardiometabolic risk.MethodsThis is a secondary analysis of data collected from the EuroFIT randomised controlled trial, which was a 12-week group-based lifestyle intervention for overweight middle-aged men delivered by coaches in football club stadia aiming to improve PA, SED, diet, and body weight. PA and SED were assessed by accelerometry, diet using the Dietary Instrument for Nutrition Education (DINE). An overall cardiometabolic risk score was derived from combining z-scores for glucose, HbA1c, insulin, lipids and blood pressure. In total, 707 men (from the overall cohort of 1113) with complete data for these variables at baseline and 12-month follow-up were included in the multivariable linear regression analyses.ResultsIn multivariable analyses, change in number of steps (explaining 5.1% of R2) and dietary factors (less alcohol, fatty and sugary food, and more fruit and vegetables) (together explaining 4.5% of R2), but not changes in standing time or SED, were significantly associated with change in body weight. Changes in number of steps (R2 = 1.7%), fatty food score (R2 = 2.4%), and sugary food score (R2 = 0.4%) were significantly associated with change in cardiometabolic risk score in univariable models. However, in multivariable models which included changes in weight as well as changes in steps and dietary variables, change in weight explained a substantially larger proportion of the change in cardiometabolic risk score, explaining 14.1% of R2 (out of an overall model R2 of 19.0%). When baseline (as well as change) values were also included in the model, 38.8% of R2 for change in cardiometabolic risk score was explained overall, with 14.1% of R2 still explained by change in weight.ConclusionChange in body weight, together with baseline cardiometabolic risk explained most of the change in cardiometabolic risk. Thus, the benefits of increasing physical activity and improving diet on cardiometabolic risk appear to act largely via an effect on changes in body weight.Trial registrationInternational Standard Randomised Controlled Trials, ISRCTN-81935608. Registered 06052015. https://www.isrctn.com/ISRCTN81935608?q=&filters=recruitmentCountry:Portugal&sort=&offset=7&totalResults=92&page=1&pageSize=10&searchType=basic-search

Highlights

Increased physical activity (PA) and a healthy diet have consistently shown to be associated independently with lower risk of a range of chronic non-communicable diseases [1, 2], and have been a primary focus of the public health guidelines for many years [2,3,4]
Main reasons for exclusion for men who showed interest in the trial were body mass index (BMI)
A final sample of 707 men was included in this secondary analysis, after excluding those with missing ActivPAL data (n=198), or missing cardiometabolic risk score data (n=322), which included 114 men who were missing both

Summary

Introduction

Increased physical activity (PA) and a healthy diet have consistently shown to be associated independently with lower risk of a range of chronic non-communicable diseases [1, 2], and have been a primary focus of the public health guidelines for many years [2,3,4]. The extent of change in CVD risk factors in response to a lifestyle intervention is related to the extent of body weight loss [10], and loss of body fat is likely to be an important mediator between changes in PA and diet and change in risk factors for cardiometabolic diseases [11]. Both dietary improvements [12] and increases in PA [13] have been shown to change biomarkers of cardiometabolic risk, without a concomitant reduction in body weight. We sought to investigate the relative contributions of changes in PA, SED, diet and body weight on cardiometabolic risk

Methods

Results

Discussion

Conclusion