Sage Journals: Discover world-class research

Abstract

Sense of purpose is a key construct explaining individual differences in health. However, much is still unclear about how sense of purpose and health affect each other over time in older age. Using four waves of data from the Health and Retirement Study (N = 11,390, 50+ years old) spanning 12 years, we investigated the reciprocal associations between sense of purpose and multiple objective and subjective markers of health (e.g., self-reported health; grip strength). Across the 12 years, better health was associated with higher levels of purpose for all investigated health markers. Cross-lagged panel models implied reciprocal associations for 6 of the 11 health markers. When looking at within-person associations (i.e., random intercept cross-lagged models), purpose emerged as a stronger predictor of later health than the other way around, with purpose changes predicting subsequent within-person changes in five health markers while changes in only two health markers predicted later changes in purpose. Taken together, these findings further support the notion that sense of purpose has protective effects against health decline. We discuss the findings in the context of lifespan theory and emphasize the potential benefit of considering sense of purpose as a target for public health interventions.

Plain language summary

Having a strong sense of purpose in life has been shown to be associated with better health in late adulthood. A strong sense of purpose means that individuals have and follow personally meaningful goals. This is associated with more active engagement with life and higher well-being. Research shows that more purposeful individuals generally have better health, but it is unclear if purpose protects against health decline and acts to promote better health or if individuals who are healthier are somehow better able to follow and enact their purpose. In other words, the direction of this relationship is not clear, and it is likely that purpose and health influence each other (bi-directionally) over time. To answer this question, we used existing data from a large study among adults 50 years or older that measured purpose and various aspects of health every 4 years over a total time span of 12 years. We found that older adults with a stronger sense of purpose were healthier overall, and better able to maintain their health over time. Healthier individuals also reported a stronger sense of purpose over time. However, when people declined in their sense of purpose, this was generally linked to declining health later on. In contrast, health declines were generally not followed by declines in sense of purpose. This suggests that having a strong sense of purpose may be beneficial to older adults’ health and that despite health declines, older adults may generally hold on to or adapt their sense of purpose.

Keywords

sense of purpose health old age bidirectional associations cross-lagged panel analysis

Introduction

With the rapidly aging population, there is an increasing need to prioritize the health and independence of older adults. This not only carries benefits in terms of their quality of life but also has wider societal implications, such as a lowered healthcare burden (Connolly et al., 2017). Importantly, older adults represent a heterogeneous group (Lowsky et al., 2014). Some stay healthy and independent longer, while others face more challenges as they age (Mueller et al., 2017). Research is needed to better understand the factors that underly these individual differences in health trajectories. One promising factor that could help explain some of these differences is the extent to which individuals feel a sense of purpose in life (Pfund & Lewis, 2020).

Sense of purpose can be described as the extent to which individuals perceive their goals as personally significant and clear, such that they foster active participation in life and align with fundamental aspects of their identity (Pfund & Lewis, 2020; Ryff, 1989). Sense of purpose, by definition, promotes engagement in life goals, gives a sense of intentionality and directedness to their existence, and, as such, enriches their lives with meaning (Lewis, 2020; Ryff & Keyes, 1995). Purpose can be seen as a motivational component of meaning in life (Martela & Steger, 2023), the latter also consisting of coherence (i.e., making sense of one’s life; cognitive component) and significance (i.e., having the feeling that one’s life is worth living; evaluative component). The directedness associated with purpose might provide a strong psychological resource in old age when health is declining, and social isolation becomes a growing concern. A strong sense of purpose might provide individuals with clear goals and the motivation needed to stay engaged despite these challenges (Irving et al., 2017; Pfund & Lewis, 2020).

Sense of purpose: A resource for healthy aging

Stronger sense of purpose has been associated with various markers of healthy aging, such as better subjective health, higher cognitive and physical functioning, and increased longevity (e.g., Boyle et al., 2009; Kim et al., 2022; Lewis et al., 2017; Scheier et al., 2006; Sone et al., 2008). Previous studies also suggest that individuals with higher purpose levels have fewer chronic conditions, but findings for specific chronic conditions are mixed (Hafez et al., 2018; Kim et al., 2022).

Kim et al. (2022) observed that adults 50 and older with higher sense of purpose had better physical health (e.g., reduced risk of lung diseases, mortality, number of chronic conditions, and better subjective health) 4 years later, even after accounting for baseline levels of purpose. Similarly, Willroth et al. (2021) found that more positive changes in sense of purpose predicted better physical health 4 and 9 years later in 3 different cohorts of middle-aged and older adults. Research has also linked a higher sense of purpose with better hearing (finding replicated across two samples; Sutin et al., 2022) and better psychosocial well-being (Gudmundsdottir et al., 2023; Kim et al., 2022; Martela & Steger, 2023). Moreover, recent findings also suggest that these associations are incremental beyond personality traits and well-being. For example, sense of purpose is a unique correlate of health after accounting for trait conscientiousness (Hill et al., 2021) and predicts mortality after accounting for life satisfaction (Martela et al., 2024).

Possible pathways from purpose to health

Multiple pathways have been suggested to explain the link between sense of purpose in life and healthy aging (Irving et al., 2017; Kim et al., 2019). One such pathway may involve the health-promoting behaviors that have been associated with a higher sense of purpose. For example, purpose has been linked to higher involvement in physical activity (e.g., Kim et al., 2022), this association being also confirmed using objective measures of mobility (i.e., purposeful individuals have a higher step count; Hooker & Masters, 2016). Further, sense of purpose has been linked to preventive healthcare use and medical checkups (Kim et al., 2014), a higher intake of fruits and vegetables (Steptoe & Fancourt, 2019), and better sleep (Kim et al., 2015, 2020; Steptoe & Fancourt, 2019). Overall, a higher sense of purpose may increase the likelihood of engaging in behaviors that promote recovery and health (e.g., engaging in physical activity) while reducing behaviors that contribute to health deterioration (e.g., smoking; Kim et al., 2019).

Sense of purpose in life might also exert beneficial effects by acting as an emotional buffer in the face of stress. For example, Schaefer et al. (2013) demonstrated in a laboratory study involving middle-aged adults that individuals with a higher sense of purpose exhibited a slower eye-blink startle reflex in response to negative stimuli, indicating reduced stress reactivity. Similarly, Polenick et al. (2018), in a study involving older adult caregiving dyads, found that caregivers with a stronger sense of purpose reported fewer emotional and physical difficulties in providing care to their spouse, with this effect being particularly pronounced among women. Further, Gudmundsdottir et al. (2023) found longitudinal links between sense of purpose, life satisfaction and positive affect as indicators of subjective well-being.

The link between purpose and health could also be explained by common factors in the types of environments that individuals with higher sense of purpose might typically find themselves in. They are more socially active and receive more social support (Mei et al., 2021), which could help maintain good physical functioning (Bath & Deeg, 2005; Cherry et al., 2013; Saadeh et al., 2020).

Sense of purpose changes over time

Importantly, sense of purpose in life is not static. It changes throughout life, with past research suggesting that sense of purpose shows vulnerability to decline with age (Hill & Weston, 2019; Willroth et al., 2021). Although these declines are small, they might result from changes in health status or changes in one’s ability to be actively involved in activities known to make individuals feel purposeful (e.g., leisure activities, social involvement, and physical activities; Lewis & Hill, 2020; Merten et al., 2022). Alternatively, role losses such as retirement, widowhood, and loss of friends could influence individuals’ life goals in a way that older adults may leave behind commitments that used to guide their lives (Chen et al., 2020; Hill & Weston, 2019; Lewis & Hill, 2020; Pinquart, 2002).

Health: A resource for sense of purpose

Most past research has shown that sense of purpose is a protective factor for good health. However, some research also suggests the reverse predictive direction, namely, that good health and good quality of life might foster a sense of purpose (Gudmundsdottir et al., 2023; Nakamura et al., 2022; Steptoe & Fancourt, 2020) or that negative health events can have a detrimental effect on sense of purpose (Lewis et al., 2020; Sias & Turtle, 2022). This could occur because declines in health and functional abilities may pose restrictions on the type of activities older adults can perform and from which they derive their sense of purpose. For example, starting to experience general weakness (e.g., lower grip strength and slower walking speed) or deficits in hearing and eyesight can make it harder to engage in some social and generative activities they might value—such as volunteering or spending time with grandchildren (Dohlman et al., 2023; Gottlieb & Gillespie, 2008; Same et al., 2020). Certain conditions may also directly or indirectly reduce individuals’ motivation to engage in such activities because of the associated discomfort, potential worsening of symptoms (Emerson et al., 2018; Meek et al., 2018), or because these conditions or their medication directly impact biological processes associated with motivation and reward processing (Muhammed et al., 2016; Swardfager et al., 2016; Taylor et al., 2016).

Yet, findings for this direction of the association are mixed, as some find limited support for a predictive effect from health (i.e., the number of and the individual effects of specific chronic conditions) on sense of purpose (Chen et al., 2020) or no changes in purpose following health adversity (i.e., diagnosis of multiple diseases or health events; Hill et al., 2021). These inconsistencies might be, at least in part, due to the different health domains investigated across studies (e.g. chronic pain, diseases). It is possible that only changes in specific domains of health tend to push an individual to leave behind or re-evaluate past goals and prevent them from finding new sources of purpose. For example, individuals might show decline in sense of purpose when they report increasing pain levels from one assessment to another, but they might not decline in purpose with worsening eyesight or hearing. Thus, examining various health markers separately addresses the potential heterogeneity in associations with purpose, offering a more comprehensive insight, whilst also facilitating a comparison with previous mixed findings.

Apart from considering various health markers, it is important to distinguish between objective and subjective measures of health. Although related, they are not interchangeable (Pinquart, 2001). Individuals may self-report as healthy despite objective health declines (Huisman & Deeg, 2010; Jylhä, 2009), and their sense of purpose may be more strongly tied to the subjective perception of their health than the objective one. In contrast, some older adults might report reduced sense of purpose when they subjectively think their health is declining, even in the absence of objective evidence showing deteriorating health. As previous work has shown, subjective health perceptions are shaped by multiple factors, such as points of reference (e.g., comparisons to others or oneself in the past; Sargent-Cox et al., 2010) or personality traits (Elran-Barak et al., 2019; Goodwin & Engstrom, 2002). Finally, associations between purpose and subjective markers of health may be biased due to common method variance. Therefore, it is important both from conceptual and methodological reasons to separate subjective and objective markers of health, as well as different types of health conditions.

Considering the directionality and level of associations

While there is research documenting sense of purpose independently as both an antecedent (e.g., Kim et al., 2022) and outcome of health (e.g., Nakamura et al., 2022), no past research, to our knowledge, has looked at both predictive directions within the same study to get a clearer picture of the directionality of these effects. This is important, as sense of purpose, a trait-like construct (Pfund, Hofer, et al., 2022), is likely to function in feedback loops throughout one’s life span, including older age, with potential changes in health and sense of purpose reinforcing each other (see similar discussion for purpose and subjective well-being: Gudmundsdottir et al., 2023, and health and personality traits: Chereches et al., 2023; Mroczek et al., 2020; Mueller et al., 2017). To illustrate, a high sense of purpose derived from one’s volunteering activities might lead an individual to be regularly involved in activities that promote physical exercise and opportunities to create meaningful connections with others. These are factors known to promote good health (Kail & Carr, 2020; Nakamura et al., 2023). In turn, a healthy lifestyle (e.g., physical exercise and social involvement) and good health might feed back into one’s sense of purpose, as with good health and well-being, one should be better able and motivated to invest in the initial purposeful aims. Indeed, past research has found such feedback loops between sense of purpose and physical activity, with the two reinforcing each other over time (Yemiscigil & Vlaev, 2021). Similar bidirectional associations have been found for purpose and positive affect or life satisfaction (Gudmundsdottir et al., 2023). Thus, in the current study, we aim to extend past research by testing both directions of the predictive effect: To what extent sense of purpose is predicted by and predicts health, respectively.

Further, the present study also extends previous longitudinal work on purpose-health relationships by considering the reciprocal associations at both the between- and within-person levels. The distinction is important, as between-person findings do not necessarily match within-person findings (Hamaker et al., 2020; Mueller et al., 2018). Yet, both provide valuable information. Between-person findings can help inform who is at a risk of showing a decline in health or sense of purpose, while within-person findings can inform if, for instance, declines in health predict declines in purpose, and vice versa. As most previous research on the health-purpose associations has not disentangled within- from between-person associations, it is imperative to test if previous findings are replicated when controlling for stable between-person differences.

Moderating effect of age

While past research showed an association between sense of purpose in life and health, it rarely considered how these associations might differ across age, such as late middle-aged (defined as 50 and 65) and older adults (defined as 65+ years). There are also competing assumptions regarding how purpose affects health with age.

On the one hand, as health issues become more severe with age, the association between purpose and health might strengthen, as purpose could serve as a crucial psychological resource to counterbalance the increasing health declines. On the other hand, in line with the preposition of lifespan theory, the increasing physiological constraints that come with advancing age may outbalance the advantages conferred by certain psychosocial resources (Baltes & Smith, 2003; Löckenhoff et al., 2008). In the context of purpose-health associations, this would mean that with advancing age and more pervasive physical declines, the protective effect of purpose for later health might lose importance.

The reverse direction—from health to later sense of purpose—is also complex. It is possible that health (and health decline) in late middle age would lead to stronger changes in later sense of purpose than it would in older age groups. This might be because older individuals, already expecting such declines, might not experience a significant decrease in their sense of purpose as their health worsens. They may have adjusted to the inevitability of these declines (Galenkamp et al., 2012; Wurm et al., 2008). In fact, research suggests that subjective and more objective health markers increasingly diverge with age (Pinquart, 2001), as the subjective criteria for what it means to be healthy change. In contrast, in late middle age, declines in health might be viewed as more unexpected or dramatic, disrupting one’s life goals and sense of direction.

The present study

In this pre-registered study (https://osf.io/m627u), we used pre-existing data from the Health and Retirement Study (HRS) to explore whether sense of purpose and various health markers (objective and subjectively reported) are bidirectionally related in middle to older age (50+ years). The measurement period spanned 12 years, with data collected on four occasions. As such, we were focusing on sustained, long-term effect of purpose on health and vice versa, rather than short-term changes. Although associations at shorter timescales are insightful when further looking into potential mechanisms, our primary focus was identifying meaningful trends that persist across longer time periods. We assumed that meaningful changes in purpose and health usually occur slowly, as well as the process by which health changes in response to purpose changes (e.g., via more physical exercise or reduced loneliness). The 4-year interval between assessments in this study allowed long-term effects to accumulate—potentially representing meaningful and sustained associations that would be of interest to intervention research.

To examine whether the reciprocal associations would be similar across different operationalizations of health (i.e., objective and subjective reports and different types of ailments), we compared findings across the following health- and disability markers: Self-rated health, number of physical conditions, body mass index (BMI), chronic pain, hearing, eyesight, grip strength, walking speed, basic activities of daily living (BADLs), and instrumental activities of daily living (IADLs). Given the somewhat exploratory nature of the study, we did not pre-select the health markers based on certain criteria but strived for a comprehensive summary of the information available in the dataset, focusing on clear markers of physical health or functional limitations rather than health behaviors. Distinguishing different dimensions of health offers a more nuanced understanding of the associations between purpose and health, which is important to guide future research into the specific processes and mechanisms underlying the associations and to inform intervention research and applications.

Our primary research aim was five-fold. First, we tested if sense of purpose and the aforementioned health markers are correlated cross-sectionally (hypothesis 1a) and also if they are correlated over time (hypothesis 1b), meaning that increases in one go hand in hand with increases in the other. To this aim, we employed bivariate latent growth curve models and expected to find positive associations between purpose and health, both in terms of stable levels (i.e., intercepts) and long-term change (i.e., slopes; Irving et al., 2017; Lewis & Hill, 2021).

Second, to extend previous research, which by and large has focused on concurrent or longitudinal associations with sense of purpose either as an antecedent or as an outcome of health, we sought to assess both simultaneously. Specifically, we tested whether initial sense of purpose predicts later health above and beyond initial health levels and vice versa. Given previous longitudinal studies showing purpose as a predictor of subsequent health (Hafez et al., 2018; Kim et al., 2017; Zilioli et al., 2015), health as an antecedent of purpose (Nakamura et al., 2022), and also research suggesting similar reciprocal links for sense of purpose and physical activity (Yemiscigil & Vlaev, 2021), we expected to see bidirectional associations between purpose and the health markers. More specifically, we expected that higher (vs. lower) scores on purpose would predict better subsequent health (hypothesis 2a) and vice versa (hypothesis 2b). To examine these research questions, we used cross-lagged panel models.

Third, we controlled for stable between-person differences in order to check if the aforementioned associations between purpose and health also hold at the within-person level, using random-intercept cross-lagged panel models. This can inform if within-person changes in sense of purpose predict within-person changes in health and vice versa. For this, we expected to observe the same bidirectional associations at the within-person level, where changes in purpose (around the stable mean) would predict changes in health (hypothesis 3a) and vice versa (hypothesis 3b).

Fourth, we also compared the strength of the effects of purpose on later health to that of health on subsequent purpose, and we did this for all time-lagged associations. This was to check, given a bidirectional relationship, which is more predictive of the other: purpose or the health markers. Given the lack of theory and past research on the matter, we did not formulate any specific hypotheses a-priori regarding these effects, but pre-registered the analyses.

Fifth, as an exploratory analysis (not preregistered), we examined if the associations would differ across age. Specifically, we defined late middle-aged adults as those of at least 50 years of age (and up to 64 years), and old adults as those 65 years and older. The late middle-aged group is less explored in previous studies, but they might differ from those 65 years and older in the extent of health decline and perceptions on health, which might influence the way and extent to which purpose and health interact. Due to limited research and competing assumptions on how these associations may change, we did not formulate specific hypotheses and report these findings as exploratory.

The present study thus contributes to the existing literature in multiple ways beyond replicating the observed association between purpose and markers of physical health by (a) formally testing bidirectionality and whether purpose is more strongly predicting health or vice versa, (b) assessing whether the associations are likely to reflect within-person processes or not, (c) decomposing stable trait-, temporal- and concurrent associations, (d) separating different types of subjectively and objectively reported health markers, and (e) assessing whether the associations are moderated by age.

Methods

This study is a re-analysis of existing and fully anonymous datasets, and as such, no additional ethical approval was obtained. The pre-registration, supplementary materials, and analysis scripts used in this study can be accessed in an OSF repository (https://osf.io/caghf/). Data analysis was conducted in R 4.1.3 (R Core Team, 2022) with the packages dplyr (Wickham et al., 2023), ggplot2 (Wickham, 2016), haven (Wickham et al., 2023), powRICLPM (Mulder, 2023), psych (Revelle, 2022), lavaan (Rosseel, 2012), and semTools (Jorgensen et al., 2025).

Participants and procedure

The data come from a total of four waves from two cohorts of the HRS, an open, ongoing panel study of adults 50 years and older in the US that started in 1992 (Sonnega et al., 2014). It is sponsored by the National Institute on Aging and approved by the Institutional Review Board at the University of Michigan. The sample is based on a multi-stage area probability sampling, making it nationally representative. After an initial interview in each household unit, survey data is gathered from all participants every two years (Sonnega et al., 2014).

Sense of purpose was measured every four years, with half of the study’s participants answering the survey from 2006 to 2018 and the other half from 2008 to 2020, creating two cohorts (Smith et al., 2017). To maximize the available sample size and power, we combined the two cohorts (2006, 2010, 2014, 2018 and 2008, 2012, 2016, 2020) and controlled for cohort differences in our analytic models.

Sense of purpose and most health markers were measured with self-reports, with the exception of walking speed and grip strength, which were measured in person (with the exception of 2020 due to social distancing measures). We only included participants who filled out the questionnaires for themselves, did not live in a nursing home during the study, reported to be cognitively healthy (i.e., no dementia or Alzheimer’s diagnosis), and provided at least two repeated measures of purpose and the health markers (done for each health marker separately). A detailed overview of the resulting sample sizes and descriptive statistics are presented in Table 1.

Table 1.

Sample Overview.

Purpose and …	N	Age (M)	Age (SD)	Women %	Not work. %	Waves purpose	Waves health
Self-reported health	11386	64.12	9.41	59.7	41.2	3.03	3.35
Vision	11332	64.08	9.38	59.7	41.4	3.03	3.34
Hearing	11388	64.12	9.41	59.7	41.2	3.03	3.35
Walking speed	5905	69.91	6.87	58.8	30.6	3.22	2.71
Grip strength	10239	64.16	9.35	59.2	41.7	3.10	2.89
Body mass index	11247	64.13	9.40	59.5	41.4	3.04	3.32
Chronic pain	11368	64.12	9.40	59.7	41.3	3.03	3.34
Physical conditions	11390	64.12	9.41	59.7	41.2	3.03	3.35
Basic ADLs	11390	64.12	9.41	59.7	41.2	3.03	3.35
Instrumental ADLs	11390	64.12	9.41	59.7	41.2	3.03	3.35

Note. N = overall sample size; Age (M/SD) = Mean and standard deviation of age at the first measurement occasion; Women % = Percentage of women at first wave; Not work. % = Percentage of participants not working at first wave; Waves purpose / health = Average number of completed waves for sense of purpose / the health marker. ADLs = Activities of daily living. Participants were 50–99 years old at the first measurement occasion (except for the walking speed sample: 53–94).

Based on the largest sample (i.e., N = 11,390), participants who completed all four measures of purpose reported to be on average more educated (d = 0.19; p < .001), more likely to not be working (OR = 1.35; p < .001), less likely to be female (OR = 0.93; p = .002), and more purposeful (d = 0.18; p < .001) at the first wave. Participants who completed all four waves of purpose also reported to be overall healthier at the first wave (average absolute d = 0.20; ranging in absolute size from d = 0.12 for chronic pain to d = 0.33 for self-reported health; all p < .001), with the exception of no significant differences in BMI.

Measures

Purpose in life

Purpose in life was assessed using a 7-item subscale from the Psychological Well-being Scale (Ryff, 1989; Ryff & Keyes, 1995). Participants responded to the statements (e.g., “I have a sense of direction and purpose in my life”) on a scale from 1 = strongly disagree to 6 = strongly agree. Reliability McDonald’s ω ranged from .74 to .77 across waves.

Self-rated health

Self-rated health was measured with a single item (“Would you say your health is excellent, very good, good, fair, or poor?”) on a 5-point scale ranging from 1 = Excellent to 5 = Poor. Scores were reverse coded so that higher values indicate better health.

Body mass index

The BMI was computed based on participants’ self-reported weight and height. We examined both the raw BMI as well as the absolute difference to a suggested ideal BMI of 26 for older adults (BMI-26; Winter et al., 2014)¹.

Physical conditions

Participants self-reported (0 = no; 1 = yes) if they were diagnosed with the following conditions: (a) diabetes, (b) hypertension, (c) stroke, (d) cancer, (e) heart condition (e.g., heart attack), (f) lung disease, or (g) arthritis. We used the number of conditions (ranging from 0 to 7) for the subsequent analyses.

Chronic pain

Participants reported whether they experienced bodily pain (Are you often troubled with pain? 0 = no; 1 = yes). If they responded yes, they were furthermore asked how severe the pain was (How bad is the pain most of the time: mild, moderate or severe? 1 = mild; 2 = moderate; 3 = severe). We combined the two questions to a score ranging from 0 = no pain to 3 = severe pain.

Hearing

Participants self-reported their hearing ability (Is your hearing excellent, very good, good, fair, or poor (using a hearing aid as usual)?) on a range from 1 = excellent to 5 = poor. We re-coded the scores so that higher values indicate better hearing.

Vision

Participants self-reported their eyesight (Is your eyesight excellent, very good, good, fair, or poor using glasses or corrective lenses as usual?) on a range from 1 = excellent to 5 = poor (and 6 = legally blind). We re-coded values so that higher scores indicated better eyesight, excluding participants who reported being legally blind.

Basic activities of daily living

Participants self-reported whether they had “any difficulty” (0 = no difficulty; 1 = any difficulty) with the following activities: (a) bathing, (b) dressing, (c) eating, (d) getting in/out of bed, (e) walking across a room and (f) using the toilet. For our analyses, we used a sum score ranging from 0 to 6.

Instrumental activities of daily living

Participants self-reported whether they had “any difficulty” (0 = no difficulty; 1 = any difficulty) with performing the following activities: (a) grocery shopping, (b) preparing meals, (c) managing money, (d) making phone calls, (e) taking medication, and (f) using a map. Each item response was coded as 0 = no difficulty or 1 = any difficulty. For our analyses, we used a sum score ranging from 0 to 6.

Grip strength

Grip strength in kilogram was measured twice for each hand. We used the maximum grip strength across the 4 measures. Grip strength was not measured in 2020 due to COVID-19 restrictions.

Walking speed

Participants were asked to walk 98.5 inches at a normal pace twice. We used the fastest measured time (in seconds) across the two measures and reversed it by subtracting it from 0 so that higher values indicate faster speed. Walking speed was not measured in 2020 due to COVID-19 restrictions.

Statistical analyses

All models were estimated with robust maximum likelihood (MLR) estimation to account for the non-normality of the health measures and full information maximum likelihood (FIML) to account for missing data. We interpreted the LGCM correlations using a benchmark of r = .10 / .20 / .30 for a small / medium / large effect size (Funder & Ozer, 2019; Gignac & Szodorai, 2016). For the standardized cross-lagged coefficients in the (RI-)CLPM we evaluated associations of .03 / .07 / .12 as small / medium / large (Orth et al., 2024). The power to detect a significant (p < .05) small effect with the smallest available sample size (N = 5,864; walking speed) was above 99% in the LGCMs, and generally above 95% for the RI-CLPMs (power determined by simulating data with the R packages lavaan and powRICLPM; see pre-registration for details). As the number of false positives is increased by the number of tests (i.e., 11 per association), we controlled for the false discovery rate (FDR) with the procedure recommended by Benjamini and Hochberg (1995). We also provide exact p-values and 95% confidence intervals for all relevant parameters. In the results section, unless otherwise specified, the reported associations were still significant after FDR correction.

Purpose measurement model and invariance

We first specified longitudinal measurement models for purpose, with four measurement occasion factors loading on the corresponding seven items. The same items were allowed to correlate with each other across time. Because model fit was insufficient (df = 302; χ² = 9748; CFI = .872; RMSEA = .052; SRMR = .050) based on common cut-offs (Bentler & Bonett, 1980; Kenny, 2020), we examined the modification indices and added residual correlations between item 1, 3 and 7 within each wave (see also Pfund, Ratner, et al., 2022). The resulting model fit was acceptable (df = 290; χ² = 4354; CFI = .945; RMSEA = .035; SRMR = .059). We tested for measurement invariance across time by constraining the factor loadings to equality across time (metric invariance: df = 308; χ² = 4417; CFI = .944; RMSEA = .034; SRMR = .059), as well as item intercepts to equality across time (metric invariance: df = 326; χ² = 4684; CFI = .941; RMSEA = .034; SRMR = .059). Based on suggested cut-offs for model fit decreases (i.e., ΔCFI <.01; Chen, 2007), the model achieved metric and scalar measurement invariance across time. For the current focus on associations, metric invariance was needed.

Structural models

We used three types of models to assess the dynamic associations between purpose and each of the health markers separately. For the first aim, we used a bivariate latent growth curve model (LGCM; see Figure 1(a)) to examine associations between (a) the baseline levels, (b) linear change across the 12 years, and (c) occasion-specific deviations from the level and linear change. In particular, this model allows us to examine codevelopment in sense of purpose and health across longer time spans. For the second and third aims, we used a cross-lagged panel model (CLPM; see Figure 1(b))² with lag-2 associations (to control for delayed effects; see Lüdtke & Robitzsch, 2022) and a random-intercept cross-lagged panel model (RI-CLPM; see Figure 1(c)), in order to examine the reciprocal associations between purpose and health across time. The interest here is in the directionality of associations over time. The RI-CLPM differs from the traditional CLPM by including a random intercept (i.e., level factor) controlling for stable between-person differences in purpose and health and stable unmeasured confounders (Hamaker et al., 2015; Mund & Nestler, 2019). As such, the cross-lagged coefficients in the RI-CLPM can be used to test whether within-person changes (i.e., deviations from the individual averages) in purpose across consecutive waves predict changes in health between the next consecutive waves, and vice versa (i.e., do within-person changes in health predict later changes in purpose). For example, if an individual shows an increase in sense of purpose compared to their usual levels, does this predict better health than usual four years later? In contrast, the CLPM cross-lags indicate if between-person differences (i.e., deviations from the group average) in purpose levels predict changes in health status across subsequent time points and vice versa. For example, are individuals with higher sense of purpose at a given time point more or less likely to show health declines in the next four years? It is worth noting that the CLPM reflects a mixture of between- and within-person effects in cases where both are present (Hamaker et al., 2015; Orth et al., 2024), making it hard to interpret CLPM results in isolation. However, comparing the two models is informative, as the extent to which the effect sizes in the CLPM are preserved when controlling for stable between-person differences in the RI-CLPM indicates whether the observed associations (if present) are better explained by between- or within-person effects. If the effects are absent in the RI-CLPM, but present in the CLPM, this implies that the associations are driven by between-person differences in stable trait levels. In this case, the cross-lagged effects would not necessarily translate to the within-person level but would rather indicate to what extent health and purpose can be considered as risk or protective factors for future development. Thus, for a more comprehensive analysis and comparison with previous work, we deem it appropriate to compare the results from both types of models. We estimated the models based on the longitudinal measurement model for purpose and the manifest indicators for health. Model fit was acceptable for all models (LGCMs: CFI = .938–.962; RMSEA = .030–.035; SRMR = .053–.054; CLPMs: CFI = .931–.959; RMSEA = .031–.034; SRMR = .055–.059; RI-CLPMs: CFI = .930–.960; RMSEA: .030–.034; SRMR - .053–.061).

Figure 1.

Path diagrams of the cross-lagged panel models as employed in the present study. Note. Panel A shows the bivariate LGCM, Panel B shows the CLPM with lag-2 associations, and Panel C shows the RI-CLPM. RI = random intercept; P = sense of purpose, HM = health marker. The 11 health variables were modeled separately. Paths a are autoregressive paths, paths c are cross-lagged paths, and w are within-time correlations (w2 and w3 represent residual correlations, i.e., correlated change). Control variables and the measurement model for purpose are not depicted.

We ran all models with and without controlling for age (in years), gender (binary), years of education, partner presence (yes vs. no), work status (currently working vs. not currently working), and HRS cohort (i.e., 2006 vs. 2008). Partner presence and work status were included as time-varying control variables to account for potential effects of the loss of the partner or retirement on purpose and health³. We constrained the autoregressive and cross-lagged effects, as well as the occasion-specific correlations to equality across time, to improve the statistical power and robustness of the estimates (Orth et al., 2024). This can be done if the time interval between measurements is equal and the effects are assumed to be similar between the different time points (Mulder & Hamaker, 2021). However, we also tested if the cross-lagged paths would differ across time by comparing a model with equality constraints across time to a model without. In the cases in which the Akaike Information Criterion (AIC), the Bayesian Information Criterion (BIC), and χ²-difference test suggested an increase in model misfit by constraining the cross-lags, we also compared the effects across time.

Results

Descriptive results

An overview of descriptive statistics and correlations for purpose and the health indicators at the first and last wave is provided in Table 2 (based on the largest dataset; N = 11,390). The health marker re-test correlations across 12 years were mostly moderate, with an average r = .52 (all p < .001), ranging from r = .20 for walking speed to r = .81 for BMI. The purpose re-test correlation was r = .52 (p < .001). Across the 12 years, all health markers except for BMI worsened to a small extent (average absolute d = 0.23; all t test p < .001), ranging from an absolute d = 0.14 for hearing to d = 0.42 for physical conditions. Sense of purpose declined by a very small degree across the 12 years (d = −0.09; p < .001). We found significant inter-individual differences in the long-term change (i.e., LGCM slope variance; all p < .001) and the occasion (i.e., LGCM occasion-specific residual variance; all p < .001) for purpose and all health markers.

Table 2.

Descriptive Statistics and Correlations.

Variable	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23
1. Purpose 1	4.70	0.90
2. Purpose 4	4.62	0.92	.52
3. SR health 1	3.35	1.03	.28	.26
4. SR health 4	3.16	0.99	.20	.29	.56
5. Vision 1	3.31	0.93	.20	.19	.41	.32
6. Vision 4	3.15	0.96	.16	.22	.34	.42	.43
7. Hearing 1	3.44	1.06	.14	.14	.28	.23	.29	.25
8. Hearing 4	3.29	1.03	.13	.19	.23	.29	.24	.36	.54
9. Walking speed 1^a	−3.27	1.68	.11	.05	.22	.13	.12	.09	.06	.07
10. Walking speed 4^a	−3.59	2.00	.12	.15	.23	.24	.18	.21	.09	.13	.20
11. Grip strength 1	32.77	11.04	.09	.11	.12	.08	.10	.09	−.07	−.05	.16	.20
12. Grip strength 4	28.63	12.02	.06	.10	.10	.11	.07	.09	−.02	.01	.12	.21	.65
13. BMI 1	28.57	5.83	−.09	−.08	−.22	−.19	−.06	−.06	−.04	−.04	−.09	−.09	.08	.01
14. BMI 4	28.59	6.04	−.08	−.06	−.20	−.17	−.06	−.04	−.04	−.02	−.06	−.04	.10	.09	.81
15. BMI-26 1	4.59	4.42	−.11	−.09	−.21	−.16	−.07	−.06	−.01	−.02	−.09	−.09	−.03	−.06	.78	.62
16. BMI-26 4	4.75	4.54	−.09	−.08	−.17	−.18	−.08	−.05	−.04	−.02	−.04	−.08	−.04	−.02	.59	.76	.73
17. Pain 1	0.59	0.93	−.15	−.16	−.41	−.27	−.17	−.13	−.09	−.07	−.13	−.16	−.13	−.11	.20	.17	.18	.16
18. Pain 4	0.74	1.00	−.10	−.16	−.29	−.37	−.13	−.16	−.10	−.12	−.04	−.17	−.08	−.08	.17	.17	.14	.16	.40
19. Conditions 1	1.81	1.26	−.15	−.17	−.43	−.34	−.19	−.17	−.16	−.14	−.14	−.19	−.15	−.16	.26	.18	.19	.14	.26	.23
20. Conditions 4	2.33	1.35	−.11	−.19	−.36	−.43	−.15	−.18	−.15	−.16	.06	−.17	−.06	−.15	.27	.20	.21	.16	.24	.27	.68
21. BADLs 1	0.19	0.67	−.14	−.11	−.34	−.21	−.17	−.13	−.10	−.08	−.27	−.18	−.10	−.08	.19	.14	.21	.15	.35	.20	.22	.17
22. BADLs 4	0.33	0.92	−.11	−.17	−.26	−.35	−.15	−.21	−.09	−.14	−.13	−.37	−.10	−.12	.15	.13	.13	.15	.21	.32	.20	.23	.38
23. IADLs 1	0.03	0.09	−.18	−.12	−.28	−.17	−.19	−.15	−.09	−.09	−.14	−.16	−.14	−.09	.09	.07	.11	.09	.22	.15	.16	.12	.43	.24
24. IADLs 4	0.05	0.13	−.16	−.22	−.27	−.32	−.19	−.28	−.10	−.17	−.14	−.32	−.17	−.17	.08	.04	.09	.08	.18	.23	.19	.20	.27	.54	.36

Note. SR health = self-reported health; BMI = body mass index; BMI-26 = absolute BMI difference from 26); B/IADLs = limitations in basic / instrumental activities of daily living. Label numbers indicate measurement waves (1 = first; 4 = last). Correlations are marked in bold if p < .001.

^aWalking speed is negative because it was reversed so that higher values (time in seconds) indicate faster speed.

Level and change correlations

To answer the first two research questions, whether purpose and health levels or changes are associated with each other, we used the LGCM. The level (i.e., intercept), change (i.e., slope), and occasion-specific (i.e., residual) correlations controlling for the covariates are presented in Figure 2 (see OSF Figure S1 for results without covariates). Purpose levels were associated with all health markers in the expected direction (average absolute ρ = .18; all p < .001), ranging in absolute size from ρ = −.09 for BMI to ρ = .34 for self-reported health. Regarding correlated change across the 12 years, we found very small to moderate significant associations with self-reported health (ρ = .21; p < .001), vision (ρ = .17; p = .007), and physical conditions (ρ = −.11; p = .002). The correlated change with BMI deviations from 26 (ρ = −.09; p = .030), hearing (ρ = .13; p = .020) and walking speed (ρ = .17; p = .028) was not significant when controlling for the FDR. Purpose deviations from the average level and linear change (i.e., occasion-specific residuals) were associated with deviations in self-reported health (ρ = .09; p < .001), physical conditions (ρ = −.04; p = .009), BADLs (ρ = −.10; p < .001) and IADLs (ρ = −.12; p < .001). Overall, we found significant correlations with all 11 health markers, three cases of correlated long-term change, and four cases of occasion-specific correlations.

Figure 2.

Level, Change and Occasion-specific Correlations. Note. SR health = self-reported health; BMI = body mass index; BMI-26 = absolute BMI difference from 26; B/IADLs = limitations in basic / instrumental activities of daily living. Presented are the latent growth curve model correlations with 95% confidence intervals.

Cross-lagged associations

We next examined the effects of purpose on later health, and vice versa, using the (random intercept) cross-lagged panel models. The standardized cross-lagged estimates are presented in Figure 3⁴ (for results without covariates, see OSF Figure S2). In the CLPM (i.e., a combination of between- and within-person effects), purpose predicted all health markers 4 years later in the expected direction with very small to moderate effects (average absolute β = .05). The associations ranged from an absolute β = .01 for BMI to β = .09 for self-rated health. In contrast, purpose was only predicted by 6 of the 11 health markers, with effect sizes being very small to small: self-rated health (β = .05), vision (β = .03), hearing (β = .03), walking speed (β = .03), grip strength (β = .04), and physical conditions (β = −.03). In 6 cases, the effect of purpose on later health was stronger than the other direction, namely, for vision (Δβ = .04; p < .001), hearing (Δβ = .03; p < .001), walking speed (Δβ = .04; p = .003), chronic pain (Δβ = −.04; p < .001, BADLs (Δβ = −.06; p < .001), and IADLs (Δβ = −.07; p < .001). Health did not have a stronger effect on purpose for any of the 11 associations. For self-rated health, grip strength, the BMI markers, and physical conditions, we found no difference in the effect size.

Figure 3.

Cross-lagged Estimates Between Purpose and Health. Note. (RI)CLPM = (random intercept) cross-lagged panel model; SR health = self-reported health; BMI = body mass index; BMI-26 = absolute BMI difference from 26; B/IADLs = limitations in basic / instrumental activities of daily living. Presented are standardized lag-1 cross-lag estimates with 95% confidence intervals.

In the RI-CLPM (i.e., within-person effects), many of the cross-lagged associations vanished, especially the effects from health to sense of purpose. This suggests that the between-person differences in purpose and health explained most associations found in the CLPM. Within-person deviations in purpose weakly predicted later deviations in hearing (β = .02; p = .022), walking speed (β = .05; p = .020), grip strength (β = .03; p < .001), BADLs (β = −.04; p < .001), and IADLs (β = −.05; p < .001)—all in the expected direction. Health deviations had a much smaller effect on later purpose, with only the effects of walking speed (β = .04; p = .007) and grip strength (β = .04; p < .001) being significant—again in the expected direction. In the RI-CLPM, the effect of purpose on IADLs was stronger than the other way around (Δβ = −.04; p = .003). For all other health markers, we found no differences between directions.

We tested whether the cross-lags would differ across time by comparing the (RI-)CLPMs with equality constraints across time to models with freely estimated parameters (see OSF Table S1 for results). For grip strength, the AIC, BIC and χ²-difference test suggested differences in the cross-lags in the RI-CLPM across time. Over time, the effects of purpose deviations on grip strength deviations increased slightly, whereas the effect of grip strength on later purpose deviations remained stable (see Figure 4).

Figure 4.

Grip Strength RI-CLPM Cross-lagged Estimates Across Time. Note. Presented are the standardized RI-CLPM cross-lagged estimates with 95% confidence intervals.

Exploratory analyses: Age-related differences

To examine potential age differences in the associations found, we ran multi-group models comparing late middle-aged (50–64 years old) to old adults in the sample (65+ years old at the first measurement occasion). The cross-lagged associations across the two age groups are presented in Figure 5 (for the LGCM associations, see OSF Figure S3). Contrary to our assumptions, no effects (i.e., level-level correlations, change-change correlations, occasion-specific correlations, and cross-lags) differed significantly between the age groups.

Figure 5.

Age Differences in Cross-lagged Estimates. Note. (RI)CLPM = (random intercept) cross-lagged panel model; m/o = middle-aged (50–64)/older adults (65+); SR health = self-reported health; BMI = body mass index; BMI-26 = absolute BMI difference from 26; B/IADLs = limitation in basic / instrumental activities of daily living. Presented are standardized cross-lag estimates with 95% confidence intervals. The RI-CLPM models for grip strength and walking speed had convergence issues in at least one age group and are thus not presented here. None of the observed differences between the two groups were significant.

Discussion

This study aimed to investigate how sense of purpose and health markers (both objective and subjective) affect each other over time in older adulthood. As predicted and in line with previous findings (Nakamura et al., 2022; Willroth et al., 2021; Zaslavsky et al., 2014; Zilioli et al., 2015), we found that purpose levels were associated with the health markers in the predicted directions, suggesting that individuals with higher levels of purpose also have better health and functional abilities in general. For some of the health markers, we also found significant change correlations in the expected directions, suggesting that increases or decreases in sense of purpose go hand in hand with better or worse health, respectively. We found these long-term change associations across the 12 years covered in this study for self-reported health, vision, hearing, walking speed, BMI differences from 26 (absolute difference to a suggested ideal BMI for older adults [Winter et al., 2014]), and physical conditions. In addition to these long-term change associations, we also found that participants reported a weaker sense of purpose in the years in which they also reported worse self-rated health, more physical conditions or limitations with activities of daily living (both basic and instrumental; B/IADLs). In terms of directional associations, we generally found more robust evidence for sense of purpose as a predictor of subsequent health compared to health as a predictor of later sense of purpose. However, many of the associations vanished when controlling for stable between-person differences, suggesting that they do not necessarily hold at the within-person level. This was especially the case for the health markers as predictors of later purpose, suggesting that individuals experiencing worsening health might not experience subsequent decreases in sense of purpose four years later. There were no differences in the associations when comparing late middle-aged (50–64 years old) to old adults (65+ years old). When describing the results for aims 1–4, we thus use the term “older adults” referring to the entire sample. We discuss the findings and their implications in more detail below.

Level and correlated change between purpose and health

In line with past cross-sectional studies (AshaRani et al., 2022; Ottenbacher et al., 2007; Zaslavsky et al., 2014), we consistently found that older adults with higher scores on sense of purpose also had better health as measured by all of the individual health markers investigated in this study. The effect sizes ranged from small to large (|.09–.34|; Funder & Ozer, 2019; Gignac & Szodorai, 2016) across the health variables, with the largest associations observed for self-rated health, vision, chronic pain, and activities of daily living. The weakest association was for BMI. The observed link between all health indices and purpose might be explained by individual differences in health behaviors, lifestyle choices, or stress levels that accumulate over time and maintain such between-person differences (de Oliveira et al., 2020; Sutin et al., 2022).

In terms of long-term correlated change, sense of purpose changes were significantly correlated with changes in 6 out of the 11 health markers, all in the predicted directions, suggesting that sense of purpose changes in tandem with certain aspects of health (i.e., when sense of purpose declines, health also declines). These change correlations can represent reciprocal associations between purpose and health (e.g., a lasting health decline leading to a decline in purpose, and vice versa) but also the influence of other changes in aspects of life that occur with age, such as changes in one’s social network or social involvement (both being linked to health and purpose; Hill, Olaru, & Allemand, 2023; Nakamura et al., 2022; Pfund, Hofer, et al., 2022). Alternatively, major life transitions taking place during older adulthood (i.e., retirement, loss of a partner, relocation, and transition to grandparenthood) might drive changes in both constructs (Chereches et al., 2024; Lee et al., 2022; Yemiscigil et al., 2021).

An interesting finding from the LGCM is that the limitations in activities of daily living did not show a significant long-term correlated change with purpose but the strongest occasion-specific associations. These correlations represent shorter-term change correlations which do not last or accumulate over time. A potential reason might be that the experience of these limitations restricts people’s ability to pursue their current purpose around the time these limitations become apparent, but that they find ways to compensate for these limitations or focus on other (purposeful) activities over time (Baltes & Baltes, 1990). This also aligns with the lack of cross-lagged effects of the limitations on later purpose, which we discuss in more detail below.

Between-person associations between sense of purpose and health

To gauge whether between-person differences in purpose predict between-person differences in health across time, and vice versa, we performed cross-lagged panel analyses. Here, sense of purpose predicted all health markers to a small to moderate degree. For all health markers, the associations were in the expected direction, implying that individuals with a higher sense of purpose at one time point were likely to have a better health status at subsequent time points, controlling for initial levels of these variables.

Regarding the reverse association of health on sense of purpose, the cross-lagged associations were significant for 6 out of the 11 health markers we investigated, all in the expected direction. This implies that certain aspects of health are predictive of a later sense of purpose, which is consistent with some previous findings (Lewis et al., 2020; although see Chen et al., 2020, who did not find this link). Furthermore, this finding lends support to the idea that good health can serve as a resource (i.e., a protective factor) for sense of purpose. Healthier adults may be better able to engage in purposeful activities and hold onto social roles later in life compared to those with worse health.

Our findings mostly converge with the results of previous studies using subsets (i.e., three instead of four waves) of the HRS data to assess the (between-person) predictive effects of physical health markers on purpose (Nakamura et al., 2022) and the predictive effects of purpose on physical health markers (Kim et al., 2022) using a different analytic technique (lagged exposure wide approach). One exception is that, unlike Nakamura et al. (2022), we did not find a predictive association between limitations in (instrumental or basic) functioning when modeled together with the reverse association. Instead, our findings support the reverse direction, namely, that individuals with lower sense of purpose tend to show increases in functional limitations at later time points, in line with the findings of Kim et al. (2022)⁵. Further, in contrast to the non-significant effect observed by Kim et al. (2022), we found a negative association between purpose and later chronic pain (but only in the CLPM model). Although the finding that health also predicts later sense of purpose suggests that the associations are bidirectional for many of the health markers, sense of purpose generally appears as a stronger predictor of later health rather than vice versa (with the exception that grip strength predicted later sense of purpose more strongly than the other way around). This aligns with the finding of Yemiscigil and Vlaev (2021), who tentatively concluded that the effect of purpose on physical activity (an indirect measure of health) may be slightly stronger (0.04–0.06 standard deviations) than the reverse effect of physical activity and purpose (note however that the difference was not significant in said study). To summarize, it thus appears that (a) individuals who maintain their sense of purpose or at least show smaller decreases as they age also show smaller subsequent declines in health and functional abilities on average; (b) individuals who show less decline in health are also more likely to hold onto their sense of purpose; and (c) sense of purpose is a more robust predictor of later health decline compared to health decline as a predictor of later sense of purpose. Although these results are informative for predicting for whom change in purpose and health occurs, it is important to note that these results cannot be directly translated to the individual level (or inferred as reflecting explanatory processes), as they are based on models that do not control for stable between-person differences (Hamaker et al., 2015).

Within-person dynamics between sense of purpose and health

To investigate whether the associations also represent within-person processes wherein changes in health predict subsequent changes in sense of purpose or vice versa, we ran the same models again with an added random intercept. The results showed that many associations were close to zero at the within-person level. Therefore, the cross-lagged findings from our analyses (described above), and potentially similar associations reported in the literature (e.g., Kim et al., 2022; Nakamura et al., 2022) seem to be largely explained by some stable between-person differences leading purposeful individuals to experience better health later on and vice versa. Nevertheless, within-person changes in sense of purpose predicted changes in some of the health markers at a subsequent time point (five of the eleven effects observed in CLPM were replicated with the RI-CLPM), as well as some of the reverse associations, albeit fewer (two out of the six effects observed with the CLPM). More specifically, larger decreases in sense of purpose preceded larger subsequent within-person declines in hearing, walking speed, grip strength, and activities of daily living (B/IADLs). Further, when individuals showed declines in walking speed or grip strength, they also generally had subsequent declines in sense of purpose.

Exploratory age moderation

The results of the exploratory age moderation, wherein we compared the results (aims 1–4) between late middle-aged (50–64 years old) and old adults (65+), did not suggest any clear differences across the groups. Apart from the non-significance of the difference tests, the cross-lagged effects were also highly similar between age groups, emphasizing the similarities in the bidirectional relationships across the age groups compared here. Because of power considerations and the composition of the HRS sample, we split the sample around the age of retirement. However, with more advanced age (e.g., 85+ years) these effects might start to change—as declines in health and ability generally become steeper with older age (Stenholm et al., 2015), perceptions of health change (Wurm et al., 2008), and future time perspective narrows (Pfund, Ratner, et al., 2022).

Implications for healthy aging

Notably, the two objectively measured health markers (i.e., grip strength and walking speed measured by trained professionals) were most robust against controlling for stable between-person differences, with the estimates being comparable in size between the two models. In both cases (i.e., CLPM and RI-CLPM) and both directions (i.e., purpose and later health; health and later purpose), the associations held. This suggests that interventions targeting adults’ sense of purpose (e.g., Schippers & Ziegler, 2019) may bring health benefits down the line (however, this would need to be tested using an intervention design). In contrast, when considering the subjective (i.e., self-reported) health markers, most of the associations vanished or decreased in size when controlling for between-person differences. Firstly, this may suggest that subjectively experienced health status (e.g., self-reported health and chronic pain) is not very sensitive to changes in sense of purpose, despite changes in objective health (i.e., hearing, walking speed, grip strength, and activities of daily living). This could be due to purpose declines co-occurring with reduced activity, ultimately leading to a loss in function, but without much immediate change in other aspects of health. Secondly, sense of purpose may be relatively robust against changes in subjective health status but still influenced by actual declines in certain areas of functioning. The two markers that predicted later sense of purpose were slower walking speed and grip strength, which might indicate fatigue and onset of health- and cognitive decline (both are robust predictors of later health and cognitive functioning, see Leong et al., 2015; Purser et al., 2005; Skillbäck et al., 2022). This, in turn, could influence people’s motivation and ability to engage in purposeful activities before actual health decline becomes more pronounced. Notably, the findings do not need to imply that this discrepancy generally occurs at the within-person level. Future research should aim to replicate and investigate the sources of this mismatch, evaluating the specific role of subjective health appraisals in purpose-health associations and whether sense of purpose influences these appraisals, independent of individuals' objective health.

Another interesting set of results we observed was the very small overall decline in sense of purpose across the 12 years covered in this study (d = −0.09) as compared to the extent of decline in some of the health markers (e.g., physical conditions: d = 0.42) and the lack of intra-individual declines in purpose following health decline. These findings suggest that sense of purpose may be somewhat resistant to changes in health and disability that occur with age, hinting that older adults may show flexibility in readjusting their sense of purpose when faced with health decline. Some earlier work has found similar results: Lewis et al. (2020) report changes in sense of purpose after stroke onset when comparing purpose to pre-stroke baseline, but no declines in purpose if stroke took place before baseline measurement and Hill et al. (2021) report no changes in sense of purpose following health adversity. Apart from findings on purpose changes following changes in health status, it has also been observed that subjective well-being predicts later purpose to a smaller extent than the other way around (Gudmundsdottir et al., 2023). This may indicate that older adults generally adapt to the challenges of increasing age in a way that they are still motivated to seek out meaningful life goals, perhaps by adjusting their aims or searching for a source of purpose in other areas of life. Such an effect would be in line with lifespan theories, such as the Selection, Optimization, and Compensation (SOC) model of Baltes & Baltes (1990). The SOC model suggests that individuals are able to adapt to health-related declines by abandoning goals or tasks that might be health-threatening, selecting and prioritizing goals or personal motives that are better in line with one’s current health situation, or using alternative means or resources to reach a goal (e.g., after health declines, a high sense of purpose might be fostered by other life domains such as close relationships with friends and family; Chen et al., 2020). It is also possible that, with increasing age, a certain level of physical discomfort or health impairments might be regarded as a “part of life,” possibly being perceived as less influential for sense of purpose (Karwetzky et al., 2021).

A relevant direction for future research on purpose could thus be to also take the actual purpose(s) of individuals (i.e., the content; Burrow et al., 2021) into account and examine if older adults adapt to changes in their working status, health, and social network by readjusting their overall life orientation (e.g., focusing more on generativity) or only the expectations regarding what they still want to achieve (e.g., focusing their prosocial orientation more on the immediate family instead of a larger group of people). This could help study whether some life orientations are more resilient to age-associated declines and whether a strong commitment to a few specific purposes (e.g., economic status) could come at a cost. Although a stronger purpose commitment has been shown to have positive effects among adolescents and emerging adults (e.g., better subjective well-being; Bronk, 2011; Hill et al., 2016; Hill et al., 2016), it is unclear whether these findings also apply to older adulthood. Considering the losses older adults face in their social network, work life, and health, it would be relevant to examine how flexible they are in funneling their drive for purpose and adjusting to changed circumstances. Indeed, it has been proposed that such psychological flexibility may be instrumental in determining if people can continue pursuing valued goals in spite of setbacks and distress (Kashdan & Rottenberg, 2010) and may, as such, be a key research target to uncover which individuals are most at risk of experiencing declines in sense of purpose.

It should be noted that although health declines did not emerge as robust predictors of subsequent intra-individual declines in purpose, the current and previous findings (Hill, Olaru, & Allemand, 2023) suggest significant individual differences in purpose trajectories. Since health status failed to predict individual fluctuations in purpose over time (as also concluded by Hill, Pfund, & Allemand, 2023), future research should also consider alternative explanations for individual differences in purpose trajectories beyond health or functional abilities.

Limitations and future directions

Some limitations of the current research warrant discussion. First, most health markers were self-reported. Third variables linked to the evaluation of both purpose and health (e.g., subjective well-being) might have driven some of the associations. Even the arguably more objective marker of diagnosed chronic diseases depends on whether respondents go to the doctor regularly to be diagnosed. Nonetheless, we found the most robust associations across all models for the two objectively measured health markers (i.e., grip strength and walking speed), suggesting some relevant links between purpose and health across time. Furthermore, the evaluation of one’s health can still be considered influential for how people structure their daily lives, for instance, only focusing on a restricted range of relatively “safe” behaviors versus engaging in a wider range of behaviors that can provide them with a sense of purpose.

Second, we may have underestimated individual differences in change trajectories and subsequently some of the purpose-health associations due to stability in the constructs. Although there was significant variability in all health markers, some were relatively more stable (e.g., BMI). This may have played a role in the lack of association between purpose and some of the health markers and may also bias comparison between the specific markers. For example, although weight changes (BMI) may be less important for sense of purpose compared to hearing and vision, it is possible that the relative stability of people’s weight in general did not allow picking up a potentially meaningful association. Relatedly, it may take larger changes in weight to bring about significant change in seeking out purposeful activities, and it may take longer for consequences of this weight gain to accumulate and bring about behavioral or biological changes linked to reduced ability or motivation for purposeful pursuits.

Third, and relatedly, the time lag between purpose assessments was four years which restricts the chance of detecting faster as well as slower effects of health change on sense of purpose and vice versa. In particular, changes in sense of purpose may be more immediate following (certain types of) health declines, with individuals potentially returning to baseline levels after some time or readjusting to their current health status (for a discussion on personality change and timing, see Bühler et al., 2024). We were primarily interested in identifying associations that persist across multiple assessment points, capturing longer term trends rather than short-lived ones. Identifying sustained associations is important, as they can imply the potential long-term impact of interventions aimed at lasting improvements in purpose. Intervention research is, of course, needed to confirm the effects, but longitudinal studies can help guide such research efforts. Although we expect that meaningful and sustained changes in purpose and health generally occur at a slow rate and that the associations between the two are a result of accumulated effects over time, we acknowledge that this may, especially for some health markers, occur faster than four years. The general lack of within-person cross-lagged associations may be explained by the two declining more simultaneously rather than sequentially, in particular for health changes predicting purpose changes. Establishing the associations across shorter intervals (provided there is sufficient change) would provide stronger support for a directional relationship and is an important next step to replicate the current findings and to explore potential mechanisms. Moving forward, future research should thus aim to investigate these associations over shorter timeframes and consider potential mechanisms explaining the links, such as social integration, engagement in health behavior and perceived emotional stress. To capture these mechanisms effectively, and determine the timeframe within which they operate, more frequent assessments are likely needed. Importantly, different health markers and the specific direction of association under study (i.e., health to purpose vs. purpose to health) may require measurement at different time scales.

Fourth, we were unable to assess what the observed health changes meant for participants (e.g., if they were unexpected or perceived as tragic), as well as their severity, since changes in sense of purpose might only follow more extreme and unpredictable health experiences (e.g., diagnosis of a chronic disease, severe loss of mobility; see Murphy & Bastian, 2020 for a similar discussion for meaning in life). It is also possible that only the accumulation of certain life events in combination with health change leads to changes in sense of purpose. Late adulthood represents a period of multiple role transitions (i.e., both role entries and role exits such as transition to grandparenthood, retirement, and loss of a loved one). Under certain circumstances, when decrements in health are coupled with role losses, we might observe changes in sense of purpose (for a similar discussion on the coupling of life events on changes in life satisfaction, see Krämer et al., 2023). For example, declines in purpose following health decline might occur only when adults also lose social roles that could have otherwise acted as compensation (e.g., loss of a partner and loss of job-related roles), or these declines might be less likely to occur if adults enter a new social role around the same time (e.g., volunteering and grandparenthood). Future research should thus aim to investigate how purpose may change in response to health changes that occur in tandem with other life events specific to older age.

Finally, work on sense of purpose should also be contrasted with work on meaning in life. Within research, the two constructs are often used interchangeably. Yet, efforts have been made to distinguish them, with meaning in life emerging as a broader construct that encompasses purpose in life as one component (Costin & Vignoles, 2020; Martela & Steger, 2016). More specifically, purpose is related to motivated planning and acting (i.e., having a direction in life) and is predominantly future-oriented, while meaning is more related to processing and interpreting reality, including the past, present and the future (George & Park, 2017; Martela & Steger, 2016). Like purpose in life, a higher meaning in life has been linked to better subsequent health outcomes, as well as the other way around, at least for some health markers (i.e., limitations in activities of daily living; Weziak-Bialowolska & Bialowolski, 2022). Future research should examine both aspects of meaning in life in more detail in relation to health and also assess if the link between meaning and health is driven mainly by the future-oriented component of meaning in life (i.e., a sense of purpose and direction) or by the cognitive and evaluative component, namely, coherence and significance (Martela & Steger, 2016).

Conclusion

This study investigated the reciprocal links between sense of purpose and health using both objective and subjective markers of health. We found evidence for between-person reciprocal associations between sense of purpose and most health markers, although many of these associations vanished when controlling for stable levels of purpose and health across the 12 years covered in this study. Yet, some associations were still observed at the within-person level, and the most consistent evidence we found across health markers was for changes in sense of purpose predicting subsequent changes in health rather than the reverse (i.e., health changes predicting changes in sense of purpose). With this finding, we add to the emerging body of evidence highlighting that sense of purpose is a predictor of long-term health, and we also showed that intra-individual changes in purpose predict subsequent changes in some aspects of health (most notably objective markers thereof). However, further research into the mechanisms that underly these associations is needed to better understand the potential causal link.

Supplemental Material

Supplemental Material - Linking sense of purpose and multiple markers of health in older adulthood: A bidirectional approach

Supplemental Material for Linking sense of purpose and multiple markers of health in older adulthood: A bidirectional approach by Flavia S. Chereches, Gudrun R. Gudmundsdottir, and Gabriel Olaru in European Journal of Personality.

Supplemental Material

Supplemental Material - Linking sense of purpose and multiple markers of health in older adulthood: A bidirectional approach

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Open Science Statement

The datasets used in this study are publicly available upon registration, free of charge, at https://hrs.isr.umich.edu/data-products. The pre-registration, analysis scripts and supplementary materials can be accessed in an OSF repository

ORCID iDs

Flavia S. Chereches

Gudrun R. Gudmundsdottir

Gabriel Olaru

Supplemental Material

Supplemental material for this article is available online.

Notes

References

AshaRani

Lai

Koh

Subramaniam

(2022). Purpose in life in older adults: A systematic review on conceptualization, measures, and determinants. International Journal of Environmental Research and Public Health, 19(10), 5860. https://doi.org/10.3390/ijerph19105860

Baltes

P. B.

Baltes

M. M.

(1990). Psychological perspectives on successful aging: The model of selective optimization with compensation. In Successful aging: Perspectives from the behavioral sciences (pp. 1–34). Cambridge University Press. https://doi.org/10.1017/CBO9780511665684.003

Baltes

P. B.

Smith

(2003). New frontiers in the future of aging: From successful aging of the young old to the dilemmas of the fourth age. Gerontology, 49(2), 123–135. https://doi.org/10.1159/000067946

Bath

P. A.

Deeg

(2005). Social engagement and health outcomes among older people: Introduction to a special section. European Journal of Ageing, 2(1), 24–30. https://doi.org/10.1007/s10433-005-0019-4

Benjamini

Hochberg

(1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B, 57(1), 289–300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x

Bentler

P. M.

Bonett

D. G.

(1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88(3), 588–606. https://doi.org/10.1037/0033-2909.88.3.588

Boyle

P. A.

Barnes

L. L.

Buchman

A. S.

Bennett

D. A.

(2009). Purpose in life is associated with mortality among community-dwelling older persons. Psychosomatic Medicine, 71(5), 574–579. https://doi.org/10.1097/PSY.0b013e3181a5a7c0

Bronk

K. C.

(2011). The role of purpose in life in healthy identity formation: A grounded model. New Directions for Youth Development, 2011(132), 31–38. https://doi.org/10.1002/yd.426

Bühler

J. L.

Orth

Bleidorn

Weber

Kretzschmar

Scheling

Hopwood

C. J.

(2024). Life events and personality change: A systematic review and meta-analysis. European Journal of Personality, 38(3), 544–568. https://doi.org/10.1177/08902070231190219

10.

Burrow

A. L.

Agans

J. P.

Jeon

H. J.

Creim

(2021). Are all purposes worth having? Integrating content and strength in purpose research. Human Development, 65(2), 100–112. https://doi.org/10.1159/000515176

11.

Chen

F. F.

(2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling. A Multidisciplinary Journal, 14(3), 464–504. https://doi.org/10.1080/10705510701301834

12.

Chen

Kim

E. S.

Shields

A. E.

VanderWeele

T. J.

(2020). Antecedents of purpose in life: Evidence from a lagged exposure-wide analysis. Cogent psychology, 7(1), Article 1825043. https://doi.org/10.1080/23311908.2020.1825043

13.

Chereches

F. S.

Ballhausen

Brehmer

Olaru

(2024). Sense of purpose and meaning in life during the transition to grandparenthood. European Journal of Personality. https://doi.org/10.1177/08902070241257667

14.

Chereches

F. S.

Brehmer

Olaru

(2023). Personality and limitations in instrumental activities of daily living in old age: Reciprocal associations across 12 years. European Journal of Personality, 37(5), 543–559. https://doi.org/10.1177/08902070221111856

15.

Cherry

K. E.

Walker

E. J.

Brown

J. S.

Volaufova

LaMotte

L. R.

Welsh

D. A.

L. J.

Jazwinski

S. M.

Ellis

Wood

R. H.

Frisard

M. I.

(2013). Social engagement and health in younger, older, and oldest-old adults in the Louisiana Healthy Aging Study. Journal of Applied Gerontology: The Official Journal of the Southern Gerontological Society, 32(1), 51–75. https://doi.org/10.1177/0733464811409034

16.

Connolly

Garvey

McKee

(2017). Factors associated with ADL/IADL disability in community dwelling older adults in the Irish longitudinal study on ageing (TILDA). Disability & Rehabilitation, 39(8), 809–816. https://doi.org/10.3109/09638288.2016.1161848

17.

Costin

Vignoles

V. L.

(2020). Meaning is about mattering: Evaluating coherence, purpose, and existential mattering as precursors of meaning in life judgments. Journal of Personality and Social Psychology, 118(4), 864–884. https://doi.org/10.1037/pspp0000225

18.

de Oliveira

D. V.

do Nascimento

M. A.

Branco

B. H. M.

de Oliveira

R. V.

do Nascimento Júnior

J. R. A.

Freire

G. L. M.

Bertolini

S. M. M. G.

(2020). Psychological factors predicting sedentary behavior of older adults: A cross-sectional study in Brazil. Journal of Aging and Physical Activity, 29(2), 280–287. https://doi.org/10.1123/japa.2020-0051

19.

Dohlman

C. S.

Zalla

L. C.

Chung

E. O.

Østbye

Maselko

(2023). Generativity and engagement in grandparenting activities among older adults in northern Sri Lanka. International journal of aging & human development, 97(2), 249–262. https://doi.org/10.1177/00914150221143955

20.

Elran-Barak

Weinstein

Beeri

M. S.

Ravona-Springer

(2019). The associations between objective and subjective health among older adults with type 2 diabetes: The moderating role of personality. Journal of Psychosomatic Research, 117, 41–47. https://doi.org/10.1016/j.jpsychores.2018.12.011

21.

Emerson

Boggero

Ostir

Jayawardhana

(2018). Pain as a risk factor for loneliness among older adults. Journal of Aging and Health, 30(9), 1450–1461. https://doi.org/10.1177/0898264317721348

22.

Funder

D. C.

Ozer

D. J.

(2019). Evaluating effect size in psychological research: Sense and nonsense. Advances in Methods and Practices in Psychological Science, 2(2), 156–168. https://doi.org/10.1177/2515245919847202

23.

Galenkamp

Huisman

Braam

A. W.

Deeg

D. J.

(2012). Estimates of prospective change in self-rated health in older people were biased owing to potential recalibration response shift. Journal of Clinical Epidemiology, 65(9), 978–988. https://doi.org/10.1016/j.jclinepi.2012.03.010

24.

George

L. S.

Park

C. L.

(2017). The Multidimensional Existential Meaning Scale: A tripartite approach to measuring meaning in life. The Journal of Positive Psychology, 12(6), 613–627. https://doi.org/10.1080/17439760.2016.1209546

25.

Gignac

G. E.

Szodorai

E. T.

(2016). Effect size guidelines for individual differences researchers. Personality and Individual Differences, 102, 74–78. https://doi.org/10.1016/j.paid.2016.06.069

26.

Goodwin

Engstrom

(2002). Personality and the perception of health in the general population. Psychological Medicine, 32(2), 325–332. https://doi.org/10.1017/S0033291701005104

27.

Gottlieb

B. H.

Gillespie

A. A.

(2008). Volunteerism, health, and civic engagement among older adults. Canadian journal on aging = La revue canadienne du vieillissement, 27(4), 399–406. https://doi.org/10.3138/cja.27.4.399

28.

Gudmundsdottir

G. R.

Pfund

G. N.

Hill

P. L.

Olaru

(2023). Reciprocal associations between sense of purpose and subjective well-being in old age. European Journal of Personality, 38(3), 391–404. https://doi.org/10.1177/08902070231176702

29.

Hafez

Heisler

Choi

Ankuda

C. K.

Winkelman

Kullgren

J. T.

(2018). Association between purpose in life and glucose control among older adults. Annals of Behavioral Medicine, 52(4), 309–318. https://doi.org/10.1093/abm/kax012

30.

Hamaker

E. L.

Kuiper

R. M.

Grasman

R. P.

(2015). A critique of the cross-lagged panel model. Psychological Methods, 20(1), 102–116. https://doi.org/10.1037/a0038889

31.

Hamaker

E. L.

Mulder

J. D.

van Ijzendoorn

M. H.

(2020). Description, prediction and causation: Methodological challenges of studying child and adolescent development. Developmental cognitive neuroscience, 46, Article 100867. https://doi.org/10.1016/j.dcn.2020.100867

32.

Hill

P. L.

Beck

E. D.

Jackson

J. J.

(2021). Maintaining sense of purpose following health adversity in older adulthood: A propensity score matching examination. The Journals of Gerontology: Serie Bibliographique, 76(8), 1574–1579. https://doi.org/10.1093/geronb/gbab002

33.

Hill

P. L.

Edmonds

G. W.

Peterson

Luyckx

Andrews

J. A.

(2016). Purpose in life in emerging adulthood: Development and validation of a new brief measure. The Journal of Positive Psychology, 11(3), 237–245. https://doi.org/10.1080/17439760.2015.1048817

34.

Hill

P. L.

Olaru

Allemand

(2023). Do associations between sense of purpose, social support, and loneliness differ across the adult lifespan? Psychology and Aging, 38(4), 345–355. https://doi.org/10.1037/pag0000733

35.

Hill

P. L.

Pfund

G. N.

Allemand

(2023). The paths to purpose: A new framework toward understanding purpose development. Current Directions in Psychological Science, 32(2), 105–110. https://doi.org/10.1177/09637214221128019

36.

Hill

P. L.

Weston

S. J.

(2019). Evaluating eight-year trajectories for sense of purpose in the health and retirement study. Aging & Mental Health, 23(2), 233–237. https://doi.org/10.1080/13607863.2017.1399344

37.

Hooker

S. A.

Masters

K. S.

(2016). Purpose in life is associated with physical activity measured by accelerometer. Journal of Health Psychology, 21(6), 962–971. https://doi.org/10.1177/1359105314542822

38.

Huisman

Deeg

D. J.

(2010). A commentary on marja jylhä's “what is self-rated health and why does it predict mortality? Towards a unified conceptual model”(69: 3, 2009, 307–316). Social Science & Medicine, 70(5), 652–657. https://doi.org/10.1016/j.socscimed.2009.11.003

39.

Irving

Davis

Collier

(2017). Aging with purpose: Systematic search and review of literature pertaining to older adults and purpose. The International Journal of Aging and Human Development, 85(4), 403–437. https://doi.org/10.1177/0091415017702908

40.

Jorgensen

T. D.

Pornprasertmanit

Schoemann

A. M.

Rosseel

(2025). semTools: Useful tools for structural equation modeling. R package version 0.5-7. https://CRAN.R-project.org/package=semTools

41.

Jylhä

(2009). What is self-rated health and why does it predict mortality? Towards a unified conceptual model. Social Science & Medicine, 69(3), 307–316. https://doi.org/10.1016/j.socscimed.2009.05.013

42.

Kail

B. L.

Carr

D. C.

(2020). More than selection effects: Volunteering is associated with benefits in cognitive functioning. The Journals of Gerontology: Serie Bibliographique, 75(8), 1741–1746. https://doi.org/10.1093/geronb/gbaa101

43.

Karwetzky

Werdecker

Esch

(2021). What matters most in life? A German cohort study on the sources of meaning and their neurobiological foundations in four age groups. Frontiers in Psychology, 12, Article 777751. https://doi.org/10.3389/fpsyg.2021.777751

44.

Kashdan

T. B.

Rottenberg

(2010). Psychological flexibility as a fundamental aspect of health. Clinical Psychology Review, 30(7), 865–878. https://doi.org/10.1016/j.cpr.2010.03.001

45.

Kenny

D. A.

(2020). Measuring model fit. https://www.davidakenny.net/cm/fit.htm

46.

Kim

E. S.

Chen

Nakamura

J. S.

Ryff

C. D.

VanderWeele

T. J.

(2022). Sense of purpose in life and subsequent physical, behavioral, and psychosocial health: An outcome-wide approach. American Journal of Health Promotion, 36(1), 137–147. https://doi.org/10.1177/08901171211038545

47.

Kim

E. S.

Delaney

S. W.

Kubzansky

L. D.

(2019). Sense of purpose in life and cardiovascular disease: Underlying mechanisms and future directions. Current Cardiology Reports, 21, 135–211. https://doi.org/10.1007/s11886-019-1222-9

48.

Kim

E. S.

Hershner

S. D.

Strecher

V. J.

(2015). Purpose in life and incidence of sleep disturbances. Journal of Behavioral Medicine, 38(3), 590–597. https://doi.org/10.1007/s10865-015-9635-4

49.

Kim

E. S.

Kawachi

Chen

Kubzansky

L. D.

(2017). Association between purpose in life and objective measures of physical function in older adults. JAMA Psychiatry, 74(10), 1039–1045. https://doi.org/10.1001/jamapsychiatry.2017.2145

50.

Kim

E. S.

Shiba

Boehm

J. K.

Kubzansky

L. D.

(2020). Sense of purpose in life and five health behaviors in older adults. Preventive Medicine, 139, Article 106172. https://doi.org/10.1016/j.ypmed.2020.106172

51.

Kim

E. S.

Strecher

V. J.

Ryff

C. D.

(2014). Purpose in life and use of preventive health care services. Proceedings of the National Academy of Sciences, 111(46), 16331–16336. https://doi.org/10.1073/pnas.1414826111

52.

Krämer

M. D.

Rohrer

J. M.

Lucas

R. E.

Richter

(2023). Life events and life satisfaction: Estimating effects of multiple life events in combined models. European Journal of Personality, 39(1). https://doi.org/10.1177/08902070241231017

53.

Lee

Sain

Harari

Kürüm

(2022). Social mobility and sense of purpose from midlife to old age: Examining the role of major life events. Research on Aging, 45(7-8), 550–562. https://doi.org/10.1177/01640275221121588

54.

Leong

D. P.

Teo

K. K.

Rangarajan

Lopez-Jaramillo

Avezum

Orlandini

Seron

Ahmed

S. H.

Rosengren

Kelishadi

Rahman

Swaminathan

Iqbal

Gupta

Lear

S. A.

Oguz

Yusoff

Zatonska

Chifamba

… Yusuf

(2015). Prognostic value of grip strength: Findings from the prospective urban rural epidemiology (PURE) study. The lancet, 386(9990), 266–273. https://doi.org/10.1016/S0140-6736(14)62000-6

55.

Lewis

N. A.

(2020). Purpose in life as a guiding framework for goal engagement and motivation. Social and Personality Psychology Compass, 14(10), 1–11. https://doi.org/10.1111/spc3.12567

56.

Lewis

N. A.

Brazeau

Hill

P. L.

(2020). Adjusting after stroke: Changes in sense of purpose in life and the role of social support, relationship strain, and time. Journal of Health Psychology, 25(12), 1831–1841. https://doi.org/10.1177/1359105318772656

57.

Lewis

N. A.

Hill

P. L.

(2020). Does being active mean being purposeful in older adulthood? Examining the moderating role of retirement. Psychology and Aging, 35(7), 1050–1057. https://doi.org/10.1037/pag0000568

58.

Lewis

N. A.

Hill

P. L.

(2021). Sense of purpose promotes resilience to cognitive deficits attributable to depressive symptoms. Frontiers in Psychology, 12, Article 698109. https://doi.org/10.3389/fpsyg.2021.698109

59.

Lewis

N. A.

Turiano

N. A.

Payne

B. R.

Hill

P. L.

(2017). Purpose in life and cognitive functioning in adulthood. Aging, Neuropsychology, and Cognition, 24(6), 662–671.

60.

Löckenhoff

C. E.

Sutin

A. R.

Ferrucci

Costa Jr

P. T.

(2008). Personality traits and subjective health in the later years: The association between NEO-PI-R and SF-36 in advanced age is influenced by health status. Journal of Research in Personality, 42(5), 1334–1346.

61.

Lowsky

D. J.

Olshansky

S. J.

Bhattacharya

Goldman

D. P.

(2014). Heterogeneity in healthy aging. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences, 69(6), 640–649. https://doi.org/10.1093/gerona/glt162

62.

Lüdtke

Robitzsch

(2022). A comparison of different approaches for estimating cross-lagged effects from a causal inference perspective. Structural Equation Modeling, 1–20. https://doi.org/10.1080/10705511.2022.2065278

63.

Martela

Laitinen

Hakulinen

(2024). Which predicts longevity better: Satisfaction with life or purpose in life? Psychology and Aging, 39(6), 589–598. https://doi.org/10.1037/pag0000802

64.

Martela

Steger

M. F.

(2016). The three meanings of meaning in life: Distinguishing coherence, purpose, and significance. The Journal of Positive Psychology, 11(5), 531–545. https://doi.org/10.1080/17439760.2015.1137623

65.

Martela

Steger

M. F.

(2023). The role of significance relative to the other dimensions of meaning in life–an examination utilizing the three dimensional meaning in life scale (3DM). The Journal of Positive Psychology, 18(4), 606–626. https://doi.org/10.1080/17439760.2022.2070528

66.

Meek

K. P.

Bergeron

C. D.

Towne

S. D.

Ahn

Ory

M. G.

Smith

M. L.

(2018). Restricted social engagement among adults living with chronic conditions. International Journal of Environmental Research and Public Health, 15(1), 158. https://doi.org/10.3390/ijerph15010158

67.

Mei

Lori

Vattathil

S. M.

Boyle

P. A.

Bradley

Jin

Bennett

D. A.

Wingo

T. S.

Wingo

A. P.

(2021). Important correlates of purpose in life identified through a machine learning approach. American Journal of Geriatric Psychiatry: Official Journal of the American Association for Geriatric Psychiatry, 29(5), 488–498. https://doi.org/10.1016/j.jagp.2020.09.018

68.

Merten

Pinto

A. A.

Paulsen

A. J.

Chen

Dillard

L. K.

Fischer

M. E.

Ryff

C. D.

Schubert

C. R.

Cruickshanks

K. J.

(2022). The association of psychological well-being with sensory and cognitive function and neuronal health in aging adults. Journal of Aging and Health, 34(4-5), 529–538. https://doi.org/10.1177/08982643211046820

69.

Mroczek

D. K.

Weston

S. J.

Willroth

E. C.

(2020). A lifespan perspective on the interconnections between personality, health, and optimal aging. In Hill

P. L.

Allemand

(Eds.), Personality and healthy aging in adulthood: New directions and techniques (pp. 191–202): Springer. https://doi.org/10.1007/978-3-030-32053-9_12

70.

Mueller

Wagner

Gerstorf

(2017). On the role of personality in late life. In Personality development across the lifespan (pp. 69–84). Elsevier. https://doi.org/10.1016/B978-0-12-804674-6.00006-5

71.

Mueller

Wagner

Smith

Voelkle

M. C.

Gerstorf

(2018). The interplay of personality and functional health in old and very old age: Dynamic within-person interrelations across up to 13 years. Journal of Personality and Social Psychology, 115(6), 1127–1147. https://doi.org/10.1037/pspp0000173

72.

Muhammed

Manohar

Ben Yehuda

Chong

T. T.-J.

Tofaris

Lennox

Bogdanovic

Husain

(2016). Reward sensitivity deficits modulated by dopamine are associated with apathy in Parkinson’s disease. Brain, 139(Pt 10), 2706–2721. https://doi.org/10.1093/brain/aww188

73.

Mulder

J. D.

(2023). Power analysis for the random intercept cross-lagged panel model using the powRICLPM R-package. Structural Equation Modeling: A Multidisciplinary Journal, 30(4), 645–658. https://doi.org/10.1080/10705511.2020.1784738

74.

Mulder

J. D.

Hamaker

E. L.

(2021). Three extensions of the random intercept cross-lagged panel model. Structural Equation Modeling: A Multidisciplinary Journal, 28(4), 638–648. https://doi.org/10.1080/10705511.2020.1784738

75.

Mund

Nestler

(2019). Beyond the cross-lagged panel model: Next-generation statistical tools for analyzing interdependencies across the life course. Advances in Life Course Research, 41, 100249. https://doi.org/10.1016/j.alcr.2018.10.002

76.

Murphy

S. C.

Bastian

(2020). Emotionally extreme life experiences are more meaningful. The Journal of Positive Psychology, 15(4), 531–542. https://doi.org/10.1080/17439760.2019.1639795

77.

Nakamura

J. S.

Chen

VanderWeele

T. J.

Kim

E. S.

(2022). What makes life purposeful? Identifying the antecedents of a sense of purpose in life using a lagged exposure-wide approach. SSM-Population Health, 19, Article 101235. https://doi.org/10.1016/j.ssmph.2022.101235

78.

Nakamura

J. S.

Kwok

Huang

Strecher

V. J.

Kim

E. S.

Cole

S. W.

(2023). Reduced epigenetic age in older adults who volunteer. Psychoneuroendocrinology, 148, Article 106000. https://doi.org/10.1016/j.psyneuen.2022.106000

79.

Orth

Meier

L. L.

Bühler

J. L.

Dapp

L. C.

Krauss

Messerli

Robins

R. W.

(2024). Effect size guidelines for cross-lagged effects. Psychological Methods, 29(2), 421–433. https://doi.org/10.1037/met0000499

80.

Ottenbacher

M. E.

Kuo

Y.-F.

Ostir

G. V.

(2007). Test-retest reliability of a psychological well-being scale in hospitalized older adults. Aging clinical and experimental research, 19(5), 424–429. https://doi.org/10.1007/BF03324725

81.

Pfund

G. N.

Hofer

Allemand

Hill

P. L.

(2022). Being social may be purposeful in older adulthood: A measurement burst design. American Journal of Geriatric Psychiatry, 30(7), 777–786. https://doi.org/10.1016/j.jagp.2021.11.009

82.

Pfund

G. N.

Lewis

N. A.

(2020). Aging with purpose: Developmental changes and benefits of purpose in life throughout the lifespan. In: Personality and healthy aging in adulthood: New directions and techniques (pp. 27–42). Springer Nature. https://doi.org/10.1007/978-3-030-32053-9_3

83.

Pfund

G. N.

Ratner

Allemand

Burrow

A. L.

Hill

P. L.

(2022). When the end feels near: Sense of purpose predicts well-being as a function of future time perspective. Aging & Mental Health, 26(6), 1178–1188. https://doi.org/10.1080/13607863.2021.1891203

84.

Pinquart

(2001). Correlates of subjective health in older adults: A meta-analysis. Psychology and Aging, 16(3), 414–426. https://doi.org/10.1037/0882-7974.16.3.414

85.

Pinquart

(2002). Creating and maintaining purpose in life in old age: A meta-analysis. Ageing International, 27(2), 90–114. https://doi.org/10.1007/s12126-002-1004-2

86.

Polenick

C. A.

Kales

H. C.

Birditt

K. S.

(2018). Perceptions of purpose in life within spousal care dyads: Associations with emotional and physical caregiving difficulties. Annals of Behavioral Medicine, 52(1), 77–87. https://doi.org/10.1093/abm/kax005

87.

Purser

J. L.

Weinberger

Cohen

H. J.

Pieper

C. F.

Morey

M. C.

Williams

G. R.

Lapuerta

(2005). Walking speed predicts health status and hospital costs for frail elderly male veterans. Journal of Rehabilitation Research and Development, 42(4), 535–546. https://doi.org/10.1682/jrrd.2004.07.0087

88.

R Core Team . (2022). R: A language and environment for statistical computing . https://www.r-project.org/

89.

Revelle

(2022). Package “Psych”: Procedures for Psychological . Psychometric, and Personality Research. https://personality-project.org/r/psych.

90.

Rosseel

(2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36. https://doi.org/10.18637/jss.v048.i02

91.

Ryff

C. D.

(1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 57(6), 1069–1081. https://doi.org/10.1037/0022-3514.57.6.1069

92.

Ryff

C. D.

Keyes

C. L. M.

(1995). The structure of psychological well-being revisited. Journal of Personality and Social Psychology, 69(4), 719–727. https://doi.org/10.1037/0022-3514.69.4.719

93.

Saadeh

Welmer

A. K.

Dekhtyar

Fratiglioni

Calderón-Larrañaga

(2020). The role of psychological and social well-being on physical function trajectories in older adults. The journals of gerontology. Series A, Biological sciences and medical sciences, 75(8), 1579–1585. https://doi.org/10.1093/gerona/glaa114

94.

Same

McBride

Liddelow

Mullan

Harris

(2020). Motivations for volunteering time with older adults: A qualitative study. PLoS One, 15(5), Article e0232718. https://doi.org/10.1371/journal.pone.0232718

95.

Sargent-Cox

K. A.

Anstey

K. J.

Luszcz

M. A.

(2010). Patterns of longitudinal change in older adults' self-rated health: The effect of the point of reference. Health Psychology: Official Journal of the Division of Health Psychology, 29(2), 143–152. https://doi.org/10.1037/a0017652

96.

Schaefer

S. M.

Morozink Boylan

Van Reekum

C. M.

Lapate

R. C.

Norris

C. J.

Ryff

C. D.

Davidson

R. J.

(2013). Purpose in life predicts better emotional recovery from negative stimuli. PLoS One, 8(11), Article e80329. https://doi.org/10.1371/journal.pone.0080329

97.

Scheier

M. F.

Wrosch

Baum

Cohen

Martire

L. M.

Matthews

K. A.

Schulz

Zdaniuk

(2006). The life engagement test: Assessing purpose in life. Journal of Behavioral Medicine, 29(3), 291–298. https://doi.org/10.1007/s10865-005-9044-1

98.

Schippers

M. C.

Ziegler

(2019). Life crafting as a way to find purpose and meaning in life. Frontiers in Psychology, 10, 2778. https://doi.org/10.3389/fpsyg.2019.02778

99.

Sias

Turtle

(2022). Health shocks and changes in life purpose: Understanding the link between purpose and longevity. medRxiv, Article 22272313. https://doi.org/10.1101/2022.03.13.22272313

100.

Skillbäck

Blennow

Zetterberg

Skoog

Rydén

Wetterberg

Guo

Sacuiu

Mielke

M. M.

Zettergren

Skoog

Kern

(2022). Slowing gait speed precedes cognitive decline by several years. Alzheimer's and Dementia, 18(9), 1667–1676. https://doi.org/10.1002/alz.12537

101.

Smith

Ryan

L. H.

Fisher

G. G.

Sonnega

Weir

D. R.

(2017). HRS psychosocial and lifestyle questionnaire 2006-2016. Survey Research Center, Institute for Social Research, University of Michigan. https://hrs.isr.umich.edu/publications/biblio/9066.

102.

Sone

Nakaya

Ohmori

Shimazu

Higashiguchi

Kakizaki

Kikuchi

Kuriyama

Tsuji

(2008). Sense of life worth living (ikigai) and mortality in Japan: Ohsaki Study. Psychosomatic Medicine, 70(6), 709–715. https://doi.org/10.1097/PSY.0b013e31817e7e64

103.

Sonnega

Faul

J. D.

Ofstedal

M. B.

Langa

K. M.

Phillips

J. W.

Weir

D. R.

(2014). Cohort profile: The health and retirement study (HRS). International Journal of Epidemiology, 43(2), 576–585. https://doi.org/10.1093/ije/dyu067

104.

Stenholm

Westerlund

Head

Hyde

Kawachi

Pentti

Kivimäki

Vahtera

(2015). Comorbidity and functional trajectories from midlife to old age: The health and retirement study. The Journals of Gerontology: Series A, 70(3), 332–338. https://doi.org/10.1093/gerona/glu113

105.

Steptoe

Fancourt

(2019). Leading a meaningful life at older ages and its relationship with social engagement, prosperity, health, biology, and time use. Proceedings of the National Academy of Sciences, 116(4), 1207–1212. https://doi.org/10.1073/pnas.1814723116

106.

Steptoe

Fancourt

(2020). An outcome-wide analysis of bidirectional associations between changes in meaningfulness of life and health, emotional, behavioural, and social factors. Scientific Reports, 10(1), 6463. https://doi.org/10.1038/s41598-020-63600-9

107.

Sutin

A. R.

Luchetti

Stephan

Terracciano

(2022). Sense of purpose in life and motivation, barriers, and engagement in physical activity and sedentary behavior: Test of a mediational model. Journal of Health Psychology, 27(9), 2068–2078. https://doi.org/10.1177/13591053211021661

108.

Swardfager

Rosenblat

J. D.

Benlamri

McIntyre

R. S.

(2016). Mapping inflammation onto mood: Inflammatory mediators of anhedonia. Neuroscience & Biobehavioral Reviews, 64, 148–166. https://doi.org/10.1016/j.neubiorev.2016.02.017

109.

Taylor

A. M. W.

Becker

Schweinhardt

Cahill

(2016). Mesolimbic dopamine signaling in acute and chronic pain: Implications for motivation, analgesia, and addiction. Pain, 157(6), 1194–1198. https://doi.org/10.1097/j.pain.0000000000000494

110.

Weziak-Bialowolska

Bialowolski

(2022). Bidirectional associations between meaning in life and the health, emotional ill-being and daily life functioning outcomes among older adults. Psychology and Health, 39(6), 711–727. https://doi.org/10.1080/08870446.2022.2105842

111.

Wickham

(2016). ggplot2: Elegant Graphics for Data Analysis. New York: Springer-Verlag. https://ggplot2.tidyverse.org.

112.

Wickham

François

Henry

Müller

Vaughan

(2023). dplyr: A Grammar of Data Manipulation. R package version 1.1.4, https://github.com/tidyverse/dplyr, https://dplyr.tidyverse.org

113.

Wickham

Miller

Smith

(2023). haven: Import and Export 'SPSS', 'Stata' and 'SAS' Files. R package version 2.5.4, https://github.com/tidyverse/haven, https://github.com/WizardMac/ReadStat, https://haven.tidyverse.org

114.

Willroth

E. C.

Mroczek

D. K.

Hill

P. L.

(2021). Maintaining sense of purpose in midlife predicts better physical health. Journal of Psychosomatic Research, 145, Article 110485. https://doi.org/10.1016/j.jpsychores.2021.110485

115.

Winter

J. E.

MacInnis

R. J.

Wattanapenpaiboon

Nowson

C. A.

(2014). BMI and all-cause mortality in older adults: A meta-analysis. The American Journal of Clinical Nutrition, 99(4), 875–890. https://doi.org/10.3945/ajcn.113.068122

116.

Wurm

Tomasik

M. J.

Tesch-Römer

(2008). Serious health events and their impact on changes in subjective health and life satisfaction: The role of age and a positive view on ageing. European Journal of Ageing, 5(2), 117–127. https://doi.org/10.1007/s10433-008-0077-5

117.

Yemiscigil

Powdthavee

Whillans

A. V.

(2021). The effects of retirement on sense of purpose in life: Crisis or opportunity? Psychological Science, 32(11), 1856–1864. https://doi.org/10.1177/09567976211024248

118.

Yemiscigil

Vlaev

(2021). The bidirectional relationship between sense of purpose in life and physical activity: A longitudinal study. Journal of Behavioral Medicine, 44(5), 715–725. https://doi.org/10.1007/s10865-021-00220-2

119.

Zaslavsky

Rillamas-Sun

Woods

N. F.

Cochrane

B. B.

Stefanick

M. L.

Tindle

Tinker

L. F.

LaCroix

A. Z.

(2014). Association of the selected dimensions of eudaimonic well-being with healthy survival to 85 years of age in older women. International Psychogeriatrics, 26(12), 2081–2091. https://doi.org/10.1017/S1041610214001768

120.

Zilioli

Slatcher

R. B.

Ong

A. D.

Gruenewald

T. L.

(2015). Purpose in life predicts allostatic load ten years later. Journal of Psychosomatic Research, 79(5), 451–457. https://doi.org/10.1016/j.jpsychores.2015.09.013

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.51 MB

0.18 MB