Sedentary Behavior and Cardiovascular Risk in Older Adults: a Scoping Review

Abstract

Sedentary behavior is an emerging risk factor for cardiovascular disease (CVD) and may be particularly relevant to the cardiovascular health of older adults. This scoping review describes the existing literature examining the prevalence of sedentary time in older adults with CVD and the association of sedentary behavior with cardiovascular risk in older adults. We found that older adults with CVD spend >75 % of their waking day sedentary, and that sedentary time is higher among older adults with CVD than among older adults without CVD. High sedentary behavior is consistently associated with worse cardiac lipid profiles and increased cardiac risk scores in older adults; the associations of sedentary behavior with blood pressure, CVD incidence, and CVD-related mortality among older adults are less clear. Future research with larger sample sizes using validated methods to measure sedentary behavior are needed to clarify the association between sedentary behavior and cardiovascular outcomes in older adults.

Introduction

Adults over the age of 60, despite constituting just 18 % of the US population [1], comprise more than half of Americans with cardiovascular disease (CVD) [2] and account for more than two-thirds of US cardiovascular-related deaths [3]. With the aging of the baby boomers, the population of older adults is expected to double from 56 million in 2010 to over 100 million by 2040 [1]; the concomitant rise in CVD prevalence will place large burdens on the healthcare system. Intensified efforts to stem the impending upsurge of CVD are therefore urgently needed.

It is well established that engagement in moderate to vigorous physical activity plays a key role in both primary [4] and secondary [5] prevention of CVD. However, very few older adults engage in the 150 minutes of moderate or 75 minutes of vigorous physical activity per week recommended in cardiovascular disease prevention guidelines [6]. A recent study revealed that older adults perform, on average, less than 10 min of moderate to vigorous physical activity per day [7]. Such limited physical activity among older adults may be due to functional limitations, fatigue, and lack of desire [8]. Thus, it is necessary to find alternative means of improving cardiovascular fitness that match the desires and abilities of older adults.

Sedentary behavior (SB) has emerged in recent years as a novel risk factor for poor health outcomes, including CVD [9]. SB is defined as “any waking activity characterized by an energy expenditure less ≤1.5 METs (i.e., metabolic equivalent tasks) and (performed in) a sitting or reclining position,” [10•] and includes common activities such as reading, watching TV, using a computer, and other seated or lying activities. SB has been associated, independent of physical activity, with increased risks of all-cause mortality [11, 12], cardiovascular mortality [13], incident CVD [12–15], myocardial infarction [16], and heart failure [17] in general populations. Recent studies have demonstrated that people with existing CVD [18•] and stroke [19•] tend to be more sedentary than people without CVD, potentially increasing their risk for recurrent cardiovascular events and health decline.

SB may be a particularly important risk factor for cardiovascular outcomes among older adults, as older adults generally spend more of their day sedentary than any other age group [20]. A joint report issued by the National Institutes of Health’s National Heart, Lung, and Blood Institute and the National Institute on Aging in June of 2015 recognized the potential impact of SB on cardiovascular outcomes in older adults and called for more research to fill critical gaps in knowledge about SB and cardiovascular and aging health outcomes [21]. To address this need, the objective of this review was to describe the currently available literature examining the prevalence of SB in older adults with CVD and the association of SB with CVD risk in older adults, as well as to propose areas for future inquiry and intervention to mitigate the risks of SB on cardiovascular health.

Methods

We undertook a scoping review in accordance with the framework developed by Arksey and O’Malley [22]. The purpose of a scoping review is to provide an overview of the type, extent, and quantity of research available on a given topic in order to rapidly disseminate existing knowledge about a topic and identify opportunities for future research. Scoping reviews are often undertaken when the field of literature being reviewed is too young or sparse to be appropriate for systematic review.

Literature Search

Searches to identify relevant articles were performed in PubMed, Ovid, and Scopus in July 2015. The review was limited to English language reports of observational studies and trials that included older adults (age ≥60 years) or broader age groups for which results specific to older adults were reported. We required that SB be quantified either through accelerometry (e.g., metabolic equivalents or counts per minute) or by self-report. Articles defining SB as lack of physical activity, rather than as time spent sitting or lying, were excluded. In our examination of the association between SB and cardiovascular risk, we considered all cardiovascular risk factors with the exception of BMI/adiposity, diabetes/other glycemic aberrations, inflammatory markers, and the metabolic syndrome, as the authors judged these markers of metabolic risk to be outside the scope of this review. Keywords and Medical Subject Heading terms searched included sedentary, sitting, television (exposure), cardi* (truncated to include cardiac, cardiovascular, etc.), heart, stroke, myocardial infarction (outcome), and older adults, elderly, senior (population). Editorials, reviews, case studies, and meeting abstracts were excluded. Bibliographies of eligible articles were searched for additional references.

Article Selection and Data Abstraction

We performed an initial screen of article titles and abstracts to exclude studies that did not meet inclusion criteria. Full review of all remaining studies was undertaken to determine eligibility for inclusion in the review. One author (AMH) independently abstracted data from all included studies using a standardized form. Information abstracted from eligible articles included study type, length of follow-up (as applicable), name of parent study (as applicable), number of participants, age of participants, operationalization of sedentary behavior, cardiovascular condition or risk factor investigated, results, whether results were adjusted for or were otherwise independent of participants’ physical activity, and other information pertinent to interpretation of the study results (labeled “notes” in Table). Due to the small number of heterogeneous studies and in accordance with the recommended protocol for scoping reviews [22], quality assessment and meta-analysis were not performed.

Table 1.

Characteristics of selected titles

Author, year, reference number	Study type, length of follow-upa, parent studya	Sample size and age distribution	Operationalization of sedentary behavior	Cardiovascular disease or risk factor	Results	Results independent of physical activity?	Notes
Chase et al. 2014 [23]	Cross-sectional, observational	N = 54 physically active adults >65 mean age = 72 ± 0.6 years, range = 65–81	SenseWear Pro accelerometer, worn on arm (biaxial) Minutes of sedentary time (<1.5 METs) per day	Triglycerides, LDL, HDL, blood pressure	Higher sedentary time was associated with higher LDL (β = 0.32, p = 0.02). Correlation between sedentary time and LDL = 0.32 (p = 0.02). Sedentary time was not associated with triglycerides, HDL, or blood pressure.	Yes	Older adult sample was purposefully extraordinarily physically active-mean MVPA = 2.6 hours per day
Chomistek et al. 2013 [15]	Prospective longitudinal, observational Mean follow-up time = 12.2 (IQR: 8.7–14.0) years Women’s Health Initiative Observational Study (WHI-OS)	N = 71,018 women ≥50 at baseline Subgroup analysis performed in women ≥70 (n = 37,792)	Self-reported sitting per day, separated into tertiles: T1: ≤5 h per day T2: 5.1–9.9 h T3: ≥10 h	Incident cardiovascular disease (non-fatal MI, fatal CHD, and stroke)	Sitting time was associated with higher risk of cardiovascular disease (HR = 1.22, 95 % CI 1.09–1.36 for ≥10 h sitting per day) for women ≥70, but not for women <70	Yes
Ensrud et al. 2014 [27]	Prospective longitudinal, observational Mean follow-up time = 4.5 ± 1.0-years Osteoporotic Fractures in Men Study (MrOS)	N = 2918 men ≥65 years Mean age = 79.0 ± 5.2 years	SenseWear Pro accelerometer, worn on arm (biaxial) Minutes of sedentary time (≤1.5 METs) per waking day, separated into quartiles: Q1: <772 min Q2: 772–844 Q3: 844–915 Q4: ≥915	CVD-related death (ICD-9 codes 396.9–442, 966.71, 785.51)	# of CVD-related deaths = 138 HR (95 % CI) for CVD-related death for each quartile of SB: Q1: reference Q2: 1.59 (0.91–2.75) Q3: 1.12 (0.63–2.00) Q4: 1.71 (0.99–2.97) P value for trend = 0.16	No	Power may have been limited to detect significant associations
Fitzgerald et al. 2015 [31]	Cross-sectional analysis of intervention data Lifestyle Interventions and Independence for Elders (LIFE) study	N = 1170 adults ≥70 years with mobility impairment Mean age = 78.8 ± 5.3 years, range = 70–89	ActigGraph accelerometer, worn on the hip (triaxial) Minutes of sedentary time (i.e., <100 counts/minute)	Prevalence of sedentary time by history of CVD Framingham Hard Coronary Heart Disease (HCHD) risk score Blood pressure total cholesterol, HDL	Sedentary time was higher among participants with CVD (78.4 % of day) than those without CVD (76.4 %). Each minute of sedentary time was associated with a 0.04 % increase in HCHD risk score (or 1 % increase for each 25 min) in older adults with CVD, 0.03 % increase in HCHD risk score in older adults without CVD. Sedentary time associated with lower total and HDL cholesterol, but not blood pressure.	No	Time spent in any activity >100 counts/minute (i.e., non-SB, including light activity) was associated with decreased cardiovascular risk
Gao et al. 2007 [24]	Cross-sectional, observational	N = 455 Hispanic adults ≥60 years Mean age of sample not reported	Hours of TV viewing time per day, separated into quartiles: Q1: 0–1.5 h/day Q2: 1.6–3.4 Q3: 3.5–5.5 Q4: 5.6–18.0	Hypertriglyceridemia, HDL, hypertension, HDL-total cholesterol ratio	Greater TV viewing time per day significantly associated with lower HDL cholesterol, greater risk of hypertension, and higher HDL-total cholesterol ratio.	yes
Gennuso et al. 2015 [25]	Cross-sectional, observational National Health and Nutrition Examination Survey 2003–2006	N = 1914 adults ≥65 years Mean age = 74.6 ± 6 years	ActiGraph accelerometer, worn on hip (uniaxial) Minutes of sedentary time (i.e., <100 counts/minute), calculated into hours and split into quartiles: Q1: 0–7.92 h/day Q2: 2.93–9.17 Q3: 9.18–10.63 Q4: ≥10.64	Blood pressure, total cholesterol, HDL, LDL, triglycerides	SB was not significantly associated with CV risk factors in overall sample. Higher SB was associated lower systolic blood pressure in participants with low physical activity. Higher SB associated with lower LDL in older adults with high physical activity.	Yes	SB was positively associated with weight, BMI, waist circumference, C-reactive protein, plasma glucose, functional limitations
Seguin et al. 2014 [28]	Prospective longitudinal, observational Mean follow-up = 12 years Women’s Health Initiative (WHI) Study	N = 40,610 women 60–69 years, 22,344 women 70–79 years (out of full sample 92,234 women age 50–79 years)	Self-report of average time spent sitting or lying down per waking day, separated into categories: ≤4, >4–8, >8–11, >11 h/per day	Third party-confirmed CVD death or CHD death	SB was not significantly associated with CVD-related deaths in 60–69 or 70–79 year olds. High SB was associated with CHD death among 60–69 year olds (HR = 1.41, 95 % CI = 1.00–1.97 but not among 70–79 year olds.	Yes	Details about conditions included in CVD death and CHD death were not provided
Stamatakis et al. 2012 [26]	Cross-sectional, observational Health Survey for England (HSE)	N = 2765 adults ≥60 years with self-reported SB N = 649 adults ≥60 years with objectively measured SB	ActiGraph GT1M accelerometer (uniaxial) Minutes of sedentary time (i.e., <100 counts/minute) per day Self-report of leisure-time sitting activities and TV viewing	Total to HDL cholesterol ratio	Both objectively measured and self-reported SB associated with worse cholesterol ratio; association using objectively measured SB was stronger. Non-TV watching SB time not associated with cholesterol.	Yes	Standing vs. sitting time could not be distinguished
Van der Berg et al. 2014 [30]	Retrospective longitudinal, observational Average time from midlife risk factor assessment to SB assessment = 31 years Age, Gene/Environment Susceptibility (AGES)-Reykjavik Study	N = 565 older adults Mean age = 80 ± 4.7 years (range = 73–92 years)	ActiGraph GT3X accelerometer, worn on hip (triaxial) Minutes per day spent sedentary (<100 counts/minute)	Self-report of heart disease, hypertension, cholesterol, triglycerides	Presence of heart disease during midlife was the strongest predictor of SB in late life. Average sedentary time per day was 38.9 min greater among older adult with history of heart disease. Midlife hypertension, cholesterol, and triglycerides were not related to SB in old age.	Yes
Warren et al. 2010 [29]	Prospective, longitudinal, observational Follow-up = 21 years Aerobics Center Longitudinal Study (ACLS)	N = 891 men ≥60 years (out of full sample of 7744 age 20–89)	Self-reported time spent riding in a car per week, categorized into quartiles: Q1: <4 h/week Q2: 4–7 Q3: 7–10 Q4: >10	National death index-confirmed death attributable to ICD-9 codes 390–449.9 or ICD–10 codes 100–178	Average time spent riding in a car per week was not associated with risk of CVD death.	Yes

Notes: CHD coronary heart disease, CVD cardiovascular disease, HDL high-density lipoprotein, LDL low-density lipoprotein, METs metabolic equivalent tasks, MVPA moderate to vigorous physical activity, SB sedentary behavior

^aAs applicable

Results

Study Selection

The literature search yielded 6137 original articles, from which 415 duplicates were removed, leaving 5722 articles for review. Of these, 5686 were excluded based on title and abstract review. Of the 36 articles reviewed in full, 16 were excluded due to not reporting results specific to adults age 60 or older, seven did not examine SB, three were not related to CVD or CVD risk, two grouped SB with low-intensity physical activity, and one did not report results that were interpretable for the purposes of this review. Three articles were added from review of bibliographies to produce a total of ten articles included in this review (Fig. 1).

Fig. 1.

Selection for selected titles

Study Design and Sample Characteristics

Characteristics of included articles are detailed in the Table 1. All studies were observational—four were cross-sectional [23–26], four were prospective longitudinal [15, 27–29], one was retrospective longitudinal [30••], and one was a cross-sectional analysis of baseline data from an intervention study [31••]. All studies were published from 2007 to 2015. Sample sizes (either full sample or portion of sample ≥60 years) ranged from 54 [23] to 62,954 [28]. By design, all studies in this review included samples exclusively age ≥60 [23–27,30••, 31••] or reported on older subgroups [15, 28, 29], but studies varied in age limits—four studies reported on participants ≥60 [24, 26, 28, 29], three studies on participants ≥65 [23, 25, 27], two studies on participants ≥70 [15, 31••], and one study on participants age ≥73 [30••]. Four studies included generally unrestricted samples of older adults, with the remainder having restricted samples: two studies that included women only [15, 28], two studies that included men only [27, 29], one study that included only Hispanic elders [24], and two studies that examined samples of older adults with distinct physical activity characteristics, one of older adults with very high physical activity [23] and the other of inactive older adults with mobility limitations [31••].

Operationalization of Sedentary Behavior

Six studies assessed sedentary behavior using accelerometers, i.e., small wearable devices that directly measure bodily acceleration over one or more axes. Accelerometers were uniaxial in two studies [25, 26], biaxial in two studies [23, 27], and triaxial in two studies [30••, 31••]. All studies that measured SB via accelerometry used cut points of ≤1.5 METs [23, 27] or <100 counts per minute [25, 26, 30••, 31••] to define SB. Five studies used self-report to measure SB via responses to questions about the number of hours per day spent sitting or lying [15, 26, 28], responses to questions about hours of TV viewing per day [24, 26], or hours spent riding in a car per week [29]. One study used both accelerometry and self-report to operationalize SB [26].

Operationalization of Cardiovascular Disease, Cardiovascular Risk, and Cardiovascular-Related Death

Two studies [30••, 31••] assessed the prevalence of SB among participants with existing cardiovascular disease; in these studies, existing cardiovascular disease was identified via self-report. Five studies assessed the association of SB with objectively measured cardiovascular risk factors [23–26, 31••], including blood pressure/hypertension [23–25, 31••], blood lipids [23–26, 31••], and a cardiac risk score [31••]; all but one [25] of these studies accounted for antihypertensive or statin use in analyses. One study [15] assessed the association of SB with incident CVD, defined as physician-confirmed diagnoses of incident coronary heart disease, non-fatal myocardial infarction, fatal coronary heart disease, or stroke [15]. Lastly, three studies evaluated risk of SB with cardiovascular-related mortality, defined as physician- or National Death Index-confirmed deaths attributable to ICD-9 or ICD-10 codes associated with cardiovascular disease [27] or third party-confirmed death attributable to coronary heart disease, stroke, congestive heart failure, angina, peripheral vascular disease, or venous thromboembolism [28, 32].

Prevalence of SB in Older Adults with Cardiovascular Disease

The two studies [30••, 31••] that reported on the prevalence of SB in older adults with cardiovascular disease found that SB was higher among participants with CVD than among older adults without CVD. Fitzgerald et al. [31••] found a small difference in sedentary time among participants with CVD (78.4 % of waking day) compared with participants without CVD (76.4 %), translating into an average of 17 more minutes of SB per day among older adults with CVD. Van der Berg et al. [30••], who performed a retrospective analysis of correlates of SB in older persons, reported average daily SB time to be 10.3 h (75.3 % of waking day) among all older adults in the sample; a history of heart disease in midlife was associated with an average 38.9 minutes more sedentary time per day. In this study, history of heart disease was the strongest predictor of SB in old age among dozens of demographic, socioeconomic, lifestyle, and biomedical factors examined in midlife, lending evidence that CVD might predispose older adults toward greater SB. Importantly, both studies accounted for geriatric impairments that might influence SB prevalence in older adults—Van der Berg et al. [30••] adjusted for health status, mobility limitations, and joint pain, and Fitzgerald et al. [31••] performed their analyses exclusively in a sample of older adults at high risk for mobility impairment.

Sedentary Behavior and Cardiovascular Risk Factors

SB was associated with increased blood lipid levels and composite cardiac risk scores; findings for the association of SB with blood pressure were mixed. Four [23, 24, 26, 31••] of five studies [23–26, 31••] that examined the association of SB with blood lipids (i.e., total cholesterol, its constituents, and triglycerides) found that increases in SB assessed via self-report [24], accelerometry [23, 25, 31••], or both [26] were associated with increased LDL cholesterol [23], decreased HDL [24, 31••], and decreased HDL to total cholesterol ratio [24, 26]. However, one study found no association between SB and blood lipids [25], and others reported mixed findings for different domains of blood lipids [23, 24]. Three [23, 25, 31••] studies examining the association of SB with systolic or diastolic blood pressure did not report significant findings, whereas one cross-sectional study [24] found that greater hours of TV viewing per day was associated with a greater than twofold risk for hypertension (OR 2.5, 95 % CI 1.0–6.0), although this risk was slightly attenuated in multivariable-adjusted models. Fitzgerald et al. [31••] reported a dose–response relationship of SB with Framingham Hard Coronary Heart Disease risk scores [33] such that each minute of SB per day was associated with a 0.04 % increase in risk score among participants with existing CVD (i.e., 1 % increase in risk for every 25 min of SB per day) and a 0.03 % increase for participants without CVD.

Sedentary Behavior and CVD Incidence or CVD-Related Death

Findings regarding the association of SB with CVD incidence or CVD-related mortality were mixed. The one study [15] that examined the association between SB and incident CVD found that women age ≥70 who reported sitting ≥10 h per day had a higher incidence of CVD (HR 1.22, 95 % CI 1.09–1.36) than older women who sat <5 h per day. Interestingly, the association between sitting time and CVD incidence among women age 50 to 69 was not significant (HR 1.08 95 % CI 0.94–1.25). Conversely, other studies [28, 29] that examined the association of self-reported SB with CVD-or coronary heart disease-related mortality found that high SB was associated with CVD-related death in younger men [29] and women (HR 1.41 95 % CI 1.00–1.97) [28] but not among older men and women 70–79 (HR 1.16, 95 % CI 0.90–1.50). Another study [27] that examined the association of SB with CVD-related mortality in older men found a non-significant trend for increases in CVD mortality according to quartiles of accelerometer-derived SB, and the authors reported that the small sample size (n = 138) may have resulted in non-significant findings. No studies reported a protective effect of SB on risk of incident CVD or CVD-related mortality.

Influence of Physical Activity on Association Between SB and Cardiovascular Risk

All but two studies [27, 31••] accounted for physical activity in some way when assessing the association of SB with outcomes in older adults—six studies adjusted for physical activity level in multivariable analyses [15, 24, 26, 28, 29, 30••], one study included only older adults with high physical activity levels [23], and one study [25] presented results for SB and cardiovascular risk stratified by physical activity level. Unfortunately, with the exception of one study [30••], statistical adjustment for physical activity occurred simultaneously with other potential confounders, so the precise modifying effect of adjustment for physical activity could not be determined. Findings from a study [30••] that reported sedentary time among participants with CVD after adjustment for physical activity (as well as sex, age, and follow-up time) reported similar but slightly attenuated results compared to the model that adjusted for sex, age, and follow-up time only, suggesting that physical activity has a minimal effect on the association between SB and CVD. Gennuso et al. [25] reported null results overall for the association between SB and blood lipids and blood pressure but found in analyses stratified by physical activity level that increased SB was paradoxically associated with lower systolic blood pressure among older adults who engaged in “insufficient” (i.e., <150 min of moderate to vigorous physical activity per week), but not among those with sufficient physical activity. They also reported an inverse relationship between SB and LDL cholesterol among more sedentary older adults with sufficient physical activity, but not among those with insufficient physical activity. However, the magnitude of differences in blood pressure and LDL were very small, and analyses were not adjusted for potential confounders such as statin or antihypertensive use, so they must be interpreted with caution.

Influence of SB Measurement Method on Association Between SB and Cardiovascular Risk

No clear evidence emerged for a difference in associations between SB and cardiovascular risk according to method of SB measurement, i.e., accelerometry or self-report. Studies using either accelerometer-derived or self-reported SB reported some positive associations between SB and cardiovascular risk, and the heterogeneity of studies in terms of sample size and characteristics, outcome ascertainment, and covariates precludes comparisons across studies using different SB measurement methods. However, one study [26] did examine the association of SB with total:HDL cholesterol ratio using multiple methods of SB measurement, including total self-reported leisure-time SB, self-reported time spent watching TV, self-reported non-TV leisure-time SB, and accelerometer-derived total sedentary time. This study found that accelerometer-derived SB was a stronger predictor of higher cholesterol ratio than any self-reported SB measure.

Discussion

In this scoping review, we found preliminary evidence that older adults with CVD spend the majority (i.e., >75 %) of their waking day sedentary, that SB is more prevalent among older adults with CVD than older adults without CVD, and that SB may be a risk factor for cardiovascular health outcomes in older adults. SB was consistently associated with worse cardiac lipid profiles and increased cardiac risk scores; the associations of SB with blood pressure, CVD incidence, and cardiovascular mortality among older adults were less clear.

Our findings of increased sedentary behavior among older adults with CVD are consistent with prior evidence of higher SB in the general CVD population. Recent studies examining differences in SB between adults (of all ages) with and without CVD reported that SB was significantly higher among participants with stroke (mean SB = 10.0 h per day) [19•], angina (9.6 h per day) [18•], coronary heart disease (9.9 h per day) [18•], congestive heart failure (10.1 h per day) [18•], and myocardial infarction (9.9 h per day) [18•], compared with participants without CVD (9.2 h per day, all p’s ≤ .01). The smaller magnitude of differences in SB observed among older adults with and without CVD in this review (i.e., approximately 20–40 min) may reflect the tendency of all older adults toward greater SB [20]. However, as Fitzgerald et al. [31••] and others reported, even small increases in SB per day can produce meaningful changes in cardiovascular risk. Future work should investigate the factors that promote higher SB in older adults with cardiovascular disease and develop interventions to decrease SB in this population.

Similar to our findings, other studies have reported significant associations of SB with poor lipid profiles and non-significant associations between SB and blood pressure [34]. It appears, based on the results observed in this review and other studies, SB may have a stronger influence on blood lipids than blood pressure, although more research is needed to confirm this finding.

Our mixed findings for the association of SB with CVD incidence and CVD-related mortality, however, contrast with results of a recent empirical studies [14, 34] and reviews [12, 13] that have consistently reported significant associations between SB and cardiovascular health outcomes in the general (i.e., not restricted to elderly) population. A meta-analysis by Biswas et al. [13] reported statistically significant associations between high sedentary time and risk for incident CVD (HR 1.14, 95 % CI 1.00–1.30) and CVD-related mortality (HR 1.18, 95 % CI 1.11–1.24), which were more conservative than estimates produced in meta-analyses by Wilmot et al. who reported a relative risk for CVD incidence of 2.47 (95 % CI 1.44–4.24) and a hazard ratio for cardiovascular-related mortality of 1.90 (95 % CI 1.36–2.66) among middle-aged adults with the greatest sedentary time compared with those with the lowest sedentary time.

There are a number of potential explanations for the reduced strength of association between SB and CVD incidence and CVD-related mortality among older adults reported in this review. First, the main mechanism proposed to underlie the effects of SB on cardiovascular health, a reduction in activity of the enzyme lipoprotein lipase (LPL) [35, 36], is also influenced in aging, so that older adults may have lower LPL activity than younger adults and thus, the effect of SB on this mechanism in older adults is blunted. However, since studies of LPL activity in the context of SB have only been performed in rats, this is only a speculative hypothesis that requires empirical inquiry in humans. Another potential explanation may be that subtle differences in the association between SB and cardiovascular risk exist between young-old, middle-old, and old-old adults, as evidenced by the differential associations of SB with CVD incidence and coronary heart disease-related mortality observed among age subgroups (e.g., 60–69 vs. 70–79) of older adults in included studies [15, 28]. However, the data at this point are too scant to draw any firm conclusions about this hypothesis. Another potential reason for mixed findings, first proposed by Seguin et al. [28], is that the association between SB and CVD incidence or CVD-related death may not necessarily be weaker in older adults but may rather be attenuated due to competing risks of death brought on by advancing age. Of note, no studies included in this review performed competing risk analysis, a potentially important analytic approach when examining outcomes in older adults [37]. Lastly, low power due to small numbers of participants experiencing CVD-related events in several studies [27, 29] make firm conclusions regarding the association of SB with CVD in older adults impossible at this time.

Influence of Physical Activity on Findings

We also found preliminary evidence to suggest that moderate to vigorous physical activity may modify the association between SB and cardiovascular risk in older adults, although not in a consistent manner. Generally, moderate to vigorous physical activity is considered to be protective against the effects of SB on cardiovascular risk [16], although some studies have found this to not be the case [13]. The majority of studies included in this review considered the physical activity levels of older adults when assessing the association between SB and cardiovascular risk, and some studies found attenuated associations after controlling for physical activity [30••]. Physical activity may be an important modifying factor, but given the very low levels of moderate to vigorous physical activity observed among older adults in the USA, i.e., <10 min per day [7], we speculate that the effect of moderate to vigorous physical activity on the association between SB and cardiovascular risk in older adults may not be as important as in younger adults. While we continue to advocate for vigilance in collecting data on moderate to vigorous physical activity when assessing SB, particularly among young-old adults, we encourage a shift in attention away from focusing on the modifying influence of moderate to vigorous physical activity on the association between SB and cardiovascular risk in older adults toward an exploration of how the balance between SB and light-intensity physical activity, the types of activity older adults are most able and willing to do, impacts cardiovascular outcomes.

Limitations of This Review

The biggest limitation of this review is the small number of studies found that examined the association of SB with CVD incidence and CVD-related mortality. Also, this review was limited to studies authored in English and published in the peer-reviewed biomedical literature, which could potentially introduce publication bias.

Research Implications

We are still in the early phases of understanding the influence of SB on cardiovascular risk in older adults, and many questions remain unanswered. First, a clearer understanding of the prevalence of SB in older adults with cardiovascular disease is needed. Future studies should analyze the prevalence of SB in older adults with specific cardiovascular diseases such as myocardial infarction, angina, heart failure, and stroke, as has been done previously in samples of all-age adults [18•, 19•, 38], while accounting for other factors associated with decreased activity in older adults such a functional limitations. More studies with adequate power are needed to examine the association of SB with cardiovascular outcomes as well. When possible, SB should be measured with accelerometers, as they have been found to more accurately quantify SB in older adults than self-report [39]. Triaxial accelerometers may be especially useful in characterizing SB in older adults because they are able to differentiate sitting from standing [40], which may hold important implications for the association of SB with cardiovascular risk [41] and other health outcomes [42] in older adults.

Qualitative research should also be undertaken to examine factors associated with higher rates of SB in older adults with CVD. A prior study [43] of a general sample of older women found that pain, fatigue, and mobility issues were motivating factors for increased SB, but there may be specific risk factors for SB in older adults with CVD, such as discomfort from angina or fear of recurrent cardiac events, that have yet to be examined. Information about the determinants of SB in older adults with CVD will be helpful in directing efforts and resources toward reducing it.

There was a noticeable absence in the included articles about longitudinal changes in SB over time—does SB change over the life course, or over the course of old age, and if so, how do changes in SB over time affect cardiovascular outcomes? Furthermore, how does an intervening cardiac event (e.g., myocardial infarction, heart failure decompensation, stroke) impact SB in older adults, and in turn, how do changes in SB after a cardiac event affect risk of recurrent cardiac events or general health decline? One study [38] of adult stroke survivors reported very high SB during the year after stroke, and the authors posited that high SB may inhibit recovery after stroke, but this study was limited by lack of data on pre-stroke SB. Assessing longitudinal changes in SB, particularly changes incurred due to incident cardiovascular events, and the association of these changes with future cardiovascular risk or other health outcomes, may help identify times when older cardiac patients are vulnerable to increases in SB so that these times can be intervened upon.

Lastly, while we acknowledge that our understanding of the risk of SB on cardiovascular outcomes in older adults is incomplete at this time, the literature available to date supports a push to start thinking about how we may intervene on SB in older adults to reduce cardiovascular risk. Reducing SB may be one of the safest, most feasible, and most acceptable ways to modify activity levels in older adults, particularly among those with existing cardiovascular disease or functional impairments, or those recovering from a cardiac event for which hospitalization-associated deconditioning or safety issues preclude their ability to participate in moderate to vigorous physical activity. Preliminary studies have shown that increases in light-intensity physical activity (i.e., walking [44], standing time [41], and short breaks in SB [45]) are associated with favorable cardiovascular and functional outcomes. Future studies should consider integrating these behaviors into interventions to reduce SB in a manner that is safe, practicable, and not overly burdensome for older adults.

Conclusions

Current evidence suggests that older adults with CVD are highly sedentary and exhibit higher SB than older adults without CVD. The state of the evidence about the influence of SB on cardiovascular risk and outcomes in older adults is in its infancy and findings are mixed, but preliminary evidence suggests that SB may be a risk factor, independent of physical activity, for incident CVD and mortality in older adults. Future studies that include sufficiently large samples of older adults and use objective and validated measurements of SB are needed to clarify the association between SB and cardiovascular risk in older adults so that interventions can be developed to stem the tide of CVD in our aging population.