Accelerometer-Measured Sedentary Behavior and Risk of CVD

Abstract

BACKGROUND Beyond serving as a marker for insufficient physical activity, sedentary behavior may directly affect future cardiovascular (CV) disease risk.

OBJECTIVES This study sought to examine associations between accelerometer-measured sedentary behavior with risk of specific CV outcomes, including potential relations with moderate to vigorous physical activity (MVPA).

METHODS Among participants of the UK Biobank prospective cohort study, we fit Cox models adjusted for demographic and lifestyle factors to assess associations between accelerometer-measured daily sedentary time with incident atrial fibrillation (AF), myocardial infarction (MI), heart failure (HF), and CV mortality. We assessed the potential effect of MVPA on associations between sedentary time and CV disease by including MVPA as an adjustment variable, as well as performing subgroup analyses stratified at the guideline-recommended MVPA threshold (ie, ≥150 min/wk). We then performed compositional analyses to estimate the effects of reallocating sedentary time to other activities.

RESULTS Among 89,530 individuals (age 62 ± 8 years, 56.4% women) undergoing 1 week of accelerometry, median sedentary time was 9.4 h/d (Q1-Q3: 8.2-10.6). In multivariable models, using the second quartile (8.2-9.4 h/d) as a referent, sedentary time in the top quartile (>10.6 h/d) was associated with greater risks of HF (HR: 1.45; 95% CI: 1.28-1.65) and CV mortality (HR: 1.62; 95% CI: 1.34-1.96), with an inflection of risk at 10.6 h/d. Higher sedentary time was also associated with greater risks of incident AF (HR: 1.11; 95% CI: 1.01-1.21) and MI (HR: 1.15; 95% CI: 1.00-1.32), with an approximately linear relation. Associations with HF and CV mortality persisted among individuals meeting guideline-recommended MVPA levels. Among individuals with >10.6 h/d of sedentary time, reallocating sedentary behavior to other activities substantially reduced the excess CV risk conferred by sedentary behavior (eg, 30-minute decrease in sedentary time for HF: HR: 0.93; 95% CI: 0.90-0.96), even among individuals meeting guideline recommended MVPA (HR: 0.93; 95% CI: 0.87-0.99).

CONCLUSIONS Sedentary behavior is broadly associated with future adverse CV outcomes, with particularly prominent effects on HF and CV mortality, where risk inflected at approximately 10.6 h/d. Although guideline-adherent MVPA partially mitigates excess risk, optimizing sedentary behavior appears to be important even among physically active individuals.

Introduction

Insufficient physical activity is a risk factor for incident cardiovascular (CV) disease. ^[1,2] Current guidelines emphasize the importance of achieving ≥150 minutes of moderate to vigorous physical activity (MVPA) per week to promote cardiovascular health.^[3] However, MVPA generally represents a small fraction (<2%) of overall physical activity.^[4,5]

By contrast, adults in the United States spend an average of 9.5 hours/day engaging in sedentary behavior.^[6] Sedentary behavior refers to waking activity with low energy expenditure (≤1.5 metabolic equivalents) while sitting, reclining, or lying down. Sedentary behavior differs from inactivity, which denotes insufficient MVPA.^[6] Despite accumulating evidence that excess sedentary behavior may be associated with CV disease risk directly rather than simply serving as a marker of insufficient MVPA, current physical activity guidelines fail to provide specific guidance on sedentary behavior.^{[7,8,9,10,11,12,13]} In particular, the specific levels of sedentary time at which CV disease risk may inflect, and potential relations between sedentary behavior and physical activity with regard to CV disease risk, remain poorly understood.^[14]

Most analyses of sedentary behavior have relied on self-reported data, which are prone to recall bias and can underestimate sedentary time by 40% to 60%.^[7,10] More recently, wearable accelerometers have facilitated more precise and reproducible quantification of free-living activity. Furthermore, recent machine learning methods validated against manually labeled activities provide substantially more accurate classification of movement behaviors^[15,16,17] than do approaches relying solely on intensity thresholds.^[18,19,20] A recent study leveraging machine learning–based behavior classification demonstrated that every 1 hour/day increase in sedentary behavior was associated with a 5% higher risk of overall CV disease.^[15] Although important, such a broad focus limits insight into relations between sedentary behavior and specific CV outcomes, which may differ on the basis of individual diseases.^[7,9,11,21]

Here, we leverage a unique resource of nearly 90,000 middle-aged and older adults who underwent prospective accelerometer-based physical activity measurement and behavior classification using a validated machine learning–based method, to examine associations of sedentary behavior with future risks of atrial fibrillation (AF), heart failure (HF), myocardial infarction (MI), and CV mortality, including the potential interplay between sedentary behavior and MVPA.

Authors and Disclosures

Ezimamaka Ajufo, MD,^a,b; Shinwan Kany, MD, MSC,^b,c,d; Joel T. Rämö, MD, PHD,^b,e; Timothy W. Churchill, MD,^f,g; J. Sawalla Guseh, MD,^f,g; Krishna G. Aragam, MD, MS,^b,f,g; Patrick T. Ellinor, MD, PHD,^b,f,h,* Shaan Khurshid, MD, MPH^b,f,h,*

From the ^aCardiology Division, Brigham and Women’s Hospital, Boston, Massachusetts, USA; ^bCardiovascular Disease Initiative, The Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA; ^cDepartment of Cardiology, University Heart and Vascular Center Hamburg-Eppendorf, Hamburg, Germany; ^dGerman Center for Cardiovascular Research (DZHK), Partner Site Hamburg/Kiel/Lübeck, Hamburg, Germany; ^eInstitute for Molecular Medicine Finland, Helsinki Institute of Life Science, University of Helsinki, Helsinki, Finland; ^fCardiovascular Research Center, Massachusetts General Hospital, Boston, Massachusetts, USA; ^gCardiology Division, Massachusetts General Hospital, Boston, Massachusetts, USA; and the ^hTelemachus and Irene Demoulas Family Foundation Center for Cardiac Arrhythmias, Massachusetts General Hospital, Boston, Massachusetts, USA. *The last 2 authors were joint supervisors of this work.

FUNDING SUPPORT AND AUTHOR DISCLOSURES Dr Ajufo is supported by the John S. LaDue Memorial Fellowship in Cardiovascular Medicine or Vascular Biology grant. Dr Kany is supported by the Walter Benjamin Fellowship from the Deutsche Forschungsgemeinschaft (521832260). Dr Rämö is supported by a research fellowship from the Sigrid Jusélius Foundation. Dr Churchill is supported by the National Institutes of Health (K23HL159262-01A1). Dr Guseh is supported by the American Heart Association (19AMFDP34990046) and the President and Fellows of Harvard College (5KL2TR002542-04). Dr Aragam is supported by grants from the National Institutes of Health (1K08HL153937) and the American Heart Association (862032); receives sponsored research support from Sarepta Therapeutics and Bayer AG; and receives a research collaboration with the Novartis Institutes for Biomedical Research. Dr Ellinor is supported by grants from the National Institutes of Health (RO1HL092577, R01HL157635), from the American Heart Association (18SFRN34230127, 961045), and from the European Union (MAESTRIA 965286); receives sponsored research support from Bayer AG, IBM Research, Bristol Myers Squibb, Pfizer and Novo Nordisk; and has served on Advisory Boards and/or consulted for Bayer AG. Dr Khurshid is supported by the NIH (K23HL169839-01) and the American Heart Association (2023CDA1050571).

ADDRESS FOR CORRESPONDENCE: Dr Shaan Khurshid, Cardiac Arrhythmia Service and Cardiovascular Research Center, Massachusetts General Hospital, 55 Fruit Street, GRB 109, Boston, Massachusetts 02114, USA. E-mail: skhurshid@mgb.org.

Accelerometer-Measured Sedentary Behavior and Risk of Future Cardiovascular Disease

Abstract

Introduction

Tables

References

Authors and Disclosures

Authors and Disclosures

Email This

Feedback