Association of Daily Sitting Time and Leisure-Time Physical Activity With Survival Among US Cancer Survivors

Key Points

Question  What are the independent and joint associations of daily sitting time and physical activity with mortality outcomes among cancer survivors in the US?

Findings  This prospective cohort study included 1535 cancer survivors followed up for a median of 4.5 years and found that those with a higher sitting time had an increased risk of all-cause, cancer, and noncancer mortality. The highest risk was among those who were inactive or insufficiently active, with sitting time of more than 8 hours per day.

Meaning  The findings of this cohort study suggest that the combination of prolonged sitting and lack of physical activity, highly prevalent among US cancer survivors, is associated with heightened risks of mortality outcomes.

Importance  Sedentary behaviors, particularly prolonged sitting and lack of physical activity, may influence survival after cancer.

Objective  To examine the independent and joint associations of daily sitting time and leisure-time physical activity with mortality outcomes among cancer survivors.

Design, Setting, and Participants  A prospective cohort of a nationally representative sample of cancer survivors, age 40 years or older (n = 1535; weighted population, 14 002 666), from the US National Health and Nutrition Examination Survey from 2007 to 2014. Participants were linked to mortality data from their interview and physical examination date through December 31, 2015. Daily sitting time and leisure-time physical activity (LTPA) were self-reported using the Global Physical Activity Questionnaire. Data analyses were performed from January 1 to May 1, 2021.

Main Outcomes and Measures  All-cause, cancer-specific, and noncancer mortality.

Results  Among 1535 cancer survivors (mean [SE] age, 65.1 [0.4] years; 828 [60.1%] females; 945 [83.1%] non-Hispanic White individuals), 950 (56.8%) reported LTPA of 0 minutes per week (min/wk) during the previous week (inactive); 226 (15.6%) reported LTPA of less than 150 min/wk (insufficiently active); 359 (27.6%) reported LTPA of 150 min/wk or more (active); 553 (35.4%) reported sitting for 6 to 8 hours per day (h/d); and 328 (24.9%) reported sitting for more than 8 h/d. Of note, 574 (35.8%) cancer survivors reported no LTPA with concurrent sitting of more than 6 h/d. During the follow-up period of up to 9 years (median, 4.5 years; 6980 person-years), there were 293 deaths (cancer, 114; heart diseases, 41; other causes, 138). Multivariable models showed that being physically active was associated with lower risks of all-cause (hazard ratio [HR], 0.34; 95% CI, 0.20-0.60) and cancer-specific (HR, 0.32; 95% CI, 0.15-0.70) mortality compared with inactivity. Sitting more than 8 h/d was associated with higher risks of all-cause (HR, 1.81; 95% CI, 1.05-3.14) and cancer-specific (HR, 2.27; 95% CI, 1.08-4.79) mortality compared with those sitting less than 4 h/d. In the joint analyses, prolonged sitting was associated with an increased risk of death among cancer survivors who were not sufficiently active. Specifically, inactive and insufficiently active survivors reported sitting more than 8 h/d had the highest overall (HR, 5.38; 95% CI, 2.99-9.67) and cancer-specific (HR, 4.71; 95% CI, 1.60-13.9) mortality risks.

Conclusions and Relevance  In this cohort study of a nationally representative sample of US cancer survivors, the combination of prolonged sitting with lack of physical activity was highly prevalent and was associated with the highest risks of death from all causes and cancer.

The population of cancer survivors is rapidly growing worldwide.¹ In the US, approximately 16.9 million adults are currently living with cancer and another 1.6 million new cases are diagnosed annually.^2,3 Given the continuous advances in early detection and treatment of cancer coupled with an aging population, the number of US cancer survivors is projected to increase to 22.1 million by 2030.³ Many cancers and cancer treatments cause long-term and late effects⁴ and shorten the life expectancy.⁵ Hence, there is a pressing need to identify accessible strategies that cancer survivors can use to improve their long-term health. Three hypothesized pathways through which physical activity may improve cancer survival are: (1) a direct effect on tumor growth and metastasis, (2) improved treatment completion rates, and (3) improved treatment efficacy.⁶ Physical activity, both before and after diagnosis, has been associated with improved survival for patients with several types of cancer,⁷ and evidence indicates that the association is stronger postdiagnosis than prediagnosis.^8,9 However, levels of physical activity are critically low among cancer survivors—more than one-third of this population participates in little or no leisure-time physical activity (LTPA) while spending prolonged time sitting.¹⁰

Sedentary behavior—any waking behavior with an energy expenditure of 1.5 metabolic equivalent of task (MET) or less in a sitting, reclining, or lying posture—has become more prevalent¹¹ and is associated with a range of negative health effects.¹² Among the general population, prolonged sitting was associated with a higher risk of cardiovascular disease, diabetes,¹³ cancer,¹⁴ and overall mortality,¹⁵ particularly among individuals with low levels of physical activity.^16-18 Given the emerging evidence, the World Health Organization’s 2020 Global Guidelines on Physical Activity and Sedentary Behavior¹⁹ recommends limiting sedentary behavior and replacing it with physical activity for health benefits. However, evidence on the effects of sedentary behavior in cancer survivors is limited. Previous cohort studies found that a higher level of postdiagnosis television watching was associated with an increased risk of all-cause mortality in patients with certain cancers (eg, breast, colon, prostate).^20-23 To our knowledge, few studies have investigated the association between prolonged sitting and survival after cancer, none at the population level. In the cancer-free population, high levels of physical activity have been shown to offset the negative effect of sedentary behavior.^16-18 Cancer survivors are likely to be more sedentary compared with cancer-free populations owing to comorbidities and/or deconditioning associated with cancer and cancer treatment. However, epidemiologic evidence remains scarce on the joint association of sitting time and physical activity with survival after cancer.

This study aimed to examine the independent and joint associations of total sitting time and LTPA with all-cause, cancer-specific, and noncancer mortality among a US nationally representative sample of cancer survivors. The knowledge generated may be critical to the development of quantitative guidelines on limiting sedentary time among cancer survivors.

This prospective cohort study used a nationally representative sample from the US National Center for Health Statistics’ National Health and Nutrition Examination Survey (NHANES), which has been conducted on 2-year cycles since 1999 to monitor the health and nutritional status of the US population. All the NHANES protocols were approved by the National Center for Health Statistics ethics review board, and written informed consent was obtained from all participants. This modeling investigation was exempt from review because it used published deidentified data sets that included no personally identifiable information. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement guidelines.

Each participant was invited to complete an in-person interview and undergo a set of physical examinations and laboratory tests in a mobile examination center. This study examined and analyzed data on sociodemographic characteristics, lifestyle factors, and medical history in adults 40 years or older with data available on daily sitting time and physical activity for 4 cycles of NHANES from 2007 to 2014.

Data on cancer diagnosis and type were collected during the in-person interview, including cancer type(s), with up to 3 diagnoses recorded, and age at each diagnosis. Participants were asked, “Have you ever been told by a doctor or other health professional that you had cancer or a malignancy of any kind?” Individuals who responded yes were defined as cancer survivors and were asked, “What kind of cancer was it?” and “How old were you when this cancer was first diagnosed?”²⁴ The years since cancer diagnosis were calculated as the difference between the participant’s current age and their age at first cancer diagnosis. Cancer types were further classified into obesity-related and non−obesity-related cancers because previous studies on physical activity and sedentary behavior have mainly focused on obesity-related cancers.²⁰ Cancers of the blood, brain, breast, colon, and rectum, esophagus, gallbladder, kidney, liver, ovary, pancreas, stomach, and uterus were classified as obesity-related; other cancer types were classified as non−obesity-related cancers.²⁵

Total daily sitting time and LTPA were self-reported by participants responding to the Global Physical Activity Questionnaire (GPAQ). The GPAQ has been previously validated to collect information on daily physical activity, LTPA, and sedentary behavior.²⁶ During the in-person interview, participants were asked to report their moderate- and vigorous-intensity recreational activities during a typical week. The total amount of LTPA was estimated as minutes of moderate-intensity recreational activities plus twice the minutes of vigorous-intensity recreational activities.^27,28 Based on the 2018 Physical Activity Guidelines for Americans, participants without any LTPA, with LTPA more than 0 minutes per week (min/wk) but less than 150 min/wk, and with LTPA of 150 min/wk or more during the previous week were classified as inactive, insufficiently active, and sufficiently active, respectively.^27,28 Furthermore, participants were asked, “On a typical day, how much time do you usually spend sitting at school, at home, getting to and from places, or with friends, including time spent sitting at a desk, traveling in a car or bus, reading, playing cards, watching television, or using a computer?” In accordance with recent studies,^16,18 responses were converted to hours per day (h/d) and categorized into 4 groups (0 to <4, 4 to <6, 6 to 8, and ≥8 h/d).

The NCHS provided mortality data that were linked to the National Death Index through December 31, 2015. The International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) was used to record the underlying cause of death. Cancer mortality was classified as death caused by malignant neoplasms (ICD-10, codes C00-C97). The duration of follow-up was defined as the interval in months from the interview date to the date of death or through December 31, 2015, for those participants who did not experience an event.²⁹ Data analyses were performed from January 1 to May 1, 2021.

Self-reported sociodemographic characteristics included sex (male or female), race and ethnicity (non-Hispanic Black, Hispanic, non-Hispanic White, other [American Indian/Native Alaskan/Pacific Islander, Asian, multiracial]), educational attainment (<high school, high school,>high school), and family poverty income ratio (total family income divided by the poverty threshold; <1.3, 1.3 to ≤3.5, ≥3.5).^11,29 Participants’ weight and height were measured during the physical examination and body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) was categorized into 3 groups (<25, 25.0-29.9, ≥30). Lifestyle factors included smoking status (never, former, current), alcohol use (never, former, current, unknown), and the Healthy Eating Index-2015³⁰ (HEI-2015; derived from the participant’s 24-h dietary recall interview). The HEI-2015 quantifies the quality of an individual’s overall diet with a score from 0 to 100 (worst-best).

Hypertension was either self-reported by participants who had received a diagnosis from a health professional or determined by NHANES-measured blood pressure (≥130 mm Hg [systolic] or ≥80 mm Hg [diastolic]). Hypercholesterolemia was self-reported by participants who had received a diagnosis from a health professional or determined by NHANES-measured total cholesterol level (≥240 mg/dL; to convert to millimoles per liter, multiply by 0.0259). A history of cardiovascular disease (CVD) and/or diabetes was self-reported by participants who had received either or both of these diagnoses from a health professional or determined by a prescription history for medications used to treat these conditions.²⁹

All analyses were conducted following the NHANES analytic guidelines and accounted for the unequal probability of selection, oversampling of certain subpopulations, and nonresponse adjustments to ensure nationally representative estimates. Statistical tests were 2-sided and statistical significance was set at P < .05. Data analyses were performed from January 1 to May 1, 2021, using Stata, version 16.0 (StataCorp LLC).

Sample sizes and weighted percentages were estimated according to the participants’ sociodemographic and lifestyle factors and their daily sitting time categories. Multivariable Cox proportional hazards regression models were applied to estimate hazard ratios (HR) and 95% CI for the associations of daily sitting time and LTPA with overall, cancer-specific, and noncancer mortality, respectively. Final stage multivariable models were adjusted for age, sex, race and ethnicity, educational attainment, family poverty income ratio, BMI, smoking status, alcohol use, HEI-2015 score, and health factors (ie, hypertension, hypercholesterolemia, history of diabetes and/or CVD, and years since the first cancer diagnosis). To examine joint associations, participants were classified by sitting time and LTPA level to estimate mortality risks using multivariable Cox proportional hazards regression models adjusting for the same set of covariates. To analyze cancer-specific and noncancer mortality, sitting categories were collapsed (<6 and ≥6 h/d) owing to small case numbers for each outcome. All analyses were conducted in overall, female, male, and survivors of obesity-related and non−obesity-related cancers, respectively. Finally, sensitivity analyses were conducted by excluding deaths that occurred during the first 2-year follow-up to lessen the probability of reverse causation.³¹

Of the 1535 cancer survivors (weighted population, 14 002 666; weighted mean [SE] age, 65.1 [0.4] years; 60.1% female) in the study cohort, 300 (7.8%) were non-Hispanic Black, 213 (5.5) were Hispanic, 945 (83.1%) were non-Hispanic White, and 77 (3.6%) were individuals of “other” race or ethnicity, including American Indian/Native Alaskan/Pacific Islander, Asian, and multiracial (Table 1). The number of cancer survivors, by cancer type and sex, is presented in eTable 1 in the Supplement. Only 27.6% of the cancer survivors were physically active (LTPA ≥150 min/wk), while 56.8% reported no LTPA during the previous week. Most of the cancer survivors (60.3%) reported sitting for more than 6 h/d (Figure 1; eTable 2 in the Supplement). Importantly, 35.8% of cancer survivors reported no LTPA and sitting for more than 6 h/d. Female survivors and those with a higher socioeconomic status or BMI spent more time sitting compared with their counterparts (Table 1).

During the follow-up period (median) of up to 9.0 (4.5) years or 6980 person-years, 293 deaths occurred; 114 patients died of cancer, 41 of heart disease, and 138 of other causes. Cancer survivors with higher sitting time had increased all-cause, cancer-specific, and noncancer mortality risks (Table 2). After adjusting for covariates and LTPA, HRs for all-cause, cancer-specific, and noncancer mortality among individuals sitting more than 8 h/d compared with those sitting less than 4 h/d were 1.81 (95% CI, 1.05-3.14), 2.27 (95% CI, 1.08-4.79), and 1.54 (95% CI, 0.72-3.31), respectively. Each 1-h increase in daily sitting was associated with 7%, 9%, and 5% increased risks of death from all-cause, cancer, and noncancer, respectively (eFigure in the Supplement). Meanwhile, physically active survivors had a lower risk of all-cause (HR, 0.34; 95% CI, 0.20-0.60), cancer (HR, 0.32; 95% CI, 0.15-0.70), and noncancer (HR, 0.36; 95% CI, 0.19-0.68) mortality than inactive survivors. These associations were consistent between female and male survivors and those with a diagnosis of obesity-related and non−obesity-related cancer (eTable 3 in the Supplement).

In the stratified analysis by LTPA levels, sitting time was not associated with survival after cancer among physically active survivors. Among those who were insufficiently active or inactive, longer sitting time was associated with elevated risks of all-cause (HR per 1-h increase, 1.08 [95% CI, 1.04-1.12]) and cancer (HR per 1-h increase, 1.09 [95% CI, 1.04-1.15]) mortality in a dose-response manner (eTable 4 and eFigure in the Supplement). In joint analyses, combinations of higher sitting time and lower LTPA were deleteriously associated with overall, cancer, and noncancer mortality risks (Figure 2). Specifically, inactive or insufficiently active cancer survivors sitting 8 or more h/d were 5.38 (95% CI, 2.99-9.67) times more likely to die of all-cause mortality compared with those who were active (sitting <6 h/d; Table 3). Similar associations were observed for cancer and noncancer deaths. These results were similar by sex and remained consistent among survivors with diagnoses of obesity-related and non−obesity-related cancers (eTables 4 and 5 in the Supplement). All results remained similar in sensitivity analyses excluding deaths that occurred during the first 2-year follow-up (data not shown).

In this US nationally representative cohort of cancer survivors, more than half reported no LTPA and two-thirds reported sitting more than 6 h/d. Altogether, about one-third of cancer survivors reported no LTPA with concurrently sitting more than 6 h/d. During as many as 9.0 years of follow-up, cancer survivors with a higher sitting time had increased all-cause, cancer, and noncancer mortality risks, independent of LTPA levels. In the joint analyses, the association of sitting with increased mortality was only observed among cancer survivors who were insufficiently active or inactive but not among those who were sufficiently active. Of note, insufficiently active or inactive cancer survivors with the longest sitting times (>8 h/d) had as much as a 5-fold increase in the risk of death from all causes, cancer, and noncancer.

To our knowledge, this is the first study to prospectively investigate the independent and joint associations of sitting time and LTPA with mortality outcomes in a nationally representative sample of cancer survivors. Given its high prevalence before and after cancer, sitting-related sedentary behavior is an important research target in cancer survivors. Previous studies frequently investigated specific sedentary behavior with little consideration of its joint association with physical activity. In 2020, Swain and colleagues²⁰ summarized data from 9 studies and concluded that a higher level of postdiagnosis television watching was associated with a mild but statistically significant higher risk (HR, 1.22; 95% CI, 1.06-1.41) of all-cause mortality in survivors of certain cancers (eg, breast, colon, prostrate). These important findings suggest a causal relationship between sitting and survival after cancer but are insufficient to make specific recommendations for cancer survivors.

The joint association of sitting and physical activity has been explored in cancer-free populations. A harmonized meta-analysis summarized evidence from 13 studies of more than 1 million adults¹⁶ and found that prolonged sitting (ie, >4 h/d) was associated with increased all-cause mortality among adults who engaged in less than 30 MET h/wk of physical activity (approximately 400 min/wk of LTPA). In that meta-analysis, the least active group (sitting >8 h/d and physical activity <2.5 MET h/wk [approximately 30 min/wk of LTPA]) had a 59% higher risk of death compared with the most active group (sitting <4 h/d and physical activity >35.5 MET h/wk). A more recent cohort study of 149 077 Australian adults ≥45 years, with a median follow-up of 8.9 years, reported that longer sitting time was associated with a higher risk of all-cause mortality in the least active group (<150 LTPA min/wk).¹⁸ To our knowledge, this is the first time that a study has indicated a strong dose-response association of prolonged sitting and mortality among cancer survivors reporting less than 150 LTPA min/wk. Most importantly, the least active cancer survivors (sitting ≥8 h/d plus LTPA <150 min/wk) experienced approximately 5.5 times the risk of death compared with those survivors who were most active (sitting <4 h/d plus LTPA ≥150 min/wk).

In its 2018 Physical Activity Guidelines for Americans,²⁸ the US Department of Health and Human Services for the first time clearly acknowledged the health risks associated with sedentary behavior and suggested that that people of all ages would benefit from more regular physical activity and less sedentary behavior. Likewise, and despite the limited evidence available, the World Health Organization in its 2020 Global Guidelines on Physical Activity and Sedentary Behavior strongly recommends limiting the amount of sedentary time and substituting it with any physical activity to improve health, particularly for individuals with long-term diseases.¹⁹ However, the low certainty of the available evidence does not support quantitative guidelines on the appropriate limit for sedentary behavior among cancer survivors. In 2019, a roundtable report from the American College of Sports Medicine⁸ identified the lack of epidemiologic data connecting sedentary behavior with cancer survival outcomes as a key knowledge gap. The present analysis addresses previously identified knowledge gaps in the literature on cancer survival and provides direct evidence of the association of prolonged sitting with worse survival after cancer, particularly among those who were inactive. These findings suggest that sufficient physical activity can attenuate the excess mortality risk from prolonged sitting. Notably, this analysis was conducted with a large and nationally representative cohort of US cancer survivors, which included cancer types that have comparatively good 5-year survival rates (eg, breast, colorectal, prostate) as well as those with poorer prognoses (eg, ovarian, pancreatic).³²

Several behavioral and biologic pathways could explain the observed associations. Many cancer survivors experience dramatic life changes that devastatingly affect their physical and psychological health. These impairments may contribute to a sedentary lifestyle and lead to shortened survival.³³ Also, hypothesized biologic mechanisms could effect metabolic and sex hormones, inflammation, and immunity as the main pathways.⁸ The growing evidence from experimental studies indicates that prolonged, uninterrupted sitting are associated with impaired glucose metabolism and increased systemic inflammation;³⁴ however, these can be attenuated by breaking up prolonged sitting.³⁵ Moreover, sitting may induce exercise resistance and diminish the benefits of physical activity,³⁶ which could potentially explain the worse survival outcomes observed among cancer survivors who reported no LTPA with concurrent prolonged sitting. Low LTPA and prolonged sitting may also contribute to increased risk of cachexia, thromboembolic events, and fatigue, which all influence survival after cancer.³⁷ Additional studies are needed to evaluate whether the observed associations are causal, and if so, to develop theory-based interventions that can be incorporated into cancer survivorship care to reduce sedentary behavior and increase physical activity.^6,38

The clear strength of this study was using a nationally representative sample of US cancer survivors, which allows the findings to be generalized at the population level. Additionally, a wide range of potential confounding factors were adjusted in the analyses. Several limitations also should be considered. First, sitting time and LTPA data were self-reported and not device-based. Wearable devices are frequently used to directly measure physical activity and monitor sedentary time; however, except for thigh-worn monitors, most devices cannot distinguish between prolonged sitting and sedentary behavior (lying down or reclining) and would have likely misclassified stationary behaviors (eg, standing) as sedentary time.³⁹ Second, sitting time and LTPA were assessed at baseline, which may not reflect behavioral changes during the follow-up period. Studies employing direct and repeated measures are needed to evaluate the longitudinal effects of sitting and LTPA on survival after cancer. Lastly, NHANES did not collect data on cancer stages and treatments; however, consistent results were observed after excluding deaths occurring during the first 2-year follow-up period, which lessens the probability of reverse causation.³¹

In this prospective cohort study of a nationally representative sample of US cancer survivors, the combination of prolonged sitting with lack of physical activity was highly prevalent and was associated with increased all-cause and cancer mortality risks. This study has provided new evidence that sedentary behavior and physical activity should be jointly considered in future observational, epidemiologic, and intervention studies. Future research may further elucidate these associations and provide more evidence that quantifies the timing, dosage, limits, and domains of sedentary behavior and physical activity to increase cancer survival.

Accepted for Publication: October 8, 2021.

Published Online: January 6, 2022. doi:10.1001/jamaoncol.2021.6590

Corresponding Author: Lin Yang, PhD, Department of Cancer Epidemiology and Prevention Research, Cancer Research and Analytics, Cancer Care Alberta, Alberta Health Services, 2210 Second St SW, Holy Cross Centre, 5th Floor, Box ACB, Calgary, AB T2S 3C3, Canada (lin.yang@albertahealthservices.ca).

Author Contributions: Mr Cao and Dr Yang had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: Cao, Yang.

Drafting of the manuscript: Cao, Yang.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Cao, Yang.

Administrative, technical, or material support: Yang.

Supervision: Friedenreich, Yang.

Conflict of Interest Disclosures: None reported.