Sedentary Time and Its Association With Risk for Disease Incidence, Mortality, and Hospitalization in Adults: A Systematic Review and Meta-analysis

 
Biswas and colleagues conducted a meta-analysis of observational studies to examine the association between sedentary time and various health outcomes (1). That analysis suggested that sedentary time was associated with an increased risk of all-cause mortality, but suffered from substantial heterogeneity (I2 = 94.96%). We wonder whether further sensitivity analyses were needed to elucidate the cause of this heterogeneity.
First, sample sizes varied across studies, ranging from 217 to 240 819. A forest plot suggested that five of the seven studies with <10 000 participants presented more exaggerated point estimates of hazard ratios when compared with the remaining eight studies. In clinical trials in medicine, large treatment effects are known to be derived from small-sized trials (2). Likewise, smaller observational studies could be hypothesized to present larger hazard ratios compared with larger studies. An analysis examining the association between sample size and all-cause mortality might therefore be intriguing.
Second, many of the studies included in the analysis of all-cause mortality were susceptible to attrition bias. Among the 13 studies, the completeness of one was unclear, whereas that of six in the long term was less than 80%. A sensitivity analysis excluding studies at risk of attrition bias should be considered.
Third, the selection criteria of participants included in the meta-analysis were somewhat unclear. For example, participants included from one study had a sitting time of ≥11 h/day (3), while those from another study sat ≥9 h/day (4). Biswas et al. admitted that operational definitions and cutoffs were applied during the categorization of sedentary time. If so, analysis using different cutoffs of calculated energy expenditures or sedentary time might identify populations at higher risk of mortality.
Finally, how the authors extracted the information from certain populations is unclear. All participants from a study seemed to be included in the analysis, while Biswas et al. quoted the number of deaths from the subgroup of sedentary time ≥11 h/day from the same study (5). Clarification of this inconsistency is needed.
The studies included in this systematic review were heterogeneous in terms of age. Every life stage has its own societal role and lifestyle. Future research should define target generations or life stages to clarify who could benefit in the long run from not being sedentary.


References
1. Biswas A, Oh PI, Faulkner GE, Bajaj RR, Silver MA, Mitchell MS, et al. Sedentary Time and Its Association With Risk for Disease Incidence, Mortality, and Hospitalization in Adults: A Systematic Review and Meta-analysis. Ann Intern Med. 2015;162(2):123-32.
2. Pereira TV, Horwitz RI, Ioannidis JP. Empirical evaluation of very large treatment effects of medical interventions. JAMA. 2012;308(16):1676-84.
3. Seguin R, Buchner DM, Liu J, Allison M, Manini T, Wang CY, et al. Sedentary behavior and mortality in older women: the Women's Health Initiative. Am J Prev Med. 2014;46(2):122-35.
4. Matthews CE, George SM, Moore SC, Bowles HR, Blair A, Park Y, et al. Amount of time spent in sedentary behaviors and cause-specific mortality in US adults. Am J Clin Nutr. 2012;95(2):437-45.
5. van der Ploeg HP, Chey T, Korda RJ, Banks E, Bauman A. Sitting time and all-cause mortality risk in 222 497 Australian adults. Arch Intern Med. 2012;172(6):494-500.

The cited review of Dr. Morris’ population studies, which were done more than fifty years ago, (1) supports his conclusions, which, in turn, supports the current Meta-analysis.
(1) Paffenbarger RS, Blair SN, Lee I-M. A history of physical activity, cardiovascular health and longevity: The contribution of Jeremy N Morris, D Sc, DPH, FRCP. Int.J. Epidemiol.2001 30 (5): 1184-1192.doi:10.1093/ise/30.5.1184

We acknowledge the concerns of Urushidani and Kuriyama regarding our study’s heterogeneity. Indeed, we focused much of our discussion section to factors that may have contributed to heterogeneity.
First, we agree that smaller sample sizes of two studies (1, 2) may exaggerate point estimates. Nonetheless, the majority of studies based on large cohorts (all but 2 based on >500 participants) show a consistent positive association, and then the precision of effect size should not change sufficiently to merit further examination.
Second, with regard to attrition bias, excluding studies at risk of attrition bias may be helpful, but conclusions may be misleading due to variability of follow-up times. This may unfairly penalize studies with lower completeness but longer follow-up duration over studies with high completeness but shorter follow-up.
Third, we agree for a more standardized approach to quantifying sedentary times. We ensured our findings were comparable by prioritizing the longest reported sitting time, and if this was not reported directly, the most comparable measure was selected (e.g. longest screen time). We resisted standardizing sedentary cut offs as the variability may lead to misleading conclusions. For example, a question asking the “time spent sitting while doing things, such as visiting friends etc.” (2) may elicit a different response to asking “about how many hours in each 24-hour day do you usually spend sitting?” (3).
Finally on the last point of inconsistencies extracting information from a study (3), although we reported deaths from all-cause mortality for sedentary time ≥11 h/day, this was based on the combined total (both) for men and women i.e. 649 deaths/222,497 participants. Information was extracted from the total population of men and women across all studies, and we combined results when presented separately for men and women.
Heterogeneity remains one of many important limitations. Additional analyses while intriguing, will likely not address the many unanswered questions that remain. Instead, we hope that our meta-analysis serves as an impetus for future research.

1. George ES, Rosenkranz RR, Kolt GS. Chronic disease and sitting time in middle-aged Australian males: findings from the 45 and Up Study. Int J Behav Nutr Phys Act. 2013;10(1):20.
2. Pavey TG, Peeters GG, Brown WJ. Sitting-time and 9-year all-cause mortality in older women. Br J Sports Med. 2012.
3. van der Ploeg HP, Chey T, Korda RJ, Banks E, Bauman A. Sitting time and all-cause mortality risk in 222 497 Australian adults. Arch Intern Med. 2012;172(6):494.