Association between different physical activities and longevity of elder individuals

By Pooja Toshniwal PahariaReviewed by Aimee MolineuxAug 29 2022

In a recent study published in JAMA Network Open, researchers investigated whether different leisure time physical activities (LT-PA) were differentially associated with all-cause mortality and mortality risks from cancer or cardiovascular disorders among elderly individuals.

Studies have reported a positive association between increased PA and longevity; however, data on the protective effects of comparable levels of different PA (such as cycling and running) on the life durations of elders need further investigation.

Moreover, previous studies have been conducted on young individuals and could not accurately evaluate dose-response relationships (since they were underpowered), and therefore the findings have limited generalizability for the elderly.

About the study

In the present study, researchers comparatively evaluated mortality risks among elder individuals who participated in different types but comparable levels of LT-PA. They explored the associated dose-response relationship curve shape.

In the NIH (national institutes of health)–AARP (American association of retired persons) diet and health prospective cohort study conducted for baseline data from 1995 to 1996, about 3.5 million questionnaires were sent by mail to AARP member individuals 50 years to 69 years of age who lived in Florida, California, New Jersey, Pennsylvania, Louisiana, and North Carolina, or metropolitan regions such as Atlanta, and Detroit. Baseline and follow-up study questionnaires were filled out by 567,169 and 313,363 individuals, respectively, to assess the relationship between cancer and diet in the follow-up period between 2004 and 2005.

The present study comprised 272,550 individuals who filled out the NIH-AARP follow-up questionnaires. Data were obtained on the durations of different PA types performed per week, and mortality risks were estimated through 31 December 2019. The study exposures were weekly metabolic equivalent (MET) hours spent in activities such as cycling, running, swimming, racquet sports, other aerobic exercises, walking, and golf.

Follow-up questionnaires were designed to obtain self-reported data on the average duration spent weekly in the previous year on PA, including running or jogging, cycling, swimming, other aerobic exercises, racquet sports such as squash, racquetball or tennis, golf, and walking. The prime outcomes and measures were all-cause, cancer, and cardiovascular mortality.

The study excluded 18,493 proxy baseline and follow-up participants, 3,859 participants with extreme or missing BMI (body mass index) values, 7,287 participants with walking disabilities, and 11,714 participants who did not answer for >3 of the seven LT-PA assessed in the study. Metabolic equivalents (ME) of the task values were allocated for every PA followed by multiplication with the self-reported PA durations to estimate the average MET hours/week.

Cox hazards regression modeling was used for the analysis, and the hazard ratios (HRs) were estimated with data adjustments for sex, weight, race, height, smoking habits, depression, consumption of alcohol, medical history (stroke, cardiovascular disorders, diabetes, cancer and, emphysema) and level of education.

Results

In total, 272,550 respondents were considered for the analysis, most of whom were men, 58%, n=157,415, with an average age of 71 years. Of the study cohort, 43% (n=118,153) deaths were reported from cancer (n=32,666) and cardiovascular disorders(n=38,300) within 12 years of follow-up.  The most commonly reported PA was walking (78%), followed by other aerobic exercises, cycling, golf, swimming, running, and racquet sports, as reported by 30%, 25%, 14%, 10%, 7%, and 4% of the respondents.

Compared to individuals who did not engage in PA, 7.5 weekly hours to <15 MET weekly hours of running and racquet sports were associated with the highest reductions in all-cause mortality risks with HR values of 0.9 and 0.8, respectively. Mortality risks were also lower for those who engaged in walking, other aerobic exercises, golf, swimming, and cycling with HR values of 0.9, 0.9, 0.95, and 0.97, respectively.

All-cause mortality risks were lower among moderately active (0.1 MET-hours to <7.5 MET-hours weekly) individuals and active (7.5 MET-hours to <15 MET-hours weekly) individuals compared to physically inactive individuals by five percent (HR 0.95) and 13% (HR 0.9), respectively.

In addition, highly-active individuals (≥15 weekly MET hours) had lower mortality risks; however, the magnitude of mortality risk reductions associated with higher weekly MET hours was relatively smaller. Of note, mortality risks were higher for individuals who reported very high levels (≥22.5 weekly MET-hours) of PA, such as swimming, running, and other aerobic exercises, in comparison to individuals who engaged in >15 hours but <22.5 weekly MET-hours of PA.

Mortality risks from cardiovascular disorders and cancer were lower among elders who played racquet (HR 0.7) and who engaged in running (HR 0.8), respectively. All PA activities showed a curvilinear-shaped dose-response relationship with the risk of mortality. The mortality risk reductions with running were null for females and inversely true for males with HR values of 1.1 and 0.8, respectively. Contrastingly, walking was more strongly associated with longevity for females than males.

Swimming for 7.5 to <15 weekly-MET hours for participants with BMI values below 25 was associated with mortality risk reduction (HR 0.9), whereas the association was null among individuals with BMI values ranging between 25 to 29.9 (HR 0.98) and BMI ≥30 (HR 1.1). Golf playing was associated with lower mortality risks among those with a high-school level of education (HR, 0.9). In contrast, the association was null for those who graduated from colleges (HR 0.99).

Overall, the study findings highlighted the protective role of LT-PA against mortality among the elderly with significant mortality risk reductions among those who engaged in PA of 7.5 to <15 weekly-MET hours.