Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jul 11.
Published in final edited form as: J Appl Gerontol. 2025 Jan 11;44(10):1691–1700. doi: 10.1177/07334648251313883

Do physical activities prevent the occurrence of bothersome pain?

Bian Liu 1,2,3,*, Chen Yang 1,4,*, Madhu Mazumdar 1,3,4, Melissa Aldridge 5, Sean Morrison 5, Lihua Li 1,3,4,5
PMCID: PMC12246169  NIHMSID: NIHMS2046008  PMID: 39797781

Abstract

We examined the associations between physical activity (PA)—measured through self-reported walking and vigorous activities—and pain occurrence (self-reported bothersome pain or frequent pain medication use), and persistent pain (pain occurring for two consecutive years). This analysis used a large, nationally representative sample of 2,279 older adults from the National Health and Aging Trends Study of 2015–2018, and applied generalized estimating equation regression with propensity score weighting. Approximately 70% and 50% of the participants reported walking and vigorous activities respectively at baseline. The cumulative incidence over the 3-year follow-up time was approximately 60%, and was similar across PA groups. The risk of pain occurrence or persistent pain did not differ by walking or by vigorous activity status (relative risks ranged from 0.97 to 1.20, and the 95% CIs included one). While pain occurrence was common among older adults, our analysis did not find it to be associated with PA.

Keywords: pain occurrence, physical activity, propensity score weighting, pain management, National Health and Aging Trends Study

Introduction

Pain is one of the most reported chronic conditions among older adults (Ettinger et al., 1994; Freedman et al., 2011; Turner et al., 2004). A large study using data from the National Health Aging Trends Study (NHATS) on Medicare beneficiaries found that over half (52.9%, 18.7 million) of older adults in the United States (US) reported bothersome pain in 2011 (Patel et al., 2013). Worryingly, a 5-year follow-up analysis of the same survey data suggests nearly 35% will experience persistently high bothersome pain (Rundell et al., 2020). Using the 2019–2021 National Health Interview Survey (NHIS) data, another nationally representative survey of the US population, it was found that the prevalence of chronic pain was 30.0% among older adults aged 65–84 years and 34.3% among those aged 85 and above (Rikard et al., 2023). Additionally, those with disabilities were more likely to experience chronic pain (58.0%) regardless of age (Rikard et al., 2023).

Pain management in older adults remains a critical public health issue. It directly impacts their well-being and ability to participate socially, making it a concern not only for seniors themselves but also for physicians across all specialties (Ritchie et al., 2023). For optimal pain prevention, treatment, and management, a multidisciplinary approach is often recommended. This may involve effective pharmacologic options including analgesics such as acetaminophen, nonsteroidal anti-inflammatory drugs and opioids, as well as nonpharmacologic approaches, such as exercise and physical activity (PA) (Ambrose & Golightly, 2015; Domenichiello & Ramsden, 2019; Geneen et al., 2017; Jones et al., 2016).

With growing evidence showing that PA can be an effective non-pharmacological alternative to treat pain and reduce reliance on long-term medications, PA has been increasingly advocated, used, and studied in chronic pain management (Ambrose & Golightly, 2015; Antcliff et al., 2021; Arnes et al., 2023; Bidonde et al., 2023; Fanning et al., 2023; Geneen et al., 2017; Skou et al., 2018; Stubbs et al., 2013; Vaegter et al., 2024). However, observational study-based research shows mixed results. While some have found PA to reduce pain in specific domains or sites, such as musculoskeletal pain and low back pain (Alzahrani et al., 2019; Heneweer et al., 2009; Naugle et al., 2016; Niederstrasser & Attridge, 2022), others have not found a significant effect (Ohlman et al., 2018). The discrepancies can be in part explained by differences in the study designs, sample sizes, and measures of both pain and PA. Small sample sizes of some cohort studies and the under-representation of older adults in clinical trials can also compromise the strength of the evidence regarding the effect of PA on pain management (Ferretti et al., 2019; Geneen et al., 2017; Merriwether et al., 2018). In addition, most examinations into the relationship between PA and pain have been cross-sectional rather than longitudinal (Ferretti et al., 2019; Naugle & Riley, 2014; Santos et al., 2018). Moreover, few studies have assessed the impact of PA on pain by specific PA characteristics, such as the type and intensity (Baumbach et al., 2022; Geneen et al., 2017). Overall, the evidence about the benefits and risks of PA among older adults is lacking (Domenichiello & Ramsden, 2019). Given these limitations, large population-based surveys with follow-up information present an innovative opportunity to study the relationship between PA and pain.

We utilized data from NHATS of 2015–2018 and applied a rigorous propensity score weighting (PSW) approach incorporating survey design elements to examine the associations between the risk of two types of pain measures (“bothersome pain” and “persistent pain”) and two types of PA measures (walking for exercise and engaging in vigorous activities). Findings may shed light on the impact of PA on pain development that can be generalizable to the US older adult population.

Methods

Data source and study population

This retrospective study used the publicly available data from NHATS, which is a large, ongoing survey that tracks a representative sample of Medicare beneficiaries aged 65 and older (Montaquila et al.). NHATS began in 2011 and collects data through annual, in-person interviews. With a stratified, multistage sampling design, NHATS allows researchers to look at both national trends in late-life functioning as well as assess individual trajectories as the population ages (Montaquila et al.). We specifically used data from Rounds 5–8 (2015–2018) to build our cohort and followed participants for three years. Details on cohort construction are provided in Supplementary Figures S1. Briefly, to focus on the occurrence of pain, we included only community-dwelling participants who reported no bothersome pain and no pain medication use at baseline in 2015 (n=2,768). We further excluded participants with missing data on the pain outcomes, and covariate variables during the first follow-up year (n=489). This resulted in a final cohort size of 2,279 participants. Compared to the excluded participants, the included participants were more likely to be non-Hispanic White, have high socio-economic status and better health condition (Table S1). The study, using identified, publicly available data from NHATS, was exempt human research determined by the institution review board of the Mount Sinai Hospital (IRB number: STUDY-23–00473).

Outcome Variables

This study assessed two pain outcomes: bothersome pain and persistent pain.

Bothersome pain:

We defined bothersome pain based on participants’ responses to two questions: “In the last month, have you been bothered by pain?” and “In the last month, how often did you take medication for pain?”. Participants who answered “Yes” to the bothersome pain or reported using pain medication “everyday”, “most days”, or “some days” were classified as having a bothersome pain. Conversely, those who answered “No” to bothersome pain question and reported using pain medication “rarely” or “never” were classified as not having bothersome pain. Since no participants had bothersome pain at baseline, we were able to assess pain occurrence using yearly incidence (i.e., bothersome pain occurrences during each follow-up year), as well as cumulative incidence (i.e., the proportion of total bothersome pain occurrences over a specified time, regardless of when the participants experienced bothersome pain during the study period).

Persistent Pain:

To distinguish chronic pain from the initial occurrence of bothersome pain, we further defined persistent pain as experiencing bothersome pain or taking pain medication for the first two consecutive years (2015–2017).

Exposure variables

We assessed physical activity using two self-reported measures: walking for exercise and engaging in vigorous activities, representing a relatively low and high impact PA intensities, respectively. Participants answered “yes” or “no” to the following questions: “In the last month, did you ever go walking for exercise?” (Walking for exercise) and “In the last month, did you ever spend time on vigorous activities that increased your heart rate and made you breathe harder?” (Engaging in vigorous activities). These vigorous activities include working out, swimming, running/biking, or playing a sport. By dichotomizing the responses (“Yes” vs. “No”) to each of these two questions, we categorized participants correspondingly for the two PA measures.

Covariates

To understand the characteristics of the study participants, we examined individual-level sociodemographic and clinical characteristics across six domains. Sociodemographic factors included age at interview (in years), sex (male/female), self-reported race and ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, and other race/ethnicity), marital status (Married/Not married), education level (below high school, high school/general education development (GED) equivalent, some college, and bachelor’s degree or higher), and income quartiles. We also assessed mental health by measuring anxiety and depression, where anxiety and depression was defined as score of 3 and above in Generalized Anxiety Disorder 2-item (GAD-2) (Kroenke et al., 2007) and Patient Health Questionnaire 2-item (PHQ-2) (Arroll et al., 2010), respectively. Physical health domain included arthritis (yes/no), hip fracture (yes/no), as well as the count of other 7 common chronic diseases (heart attack, heart disease, osteoporosis, diabetes, lung disease, stroke, and cancer) categorized into three groups (0, 1–2, and ≥3). The physical function domain included functional limitation and physical capacity. Functional limitation was defined as having difficulties in any of the three categories: activities of daily living (ADL, including eating, bathing, toileting, and dressing), instrumental activities of daily living (IADL, including laundry, grocery shopping, meal preparation, banking or paying bills, and medication tracking), and mobility (including going outside, moving around inside, and getting out of bed) (Shirazi et al., 2023). Physical capacity assessed participants’ abilities to carry out six nested pairs of tasks (walking 3 and 6 blocks; going up 10 and 20 stairs; carrying 10 and 20 pounds; bending over and kneeling down; reaching overhead and putting a heavy book on a shelf overhead; grasping small objects and opening a sealed jar) (Ji & Xiang, 2023). For each task, participants were evaluated by a score of 0, 1 or 2 with a higher value indicating being able to do more challenging tasks, and the average score over all six tasks was used. Finally, the healthcare services utilization domain included having a hospital stay in the past month (yes/no), experiencing a flare-up of pain in either hand or wrist currently (yes/no), having a surgery or serious injury on arm/shoulder in the last 3 months (yes/no), having falls in the last 12 months (yes/no), and ever receiving rehabilitation (yes/no).

Statistical Analysis

All analyses accounted for the complex survey design of NHATS (strata, clusters, and survey weights). Baseline characteristics of the participants were compared by the dichotomized PA status. Categorical variables like sex and race were analyzed using Rao-Scott Chi-square tests, while age (continuous) was compared using survey-weighted two sample t-tests. Pain outcomes were reported as incidence rate and cumulative incidence stratified by PA status.

To create balance between groups based on PA status, we used propensity score weighting (PSW) analysis using the pooled logistic regression method (Linden & Adams, 2010). We derived the propensity score by modeling the probability of walking (yes/no) and vigorous activities (yes/no) using all aforementioned covariates for current and past years. Specifically, the propensity scores for the first year (2015) were derived based on the covariates of the first year (2015); the propensity scores for the second year (2016) were derived based on the covariates of both years as well as the outcomes of the first year; the propensity scores for the third year (2017) were derived based on the covariates of all three years and outcomes of the first two years. To examine the impact of PA on pain, we then employed a generalized estimating equation (GEE) regression weighted by a “composite weight”, i.e., the product of the propensity score based weights and survey weights, as recommended by others (Cook et al., 2009; Yang et al., 2023). The GEE regression was used to account for the repeated measures of both the exposures and outcome, since behavior of exercising and pain occurrence may change over time. As a robustness check, we also conducted sensitivity analyses using multivariable GEE regressions adjusting for the aforementioned covariates.

All analyses were performed using SAS software (SAS Institute, version 9.4). We reported relative risk (RR) with 95% confidence intervals (CIs) generated by SAS NLMeans macro.(SAS Inc.) All tests were two sided, with a p-value ≤ 0.05 considered statistically significant.

Results

Baseline characteristics of study participants

Table 1 provides the baseline study participant characteristics by the two PA measures. At baseline in 2015, the weighted proportion of participants who reported having walked and performed vigorous activities in the past month was 70.7% and 50.1%, respectively. The weighted proportion of participant who reported having both walking and vigorous activities, walking only or vigorous activity only was 42.2%, 28.5% and 7.9% respectively. The crosstab pattern was relatively stable over time (Table S2). Compared to their inactive counterparts, those who walked for exercise or had vigorous activities tended to be younger (mean age 76 vs 73 years, p<0.001), have higher education (e.g., >40% vs <25% for having a Bachelor’s degree, p<0.001), have higher income (e.g., >43% vs <26% for the top quartile income, p<0.001), and have fewer comorbidities and better health and functional status.

Table 1.

Participant characteristics at baseline by physical activity measures, NHATS 2015–2018

Walking Vigorous Activities
Yes No P value Yes No P value
Unweighted, N (weighted %) 1,516 (70.7) 763 (29.3) <0.001 981 (50.1) 1,298 (49.9) 0.917
Age at interview, mean (SD) 73.69 (6.8) 76.64 (7.7) <0.001 72.85 (6.2) 76.27 (7.7) <0.001
Sex Female 749 (47.6) 430 (52.9) 0.056 414 (42.7) 765 (55.7) <0.001
Race and Ethnicity
 White, non-Hispanic 1,103 (81.5) 544 (82.0) 0.267 769 (85.8) 878 (77.5) <0.001
 Black, non-Hispanic 278 (7.0) 158 (8.0) 146 (5.6) 290 (9.0)
 Other a 43 (3.7) 13 (2.2) 25 (3.4) 31 (3.1)
 Hispanic 92 (7.8) 48 (7.9) 41 (5.2) 99 (10.4)
Married 825 (62.5) 337 (51.5) <0.001 595 (66.0) 567 (52.5) <0.001
Education
 < High School 246 (12.1) 173 (18.9) <0.001 107 (8.3) 312 (19.9) <0.001
 High School/GED 327 (20.5) 252 (33.0) 200 (19.1) 379 (29.2)
 Some College 391 (27.4) 202 (28.8) 269 (29.1) 324 (26.4)
 ≥Bachelors 552 (40.1) 136 (19.3) 405 (43.4) 283 (24.5)
Income quartile
 Quartile 1 (Low income) 273 (13.7) 196 (22.0) <0.001 119 (9.3) 350 (23.0) <0.001
 Quartile 2 311 (15.6) 225 (26.7) 179 (14.8) 357 (22.9)
 Quartile 3 403 (27.5) 190 (26.1) 266 (26.0) 327 (28.2)
 Quartile 4 (High income) 529 (43.2) 152 (25.2) 417 (49.9) 264 (25.9)
Anxiety 54 (3.5) 54 (7.5) 0.003 33 (3.6) 75 (5.8) 0.073
Depressed 72 (4.2) 89 (9.7) <0.001 40 (3.5) 121 (8.2) <0.001
Arthritis 580 (33.2) 353 (42.2) 0.001 362 (32.7) 571 (39.1) 0.008
Hip Fracture (since age 50) 49 (2.5) 41 (4.5) 0.056 21 (1.9) 69 (4.3) 0.004
Counts of other self-reported chronic diseases
 0 269 (21.7) 84 (13.1) <0.001 192 (23.9) 161 (14.4) <0.001
 1 – 2 873 (58.7) 419 (56.6) 575 (58.5) 717 (57.7)
 3 + 374 (19.5) 260 (30.3) 214 (17.6) 420 (27.8)
Hospital stays in the last 12 months 212 (12.1) 153 (19.0) 0.001 113 (9.8) 252 (18.5) <0.001
Flare up pain hand/surgery arm or shoulder 102 (6.6) 94 (12.3) <0.001 72 (7.3) 124 (9.2) 0.136
Received rehab in the last 12 months 188 (12.1) 105 (12.7) 0.695 119 (12.3) 174 (12.3) 0.989
Functional limitationb (ADL, IADL, mobility) 311 (17.1) 271 (30.4) <0.001 171 (15.0) 411 (27.0) <0.001
Physical capacity c, mean (SD) 1.85 (0.3) 1.55 (0.6) <0.001 1.90 (0.2) 1.62 (0.5) <0.001
Open in a new tab

Note: GED, general education development; ADL, activities of daily living; IADL instrumental activities of daily living

a

Other includes Am Indian/Asian/Native Hawaiian/Pacific Islander/other specify, non-Hispanic.

b

Self-care activities (activities of daily living) were defined as eating, bathing, toileting, and dressing; household activities (instrumental activities of daily living) included laundry, grocery shopping, meal preparation, banking or paying bills, and medication tracking; and mobility activities were defined as going outside, moving around inside, and getting out of bed. For each activity, participants were assigned to one of three categories: having no difficulty, having some difficulty but able to perform activity without assistance, or unable to do and requiring assistance. For each category (self-care, household, and mobility), the presence of functional limitation was formulated as a dichotomous variable, with no functional limitation (e.g., no difficulty) coded as ‘no limitation’ and any functional limitation coded as having a ‘limitation.’ An overall value for functional limitation was additionally formulated as a dichotomous variable, with an overall limitation present if the participant had a limitation in any of three categories (ADL, IADL, or mobility).”(Shirazi et al., 2023)

c

A composite score of physical capacity was created by averaging the scores across the six pairs of tasks: (a) walking 3 and 6 blocks, (b) going up 10 and 20 stairs, (c) carrying 10 and 20 pounds, (d) bending over and kneeling down, (e) reaching overhead and putting a heavy book on a shelf overhead, (f) grasping small objects and opening a sealed jar. The responses for each pair of tasks were scored as (0) unable to do either, (1) able to do the less challenging task but not more challenging, and (2) able to do the more challenging task. The final score was set to missing if all items had missing values. The final score ranges from 0 to 2, with increasing scores indicating better physical capacity.(Ji & Xiang, 2023)

Yearly and cumulative incidence of bothersome pain

Overall, the yearly incidences of bothersome pain were 34.7%, 25.9%, and 18.0% for the first (2015–2016), second (2016–2017), and third (2017–2018) follow-up year, respectively. This yearly incidence included 5.5%, 5.8 %, and 4.1% of participants who were on pain medication while not reporting bothersome pain (Table S3). The incidence rate was, in general, slightly higher among the inactive groups, though the differences were not statistically significant. The cumulative incidence over the 3-year follow-up from 2015 to 2018 was 59.6% (Figure 1). The cumulative incidence was slightly higher among those who reported walking (61.2% vs 59%, p=0.50) or having vigorous activities (61.3% vs 58.0%, p=0.21) than their inactive counterparts, though the difference was not significant (Figure 1).

Figure 1.

Figure 1.

Open in a new tab

Yearly Incidence and cumulative incidence of bothersome pain, overall and by physical activity status.

Associations between PA measures and bothersome pain

The proportion of bothersome pain for those who had walking for exercise in the following 3 years was 34.5%, 34.7%, 35.4% compared to 35.1%, 41.1%, and 36.5% for those who had no walking in the last month (p=0.822, 0.002, and 0.743, respectively); 33.5%, 33.8%, and 33.6% of those who had vigorous activities in the last month reported bothersome pain in the following 3 years, compared to 35.8%, 38.9%, and 37.8% for those who had no vigorous activities (p=0.263, 0.046, and 0.140, respectively) (Figure 2).

Figure 2.

Figure 2.

Open in a new tab

Associations between physical activity (PA) and bothersome pain.

Results from the GEE analysis with PSW showed that neither walking (RR=1.02 (95% CI: 0.91–1.13) nor having vigorous activities (RR=0.99 (95% CI: 0.90–1.08) was significantly associated with the risk of developing bothersome pain (Figure 2). Similar non-significant results were found from traditional multivariable GEE regression models for the associations between walking (RR=1.02 (95%CI: 0.91–1.13)) and vigorous activities (RR=0.99 (95%CI: 0.90–1.08) with the occurrence of bothersome pain (Figure 2).

Associations between PA measures and persistent pain

Results from the analyses using reported bothersome pain for two consecutive years as a measure of persistent bothersome pain are shown in Figure 3. The incidence of having persistent pain from 2015 to 2017 was 19.0% vs 19.9% for those with and without walking, and the difference was not statistically significant (p=0.731). The incidence of persistent pain was 17.8% vs 20.8% for those with and without vigorous activities (p=0.117). Results from the propensity score weighted GEE regression showed that neither walking (RR=1.20 (95% CI: 0.84–1.56)) nor vigorous activities (RR=0.97 (95% CI: 0.73–1.21)) was significantly associated with persistent pain. Similar results were found using traditional multivariable regression models (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

Associations between physical activity (PA) and persistent pain.

Discussion

Utilizing robust statistical methods on a nationally representative sample of Medicare-enrolled older adults, our study provided important information about the descriptive population-level statistics on pain occurrence and PA, as well as their associations. Our analysis revealed that a significant proportion of older Americans experienced new bothersome pain, with approximately 60% of older adults developing bothersome pain within 3 years, and the speed of bothersome pain occurrence ranged from 18 to 35 new cases per 100 person-years. Furthermore, our data suggests that persistent pain affects approximately 20% of this population. The magnitude of the occurrence of bothersome pain suggests a substantial increase in pain and potentially bothersome pain related burden in older adults in a relatively short time frame. This makes pain management particularly important.

Interestingly, despite the baseline assessment indicating that a majority of participants engaged in physical activities (approximately 70% reported walking and 50% reported vigorous activities in the past month), no statistically significant associations were found between PA status and occurrence of bothersome pain during the three-year follow-up. This represents a novel contribution to the existing body of research, as it is, to the best of our knowledge, the first such investigation focused on a nationally representative sample of older adults in the United States.

Despite a growing body of research on the link between PA and pain management in older adults, the results remain mixed (Ellingson et al., 2012; Naugle & Riley, 2014; Umeda et al., 2016). While our study, which measured walking and vigorous activities, did not find a predictive association with bothersome pain, other studies report a protective effects of PA (Baumbach et al., 2022; Fancourt & Steptoe, 2018; Ferretti et al., 2019; Niederstrasser & Attridge, 2022; O’Neill et al., 2021). This inconsistency may be due to several factors. First, study designs differ in follow-up duration (ours being 3 years compared to others spanning 4–10 years) and participant age (ours focused on older adults with a mean age of 75, while others included participants as young as 40). Second, variations exist in how pain and PA were measured (questionnaires, metabolic equivalent assessments, accelerometers). Finally, the choice of statistical analysis can impact the detection of associations. For example, only high levels of PA (i.e., heavy manual work or vigorous activity more than once a week), not moderate weekly activity, when compared to a sedentary lifestyle, was found to be associated with a reduced risk of suffering from musculoskeletal pain among older adults aged 50 or older in the English Longitudinal Study of Ageing across a decade (Fancourt & Steptoe, 2018; Niederstrasser & Attridge, 2022). In the German Ageing Survey of individuals aged 40 or older who were followed for 3 years, both the onset of pain and the disappearance of pain were associated with increased total PA and PA intensity (Baumbach et al., 2022). In the Irish Longitudinal Studying on Ageing, older adults aged 50 or older in the “No Pain” class with low or moderate PA, compared to those with higher PA, were more likely to transition into the “Higher Impact Pain” class during a 4-year follow-up time (O’Neill et al., 2021). In the Coronary Artery Risk Development in Young Adults Study (CARDIA) of participants with a mean age of 45 years, knee symptoms including pain was found to have a bidirectional association with accelerometer estimated sedentary time and PA at the 5- and 10- year follow ups (Whitaker et al., 2021).

Compared to these studies, the participants in our study tended to be older, as the baseline age to participate in NHATS was 65 years, and the mean age of our participants was approximately 75 years. While we were able to investigate the PA-pain associations using repeated measures and two different measures of PA, none of the associations were statistically significant, though the point estimate of the relative risk of vigorous activity was slightly below 1. In addition, the null findings were for both the operationally defined pain measures, bothersome pain and persistent bothersome pain. These mixed findings reflect the complex interplays between PA and pain, and suggest further research is needed. In particular, emphases should be given on accurate measurement of both pain and PA among representative samples of older adults over time, as different types of PA may differentially impact pain inhibitory and modulation process, as well as pain progression (Ambrose & Golightly, 2015; Fanning et al., 2023; Hodges & Smeets, 2015; Naugle et al., 2016; Naugle et al., 2017).

Our study benefits from using a large, nationally representative sample and the application of robust PSW statistics. However, some limitations require consideration. First, self-reported data were used for both PA and pain assessment, which were subject to potential recall bias and misclassification. For example, for a small portion (4%–6%) of participants who reported no bothersome pain but used pain medications, there were uncertainties regarding their bothersome pain occurrence.” Secondly, while we captured two PA intensities (walking as low impact activity and vigorous activities as high impact activities), details regarding intensity and duration were absent. This might explain the higher reported activity levels in our study (70% walking, 50% vigorous activity) compared to a similar study using both subjective and objective measures (50% inactive, 35% moderately active, and 15% highly active) (Li et al., 2023). Future research incorporating objective measures such as accelerometers for a more comprehensive PA characterization would be valuable (Schrack et al., 2023). Similarly, a more granular assessment of pain, including specific anatomical locations like back or knees, and information on pain medication and pain management, could provide deeper insights. Building on this study, future studies should investigate the association of functional limitation and managed, as well as unmanaged pain. Finally, the impact of PA on pain occurrence might be more evident with a longer follow-up period exceeding three years. Further longitudinal studies with extended monitoring of both PA and pain are warranted to explore this possibility. By addressing these limitations in future studies, we can gain a clearer picture of the relationship between PA and pain management in older adults.

Conclusion

Pain is a major concern for a significant portion of older adults, and its risk grows with age. Physical activity has emerged as potential non-pharmacological approach for pain management, but research on its effect on preventing pain in older adults is inconclusive (Antcliff et al., 2021; Kelley et al., 2011; Tse et al., 2011). Our study, which examined a nationally representative sample of the US older adults, did not find a statistically significant link between PA (captured by walking and vigorous activities) and the development of bothersome or persistent pain over a three-year period. These findings suggest the need for further research, particularly well-designed, prospective studies with randomized controlled trials, to determine the role of PA in preventing pain among older adults.

Supplementary Material

1

What this paper adds

  • Physical activity (PA) has been reported to relieve pain and improve pain related symptoms, yet its impact in preventing pain occurrence among older adults remains unclear.

  • We provided important population-level information about the epidemiological statistics on pain occurrence and PA, as well as their associations by applying rigorous propensity score weighting methods to a nationally representative sample of Medicare-enrolled older adults.

  • While a majority of participants engaged in PA at baseline, no statistically significant associations were found between PA status and occurrence of bothersome pain during a 3-year follow-up.

Applications of study findings

  • Our study revealed a substantial increase in bothersome pain occurrence among older Americans within a relatively short 3-year time frame, highlighting the importance of pain management for this vulnerable group.

  • Our null finding adds to the body of research which has showed mixed results on the link between PA and pain development in older adults, and further suggests the need of in-depth investigations.

Acknowledgements

The authors wish to thank the participants and researchers of the National Health and Aging Trends Study (NHATS). NHATS is produced and distributed by www.nhats.org with funding from the National Institute on Aging (grant number U01AG032947).

Funding

This work was supported by a pilot award from the National Institute on Aging grant (grant number: P30AG028741) to Claude D. Pepper Older Americans Independence Center of the Icahn School of Medicine at Mount Sinai.

Footnotes

Declaration of Conflicting interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval

The study was reviewed by the institution review board of Mount Sinai Hospital and determined exempted human research (IRB number: STUDY-23–00473). This secondary data analysis used publicly available de-identified data from the National Health and Aging Trends Study (NHATS).

Reference

  1. Alzahrani H, Mackey M, Stamatakis E, Zadro JR, & Shirley D (2019). The association between physical activity and low back pain: a systematic review and meta-analysis of observational studies. Scientific Reports, 9(1), 8244. 10.1038/s41598-019-44664-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ambrose KR, & Golightly YM (2015). Physical exercise as non-pharmacological treatment of chronic pain: Why and when. Best Practice & Research: Clinical Rheumatology, 29(1), 120–130. 10.1016/j.berh.2015.04.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Antcliff D, Keenan AM, Keeley P, Woby S, & McGowan L (2021). Testing a newly developed activity pacing framework for chronic pain/fatigue: a feasibility study. BMJ Open, 11(12), e045398. 10.1136/bmjopen-2020-045398 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Arnes AP, Nielsen CS, Stubhaug A, Fjeld MK, Johansen A, Morseth B, Strand BH, Wilsgaard T, & Steingrimsdottir OA (2023). Longitudinal relationships between habitual physical activity and pain tolerance in the general population. PloS One, 18(5), e0285041. 10.1371/journal.pone.0285041 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Arroll B, Goodyear-Smith F, Crengle S, Gunn J, Kerse N, Fishman T, Falloon K, & Hatcher S (2010). Validation of PHQ-2 and PHQ-9 to screen for major depression in the primary care population. Annals of Family Medicine, 8(4), 348–353. 10.1370/afm.1139 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Baumbach L, Konig HH, & Hajek A (2022). Association between onset and disappearance of pain and changes in physical activity among older adults - longitudinal evidence from the German Ageing Survey. Psychogeriatrics, 22(1), 67–76. 10.1111/psyg.12777 [DOI] [PubMed] [Google Scholar]
  7. Bidonde J, Fisher E, Perrot S, Moore RA, Bell RF, Makri S, & Hauser W (2023). Effectiveness of non-pharmacological interventions for fibromyalgia and quality of review methods: an overview of Cochrane Reviews. Seminars in Arthritis and Rheumatism, 63, 152248. 10.1016/j.semarthrit.2023.152248 [DOI] [PubMed] [Google Scholar]
  8. Cook BL, McGuire TG, Meara E, & Zaslavsky AM (2009). Adjusting for Health Status in Non-Linear Models of Health Care Disparities. Health Services and Outcomes Research Methodology, 9(1), 1–21. 10.1007/s10742-008-0039-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Domenichiello AF, & Ramsden CE (2019). The silent epidemic of chronic pain in older adults. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 93, 284–290. 10.1016/j.pnpbp.2019.04.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ellingson LD, Shields MR, Stegner AJ, & Cook DB (2012). Physical activity, sustained sedentary behavior, and pain modulation in women with fibromyalgia. Journal of Pain, 13(2), 195–206. 10.1016/j.jpain.2011.11.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ettinger WH Jr., Fried LP, Harris T, Shemanski L, Schulz R, & Robbins J (1994). Self-reported causes of physical disability in older people: the Cardiovascular Health Study. CHS Collaborative Research Group. Journal of the American Geriatrics Society, 42(10), 1035–1044. 10.1111/j.1532-5415.1994.tb06206.x [DOI] [PubMed] [Google Scholar]
  12. Fancourt D, & Steptoe A (2018). Physical and Psychosocial Factors in the Prevention of Chronic Pain in Older Age. Journal of Pain, 19(12), 1385–1391. 10.1016/j.jpain.2018.06.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fanning J, Brooks AK, Robison JT, Irby MB, Ford S, N’Dah K, & Rejeski WJ (2023). Associations between patterns of physical activity, pain intensity, and interference among older adults with chronic pain: a secondary analysis of two randomized controlled trials. Front Aging, 4, 1216942. 10.3389/fragi.2023.1216942 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Ferretti F, Silva M. R. d., Pegoraro F, Baldo JE, & Sá C. A. d. (2019). Chronic pain in the elderly, associated factors and relation with the level and volume of physical activity. Brazilian Journal Of Pain, 2(1). 10.5935/2595-0118.20190002 [DOI] [Google Scholar]
  15. Freedman VA, Kasper JD, Cornman JC, Agree EM, Bandeen-Roche K, Mor V, Spillman BC, Wallace R, & Wolf DA (2011). Validation of new measures of disability and functioning in the National Health and Aging Trends Study. Journals of Gerontology. Series A: Biological Sciences and Medical Sciences, 66(9), 1013–1021. 10.1093/gerona/glr087 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Geneen LJ, Moore RA, Clarke C, Martin D, Colvin LA, & Smith BH (2017). Physical activity and exercise for chronic pain in adults: an overview of Cochrane Reviews. Cochrane Database of Systematic Reviews, 4(4), CD011279. 10.1002/14651858.CD011279.pub3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Heneweer H, Vanhees L, & Picavet HS (2009). Physical activity and low back pain: a U-shaped relation? Pain, 143(1–2), 21–25. 10.1016/j.pain.2008.12.033 [DOI] [PubMed] [Google Scholar]
  18. Hodges PW, & Smeets RJ (2015). Interaction between pain, movement, and physical activity: short-term benefits, long-term consequences, and targets for treatment. Clinical Journal of Pain, 31(2), 97–107. 10.1097/AJP.0000000000000098 [DOI] [PubMed] [Google Scholar]
  19. Ji Y, & Xiang X (2023). The Reciprocal Relationship of Physical Capacity and Mental Health: A Random Intercept Cross-Lagged Panel Model Analysis. Journal of Gerontological Social Work, 66(7), 924–941. 10.1080/01634372.2023.2197017 [DOI] [PubMed] [Google Scholar]
  20. Jones MR, Ehrhardt KP, Ripoll JG, Sharma B, Padnos IW, Kaye RJ, & Kaye AD (2016). Pain in the Elderly. Current Pain and Headache Reports, 20(4), 23. 10.1007/s11916-016-0551-2 [DOI] [PubMed] [Google Scholar]
  21. Kelley GA, Kelley KS, Hootman JM, & Jones DL (2011). Effects of community-deliverable exercise on pain and physical function in adults with arthritis and other rheumatic diseases: a meta-analysis. Arthritis Care & Research, 63(1), 79–93. 10.1002/acr.20347 [DOI] [PubMed] [Google Scholar]
  22. Kroenke K, Spitzer RL, Williams JB, Monahan PO, & Lowe B (2007). Anxiety disorders in primary care: prevalence, impairment, comorbidity, and detection. Annals of Internal Medicine, 146(5), 317–325. 10.7326/0003-4819-146-5-200703060-00004 [DOI] [PubMed] [Google Scholar]
  23. Li W, Li Y, Wen M, & Liu B (2023). Physical Activity Patterns and Cognitive Health among Older Adults in the United States. Journal of Applied Gerontology, 42(3), 409–418. 10.1177/07334648221139480 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Linden A, & Adams JL (2010). Evaluating health management programmes over time: application of propensity score-based weighting to longitudinal data. Journal of Evaluation in Clinical Practice, 16(1), 180–185. 10.1111/j.1365-2753.2009.01361.x [DOI] [PubMed] [Google Scholar]
  25. Merriwether EN, Frey-Law LA, Rakel BA, Zimmerman MB, Dailey DL, Vance CGT, Golchha M, Geasland KM, Chimenti R, Crofford LJ, & Sluka KA (2018). Physical activity is related to function and fatigue but not pain in women with fibromyalgia: baseline analyses from the Fibromyalgia Activity Study with TENS (FAST). Arthritis Research & Therapy, 20(1), 199. 10.1186/s13075-018-1671-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Montaquila J, Freedman V, Edwards B, & Kasper J (2012, May 10). National Health and Aging Trends Study (NHATS) Round 1 Sample Design and Selection. NHATS Technical Paper #1. Retrieved October 13, 2023 from https://www.nhats.org/sites/default/files/2021-01/NHATS%20Round%201%20Sample%20Design%2005_10_12_2.pdf [Google Scholar]
  27. Naugle KM, Naugle KE, & Riley JL 3rd. (2016). Reduced Modulation of Pain in Older Adults After Isometric and Aerobic Exercise. Journal of Pain, 17(6), 719–728. 10.1016/j.jpain.2016.02.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Naugle KM, Ohlman T, Naugle KE, Riley ZA, & Keith NR (2017). Physical activity behavior predicts endogenous pain modulation in older adults. Pain, 158(3), 383–390. 10.1097/j.pain.0000000000000769 [DOI] [PubMed] [Google Scholar]
  29. Naugle KM, & Riley JL 3rd. (2014). Self-reported physical activity predicts pain inhibitory and facilitatory function. Medicine and Science in Sports and Exercise, 46(3), 622–629. 10.1249/MSS.0b013e3182a69cf1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Niederstrasser NG, & Attridge N (2022). Associations between pain and physical activity among older adults. PloS One, 17(1), e0263356. 10.1371/journal.pone.0263356 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. O’Neill A, O’Sullivan K, & McCreesh K (2021). Lower levels of physical activity are associated with pain progression in older adults, a longitudinal study. European Journal of Pain (London, England), 25(7), 1462–1471. 10.1002/ejp.1759 [DOI] [PubMed] [Google Scholar]
  32. Ohlman T, Miller L, Naugle KE, & Naugle KM (2018). Physical Activity Levels Predict Exercise-induced Hypoalgesia in Older Adults. Medicine and Science in Sports and Exercise, 50(10), 2101–2109. 10.1249/MSS.0000000000001661 [DOI] [PubMed] [Google Scholar]
  33. Patel KV, Guralnik JM, Dansie EJ, & Turk DC (2013). Prevalence and impact of pain among older adults in the United States: findings from the 2011 National Health and Aging Trends Study. Pain, 154(12), 2649–2657. 10.1016/j.pain.2013.07.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Rikard SM, Strahan AE, Schmit KM, & Guy GP Jr. (2023). Chronic Pain Among Adults - United States, 2019–2021. MMWR: Morbidity and Mortality Weekly Report, 72(15), 379–385. 10.15585/mmwr.mm7215a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Ritchie CS, Patel K, Boscardin J, Miaskowski C, Vranceanu AM, Whitlock E, & Smith A (2023). Impact of persistent pain on function, cognition, and well-being of older adults. Journal of the American Geriatrics Society, 71(1), 26–35. 10.1111/jgs.18125 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Rundell SD, Phelan EA, Patel KV, Jones BL, & Marcum ZA (2020). Longitudinal Patterns of Pain Reporting Among Community-dwelling Older Adults. Clinical Journal of Pain, 36(12), 912–922. 10.1097/AJP.0000000000000874 [DOI] [PubMed] [Google Scholar]
  37. Santos MCS, de Andrade SM, Gonzalez AD, Dias DF, & Mesas AE (2018). Association Between Chronic Pain and Leisure Time Physical Activity and Sedentary Behavior in Schoolteachers. Behavioral Medicine, 44(4), 335–343. 10.1080/08964289.2017.1384358 [DOI] [PubMed] [Google Scholar]
  38. SAS Inc. Sample 62362: Estimate and test differences, ratios, or contrasts of means in generalized linear models. Retrieved April 25th, 2024 from https://support.sas.com/kb/62/362.html
  39. Schrack J, Zhang T, Wanigatunga A, Zipunnikov V, & Freedman V (2023). Accelerometer-Derived Physical Activity in the National Health and Aging Trends Study. Innovation in Aging, 7(Supplement_1), 165–166. 10.1093/geroni/igad104.0542 [DOI] [Google Scholar]
  40. Shirazi D, Haudenschild C, Lynch DH, Fanous M, Kahkoska AR, Jimenez D, Spangler H, Driesse T, & Batsis JA (2023). Obesity, multiple chronic conditions, and the relationship with physical function: Data from the national health and aging trends survey. Archives of Gerontology and Geriatrics, 107, 104913. 10.1016/j.archger.2022.104913 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Skou ST, Bricca A, & Roos EM (2018). The impact of physical activity level on the short- and long-term pain relief from supervised exercise therapy and education: a study of 12,796 Danish patients with knee osteoarthritis. Osteoarthritis and Cartilage, 26(11), 1474–1478. 10.1016/j.joca.2018.07.010 [DOI] [PubMed] [Google Scholar]
  42. Stubbs B, Binnekade TT, Soundy A, Schofield P, Huijnen IP, & Eggermont LH (2013). Are older adults with chronic musculoskeletal pain less active than older adults without pain? A systematic review and meta-analysis. Pain Medicine, 14(9), 1316–1331. 10.1111/pme.12154 [DOI] [PubMed] [Google Scholar]
  43. Tse MM, Wan VT, & Ho SS (2011). Physical exercise: does it help in relieving pain and increasing mobility among older adults with chronic pain? Journal of Clinical Nursing, 20(5–6), 635–644. 10.1111/j.1365-2702.2010.03548.x [DOI] [PubMed] [Google Scholar]
  44. Turner JA, Franklin G, Heagerty PJ, Wu R, Egan K, Fulton-Kehoe D, Gluck JV, & Wickizer TM (2004). The association between pain and disability. Pain, 112(3), 307–314. 10.1016/j.pain.2004.09.010 [DOI] [PubMed] [Google Scholar]
  45. Umeda M, Lee W, Marino CA, & Hilliard SC (2016). Influence of moderate intensity physical activity levels and gender on conditioned pain modulation. Journal of Sports Sciences, 34(5), 467–476. 10.1080/02640414.2015.1061199 [DOI] [PubMed] [Google Scholar]
  46. Vaegter HB, Kinnunen M, Verbrugghe J, Cunningham C, Meeus M, Armijo-Olivo S, Bandholm T, Fullen BM, Wittink H, Morlion B, & Reneman MF (2024). Physical activity should be the primary intervention for individuals living with chronic pain A position paper from the European Pain Federation (EFIC) ‘On the Move’ Task Force. European Journal of Pain (London, England), 28(8), 1249–1256. 10.1002/ejp.2278 [DOI] [PubMed] [Google Scholar]
  47. Whitaker KM, Pettee Gabriel K, Laddu D, White DK, Sidney S, Sternfeld B, Lewis CE, & Jacobs DR Jr. (2021). Bidirectional associations of accelerometer measured sedentary behavior and physical activity with knee pain, stiffness, and physical function: The CARDIA study. Preventive Medicine Reports, 22, 101348. 10.1016/j.pmedr.2021.101348 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Yang C, Cuerden MS, Zhang W, Aldridge M, & Li L (2023). Propensity score weighting with survey weighted data when outcomes are binary: a simulation study. Health Services and Outcomes Research Methodology. 10.1007/s10742-023-00317-y [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2046008-supplement-1.pdf (110.6KB, pdf)

RESOURCES