
==== Front
BMJ Open Sport Exerc Med
BMJ Open Sport Exerc Med
bmjosem
bmjsem
BMJ Open Sport & Exercise Medicine
2055-7647
BMJ Publishing Group BMA House, Tavistock Square, London, WC1H 9JR

10.1136/bmjsem-2023-001875
bmjsem-2023-001875
Original Research
1506
Examining the relationship between different physical activities and health and well-being in middle-aged and older men: an isotemporal substitution analysis
http://orcid.org/0000-0001-9903-9287
Blake Henry T 1henry.blake@unisa.edu.au

http://orcid.org/0000-0002-9652-6757
Crozier Alyson J 1alyson.crozier@unisa.edu.au

https://twitter.com/jonbuckley9
http://orcid.org/0000-0003-0298-2186
Buckley Jonathan D 1jon.buckley@unisa.edu.au

https://twitter.com/B_JStenner
http://orcid.org/0000-0003-1996-1613
Stenner Brad J 1brad.stenner@unisa.edu.au

1 Alliance for Research in Exercise, Nutrition and Activity (ARENA), Allied Health and Human Performance, University of South Australia, Adelaide, South Australia, Australia
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Additional supplemental material is published online only. To view, please visit the journal online (https://doi.org/10.1136/bmjsem-2023-001875).

None declared.

DrHenry TBlake; henry.blake@unisa.edu.au
2024
28 8 2024
10 3 e00187525 7 2024
Copyright © Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.
2024
https://creativecommons.org/licenses/by-nc/4.0/ This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Abstract

Background

Well-being declines during men’s middle years, and while physical activity (PA) can have a positive influence, the type of PA that provides the greatest benefit is not clear. This cross-sectional study explored how participation in different types of PA is associated with the well-being and self-rated health of middle-aged and older men.

Methods

A representative sample of South Australian adults completed a survey which assessed time spent in a variety of physical activities and self-rated health, life worth, life satisfaction and previous day happiness and anxiety. This study focused on a subsample of men≥35 years of age (n=1019). Partition and isotemporal substitution analyses were used to explore relationships between different types of PA and self-rated health and well-being outcomes.

Results

Weekly PA was positively associated with all outcomes among middle-aged men (35–64 years), but only with self-rated health and life satisfaction in older men (65+ years). Substituting out 30 mins of team sport participation with cycling, fitness/gym-based activities or other recreational activities was associated with higher self-rated health in middle-aged men. In older men, substituting 30 mins of dance with most other types of PA was related with greater self-rated health and life worth.

Conclusions

Total weekly PA was consistently related to the self-rated health and well-being of middle-aged men, but less so among older men. Despite some substitution effects for self-rated health, for the most part well-being was not influenced by activity substitution, particularly among middle-aged men.

Aging
Health promotion
Well-being
Male
Physical activity
Australian Government Research Training Program Scholarship Norwood Football Club Top-up scholarship
==== Body
pmcWHAT IS ALREADY KNOWN ON THIS TOPIC

WHAT THIS STUDY ADDS

Weekly PA was positively associated with self-rated health and all well-being outcomes among middle-aged men, but only with self-rated health and life satisfaction among older adult men.

Select PA types demonstrated a statistically significant substitution relationship, such as substituting out team sport and higher self-rated health among middle-aged men and substituting out dance and higher self-rated health and life worth among older men.

Most PA types and subjective and eudaimonic well-being outcomes did not demonstrate statistically significant substitution relationships.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

The findings of this study provide insight that select PA types may be associated with better self-rated health; however, that activity type does not appear influential for subjective and eudaimonic well-being. The type of PA that is engaged in may not be the most important consideration when promoting PA to promote the well-being of men. The study findings and limitations provide a basis for future research.

Introduction

Well-being, while often equated to a single construct, is multifaceted.1 A more complete picture considers both subjective and eudaimonic well-being.2 How one evaluates their life (ie, life satisfaction) and their emotional experiences (ie, positive and negative affect) is conceptualised as subjective well-being,3 while broader aspects of ‘doing well’, such as life purpose, reflect eudaimonic well-being.2

Holistically, positive well-being has been linked to numerous health benefits, including decreased mortality risk,4 improved productivity and social behaviour4 and reduced odds of future depression.5 Indeed, people desire greater well-being, willing to trade-off income and career success for increased well-being.6 However, many factors influence overall well-being, such as age and physical activity (PA). For instance, low life satisfaction and high worry accompanies middle adulthood, with these improving into older age.3 Physical activity appears positively related to evaluative well-being (eg, life satisfaction) and positive affect (eg, happiness), but not negative affect (eg, stress).7 8 In addition to well-being directly, PA is positively related to how one perceives their overall health (self-rated health),9 which in turn has been identified to influence well-being.10

One’s sex also impacts well-being, as men report lower life satisfaction11 and purpose in life12 compared with women. In addition, men are less likely to seek physical and mental health support and use preventative health services compared with women.13 14 Despite this lack of help-seeking, PA has shown promise to engage men in health promotion,15 and thus is an important tool to promote well-being in men.

While PA may be beneficial for the health and well-being of men, we do not know which type of PA is best. As Ekkekakis and Backhouse discuss,16 individuals do not partake in PA in a vacuum, and therefore it is important to contextualise PA. Few studies have compared the effect of PA type on well-being outcomes. There is some evidence however that different types of PA are related differently to mental health,17 well-being18 19 and self-rated health.19 For example, Kekalainen et al19 identified walking in nature, but not other exercise modes assessed, to positively relate to the subjective well-being of 50-year-old men. Further, Werneck et al18 identified participation in team sport or jogging/running, but not other types of leisure-time PA, to be associated with higher future mental well-being among middle-aged men. Despite some evidence of differential effects, prior literature has either clustered PA types (eg, by combining group fitness and team sport) that may have different influences on men’s health and well-being19 or consisted of small sample sizes with limited types of activity examined.20

Evaluating and comparing different PA types can be challenging, as individuals seldom participate in only one type of PA. While traditional single effects models offer insights into the relationship between increased time spent in a specific PA type and the health outcome of interest, they can overlook the potential influence of concurrent participation in other PA types. It is important to consider that increasing time spent in one physical behaviour needs to come at the expense of time spent engaged in another behaviour, and therefore modelling to assess how such substitutions can influence health is becoming common.21 Isotemporal substitution modelling explores the associations between time spent doing one behaviour and replacing it with another, while holding total time constant.22 As this statistical approach provides some insight into the superiority of a particular behaviour compared with another on the outcome of interest, its application to assessing time spent in different types of PA is of interest. For instance, the seminal work from Mekary et al22 revealed distinct associations between substituting different intensities of walking and jogging/running with weight gain in premenopausal women. More recently, Ortola et al23 identified that substituting time spent doing physically active housework or gardening with walking or cycling and sports was associated with healthier ageing in older adults, whereas substituting walking with cycling and playing sports was not.23 To our knowledge, no other studies have used isotemporal substitution with a wider variety of PA types to evaluate influences on the well-being of middle-aged and older men.

Therefore, the present study examined how changing time spent doing different types of PA could influence subjective and eudaimonic well-being, as well as self-rated health, in middle-aged and older men.

Methods

Overview

This study used data from the cross-sectional ‘Active Lives’ South Australia survey conducted between April and May 2019, commissioned by the South Australian Office for Recreation, Sport and Racing. South Australian residents 18 years old and above were identified within the Integrated Public Number Database, a centralised database containing all Australian telephone numbers and associated customer details, and were contacted using random digit dialling. The survey was delivered using a Computer-Assisted Telephone Interviewing system or online. Adapted from Sport England’s ‘Active Lives’ survey,24 self-reported well-being and self-rated health and duration in seven types of PA were captured. A total of 3726 eligible South Australian adults were contacted, 2999 of whom completed the survey and 1073 of whom were men aged 35 years and above. This paper adheres to the Strengthening the Reporting of Observational Studies in Epidemiology checklist for cross-sectional studies,25 and no patient and public involvement occurred in the research methods.

Study variables

Physical activity

Participants reported frequency and duration for walking, cycling, dancing, fitness/gym-based activities, team sport, individual sport and other recreational activities. Average weekly duration spent engaging in each PA type was determined. Total weekly PA duration was determined by summing participants weekly durations in all seven PA types. See online supplemental file 1 for PA questions.

Subjective and eudaimonic well-being

Subjective well-being was assessed using three separate questions: life satisfaction was assessed by asking individuals, ‘overall, how satisfied are you with life nowadays?’; affective evaluations were assessed by two questions: ‘overall, how happy did you feel yesterday’ (previous day happiness) and ‘overall, how anxious did you feel yesterday’ (previous day anxiety). The use of a previous day recall for affective evaluations is more likely to be a true reflection of an emotional state, as intended, and not an evaluative response which can occur when asked over a longer period (eg, week or month).3 Eudaimonic well-being was assessed by a single question: ‘overall, to what extent do you feel the things you do in your life are worthwhile?’ (life worth). Response options were 0 (‘not at all’) to 10 (‘completely’). The four questions, originally developed and tested by the UK Office for National Statistics and since applied in Australian settings,26 address the distinct components of subjective and eudaimonic well-being.27

Self-rated health

Self-rated health was captured with the following question: “In general would you say that your health is…” with possible responses along a 5-point Likert scale of 1=excellent, 2=very good, 3=good, 4=fair and 5=poor. Scores were reversed to aid interpretation and align with the positive well-being outcomes (eg, 5=excellent and 1=poor). The single-item measure of self-rated health is considered to have good predictive validity due to its relationship with future mortality,28 concurrent validity due to its relationship with physical and mental health,29 and has displayed moderate test–retest reliability when delivered over the phone.30

Data handling and analysis

Participant data were excluded for the following reasons: (1) sessional duration exceeding 16 hours for a single PA type (assuming 8-hour sleep period) (n=7); (2) improbable PA frequency and duration combinations (eg, cycling eight times/week with a total weekly duration of 5 min) (n=4); (3) responses indicating ‘don’t know’ or ‘refused’ for the frequency and/or duration of a PA type (n=34); and (4) missing data for covariate variables (n=9). To minimise the impact of extreme durations and over-reporting of PA, outliers (z≤ −3.29 or z≥3.29) were replaced with the next highest value (n=39).

Analyses employed STATA/IC V.15.1 (StataCorp, Texas, USA). For comparisons between middle-aged and older men, χ2 and Fisher’s exact approach addressed demographic variables, while t-tests examined well-being, self-rated health and PA durations. To assess the relationship between weekly PA and well-being and self-rated health, a single effects regression model was used while partition and isotemporal substitution models (described below) were used for the assessment between each PA type and well-being and self-rated health. Prior to analysis, duration of total weekly PA and time spent participating in each of the seven types of PA were converted to 30 min increments (eg, 30 min=1, 15 min=0.5). Linear regression was used to assess life satisfaction, life worth and self-rated health. As linear regression model assumptions were not met for previous day happiness and anxiety, these were dichotomised, and logistic regression was used. These variables were split into two levels: (1) at or better than the group median and (2) inferior to the group median. Covariates in all models included physical disability status (yes or no), socioeconomic status and employment status, based on the available dataset and prior literature.31

A full description and rationale behind partition and isotemporal approaches, including a thorough comparison between the approaches, have been previously described.22 In brief, the partition model, as the initial step, included all seven PA types as independent variables regressed on each outcome. This analysis yielded the change in the outcome variable or odds of an outcome with a 1-unit (30 min) increase in a specific PA type, while adjusting for the other six PA types. Subsequently, isotemporal substitution models included a total weekly PA duration variable and all PA variables, except the PA type of interest. Practically speaking, for each outcome of interest, six regressions were completed per group, with each omitting one of the PA types at a time. Completing this process across all seven PA types provided results that represented the change in the outcome or odds of an outcome when 30 min of the excluded PA type was replaced with 30 min of another PA type, while mutually adjusting for the other five types of PA. While a partition model represents the effect of adding more time spent in a particular type of PA, isotemporal modelling represents the effect of swapping time spent in one PA type with another.22 Given its more direct comparison between PA types, isotemporal substitution was the primary focus of this study.

Variance inflation factors for PA and covariate variables in linear partition models were below five, indicating no problematic multicollinearity.32 Statistical significance was set at an alpha level of 0.05.

Results

Participant characteristics

Data for 1019 male participants were included. Sample size for each analysis can be found within respective tables and figures. See table 1 for participant characteristics. Middle-aged men reported superior self-rated health but worse subjective and eudaimonic well-being than older men. Compared with older men, middle-aged men reported more weekly PA and weekly minutes spent walking, dancing and undertaking other recreational activities, but less time spent in individual sports.

Table 1 Participant demographic characteristics

Category	Middle-aged men (35–64 years) (n=544)	Older men (65+ years) (n=475)	
Mean (SD)	Mean (SD)	
Average age (years)	51.75 (8.4)	72.91 (6.42)	
Socioeconomic status	N (%)	N (%)	
 Lowest quintile	87 (15.99)	61 (12.84)	
 Low quintile	112 (20.59)	116 (24.42)	
 Middle quintile	116 (21.32)	95 (20)	
 High quintile	111 (20.4)	67 (14.11)	
 Highest quintile	118 (21.69)	136 (28.63)*	
Education level	 	 	
 No schooling to secondary	180 (33.09)	180 (37.89)	
 Trade, certificate, diploma	173 (31.8)	138 (29.05)	
 Degree or higher	187 (34.38)	155 (32.63)	
 Not stated	4 (0.74)	2 (0.42)	
Employment status	 	 	
 Working (paid)	405 (74.45)	99 (20.84)***	
 Retired	45 (8.27)	355 (74.74)***	
 Unemployed or unable to work	72 (13.24)	11 (2.32)***	
 Other	19 (3.49)	10 (2.11)	
 Not stated	3 (0.55)	0 (0.00)	
Household income	 	 	
 Up to AUD $80 000	222 (40.8)	339 (71.4)***	
 More than AUD $80 000	251 (46.1)	73 (15.4)***	
 Not stated	71 (13.1)	63 (13.2)	
Country of birth	 	 	
 Australia	417 (76.65)	367 (77.26)	
 UK and Ireland	52 (9.56)	70 (14.74)*	
 Other	73 (13.42)	38 (8)**	
 Not stated/refused	2 (0.37)	0 (0)	
Marital status	 	 	
 Married/living with partner	371 (68.2)	337 (70.95)	
 Separated/divorced	93 (17.1)	73 (15.37)	
 Widowed	10 (1.84)	40 (8.42)***	
 Never married	66 (12.13)	20 (4.21)***	
 Not stated	4 (0.74)	5 (1.05)	
Aboriginal status	 	 	
 Aboriginal	2 (0.37)	7 (1.47)	
 Non-Aboriginal	537 (98.71)	465 (97.89)	
 Not stated	5 (0.92)	3 (0.63)	
Disability status	 	 	
 Yes	119 (21.88)	161 (33.89)***	
 No	421 (77.39)	314 (66.11)***	
 Don’t know/refused	4 (0.74)	0 (0)	
Well-being outcomes	Mean (SD)	Mean (SD)	
 Self-rated health (0–5)	3.27 (1.12)	3.14 (1.06)*	
 Life satisfaction (0–10)	7.36 (1.92)	7.77 (1.73)***	
 Life worth (0–10)	7.69 (1.91)	8.14 (1.82)***	
 Previous day happiness (0–10)	7.44 (2.08)	7.94 (1.92)***	
 Previous day anxiety (0–10)	2.72 (2.79)	2.04 (2.64)***	
Physical activity (min/week)	Mean (range)	Mean (range)	
 Total physical activity	378.46 (0–3480)	300.13 (0–3600)**	
 Walking	244.91 (0–1920)	172.13 (0–960)***	
 Cycling	24.96 (0–600)	17.19 (0–840)	
 Dance	6.63 (0–525)	1.82 (0–180)*	
 Fitness/gym	41.24 (0–780)	31.11 (0–1080)	
 Team sport	15.13 (0–960)	22.07 (0–1440)	
 Individual sport	19.43 (0–480)	39.64 (0–1440)**	
 Other recreational activities	26.17 (0–450)	16.18 (0–450)*	
; standard deviation; N, number; *p*p<0.05; **p**p<0.01; ***p***p<0.001.Statistical significance represents a difference between middle-aged and older men.

Statistical significance represents a difference between middle-aged and older men.

NNumberSDstandard deviation

Overall PA and well-being and self-rated health

Associations between total weekly PA and well-being and self-rated health are presented in table 2. In middle-aged men, higher total weekly PA was related to greater self-rated health, life satisfaction and life worth, and better odds of having higher previous day happiness and lower odds of having higher previous day anxiety. In older men, weekly PA was associated with better self-rated health and life satisfaction.

Table 2 Regression between weekly physical activity and self-rated health and well-being outcomes

	Middle-aged men (35–64 years)	Older men (65+ years)	
Outcome	Weekly PA, β (95% CI)	Weekly PA, β (95% CI)	
Self-rated health	0.01 (0.01 to 0.02)***	0.02 (0.01 to 0.02)***	
Life satisfaction	0.02 (0.02 to 0.03)***	0.02 (0.003 to 0.03)*	
Life worth	0.02 (0.01 to 0.03)***	0.01 (-0.002 to 0.02)	
 	Weekly PA, OR (95% CI)	Weekly PA, OR (95% CI)	
Previous day happiness	1.02 (1.01 to 1.04)**	1.01 (0.99 to 1.03)	
Previous day anxiety	0.99 (0.97 to 0.997)*	0.99 (0.97 to 1.00)	
Beta coefficients represent change in the outcome with a one 1-unit (30-minute min) increase in weekly physical activity. Odds ratiosORs represent the odds of having higher previous day happiness or anxiety with a 30-minute min increase in weekly physical activity. Models included disability status, socio-economic status, and working status as covariates.

p*p<0.05; **p**p<0.01; ***p***p<0.001.

CIconfidence intervalORodds ratioPAphysical activityβbeta coefficient

PA type and well-being and self-rated health

Results for all partition analyses are provided in table 3, while figures14 display isotemporal substitution analyses.

Table 3 Partition analysis for physical activity type and each outcome among middle-aged men and older men

	Middle-aged men (35–64 years)	
Self-rated health(n=535)	Life satisfaction(n=536)	Life worth(n=535)	Previous day happiness(n=537)	Previous day anxiety(n=536)	
Partitionanalysis	β (95% CI)	β (95% CI)	β (95% CI)	OR (95% CI)	OR (95% CI)	
Walking (per 30 min/week)	0.01 (0.002 to 0.02)*	0.02 (0.01 to 0.03)**	0.02 (0.01 to 0.03)**	1.01 (0.99 to 1.03)	0.99 (0.98 to 1.01)	
Cycling (per 30 min/week)	0.03 (0.004 to 0.06)*	0.05 (−0.01 to 0.10)	0.03 (−0.02 to 0.09)	1.05 (0.97 to 1.13)	0.97 (0.90 to 1.04)	
Dancing (per 30 min/week)	0.04 (−0.02 to 0.09)	0.05 (−0.04 to 0.15)	0.06 (−0.04 to 0.15)	1.24 (0.92 to 1.67)	0.94 (0.82 to 1.08)	
Fitness/gym (per 30 min/week)	0.03 (0.004 to 0.05)*	0.02 (−0.02 to 0.06)	0.02 (−0.02 to 0.06)	1.04 (0.97 to 1.10)	0.97 (0.91 to 1.03)	
Team sport (per 30 min/week)	−0.02 (−0.05 to 0.01)	0.04 (−0.02 to 0.09)	0.04 (−0.01 to 0.09)	1.00 (0.94 to 1.07)	0.96 (0.89 to 1.04)	
Individual sport (per 30 min/week)	0.01 (−0.04 to 0.04)	0.08 (0.01 to 0.15)*	0.06 (−0.01 to 0.13)	1.04 (0.94 to 1.15)	0.96 (0.87 to 1.05)	
Other recreational activities (per 30 min/week)	0.05 (0.02 to 0.09)**	0.003 (−0.06 to 0.07)	0.01 (−0.06 to 0.08)	1.10 (0.99 to 1.23)	0.99 (0.91 to 1.08)	
        	Older men (65 years and older)	 	
Self-rated health(n=473)	Life satisfaction(n=473)	Life worth(n=466)	Previous day happiness(n=474)	Previous day anxiety(n=473)	
Partition analysis	β (95% CI)	β (95% CI)	β (95% CI)	OR (95% CI)	OR (95% CI)	
Walking (per 30 min/week)	0.02 (0.01 to 0.04)***	0.02 (−0.01 to 0.04)	0.02 (−0.01 to 0.04)	1.00 (0.97 to 1.03)	1.00 (0.97 to 1.03)	
Cycling (per 30 min/week)	0.05 (0.02 to 0.08)**	0.05 (−0.01 to 0.10)	−0.01 (−0.07 to 0.05)	1.05 (0.95 to 1.16)	0.97 (0.90 to 1.05)	
Dancing (per 30 min/week)	−0.22 (−0.42 to −0.01)*	−0.18 (−0.54 to 0.18)	−0.49 (−0.89 to −0.11)*	0.64 (0.38 to 1.08)	1.12 (0.70 to 1.78)	
Fitness/gym (per 30 min/week)	0.02 (0.001 to 0.05)*	0.02 (−0.02 to 0.06)	0.02 (−0.02 to 0.07)	1.00 (0.95 to 1.06)	1.02 (0.97 to 1.07)	
Team sport (per 30 min/week)	0.01 (−0.02 to 0.03)	0.01 (−0.03 to 0.05)	0.02 (−0.03 to 0.06)	1.03 (0.96 to 1.11)	0.94 (0.86 to 1.01)	
Individual sport (per 30 min/week)	0.01 (−0.01 to 0.03)	0.01 (−0.02 to 0.04)	0.01 (−0.02 to 0.04)	1.04 (0.98 to 1.11)	0.96 (0.91 to 1.01)	
Other recreational activities (per 30 min/week)	−0.03 (−0.07 to 0.01)	0.01 (−0.06 to 0.08)	0.02 (−0.06 to 0.09)	0.98 (0.89 to 1.09)	1.02 (0.93 to 1.12)	
Partition analysis represents change in the outcome or odds of an outcome with a separate one unit (30-minute min) increase in a certain type of physical activity, while mutually adjusting for the other six types of physical activity. Models included disability status, socio-economic status, and working status as covariates.

p*P<0.05; **p**p<0.01; ***p***p<0.001.

βbeta coefficient

Figure 1 Isotemporal substitution analysis and self-rated health and well-being among middle-aged men. Isotemporal substitution analysis represents change in the outcome when 30 min of one type of physical activity is replaced with 30 min of another type of physical activity, while mutually adjusting for the other five types of physical activity. Models included disability status, socioeconomic status and working status as covariates. β, beta coefficient. *P<0.05; **p<0.01.

Figure 2 Isotemporal substitution analysis and affect among middle-aged men. Isotemporal substitution analysis represents odds of an outcome when 30 min of one type of physical activity is replaced with 30 min of another type of physical activity, while mutually adjusting for the other five types of physical activity. Models included disability status, socioeconomic status and working status as covariates.

Figure 3 Isotemporal substitution analysis and self-rated health and well-being among older men. Isotemporal substitution analysis represents change in the outcome when 30 min of one type of physical activity is replaced with 30 min of another type of physical activity, while mutually adjusting for the other five types of physical activity. Models included disability status, socioeconomic status and working status as covariates. β, beta coefficient. *P<0.05; **p<0.01.

Figure 4 Isotemporal substitution analysis and affect among older men. Isotemporal substitution analysis represents odds of an outcome when 30 min of one type of physical activity is replaced with 30 min of another type of physical activity, while mutually adjusting for the other five types of physical activity. Models included disability status, socioeconomic status and working status as covariates.

Self-rated health

Regarding self-rated health, when adjusting for all other types of PA included within this study (partition analyses), a 30 min/week increase in walking, cycling or fitness/gym-based activities related to greater self-rated health in both middle-aged men and older men. Further, a 30 min/week increase in other recreational activities was related to greater self-rated health of middle-aged men, while a 30 min/week increase in dancing was associated with lower self-rated health in older men.

In the isotemporal substitution models, reallocating 30 min/week from team sport to cycling, fitness/gym-based activities, or other recreational activities, or substituting walking with other recreational activities was related to higher self-rated health in middle-aged men. Among older men, substituting out 30 min of dance and replacing this with walking, cycling, fitness/gym-based activities, individual sport or team sport was associated with greater self-rated health. Further, in older men, substituting out other recreational activities with walking, cycling or fitness/gym-based activities and individual or team sport out for cycling was related to greater self-rated health.

Life satisfaction

In relation to life satisfaction, when adjusting for other activities (partition analysis), a 30 min/week increase in walking or individual sport was positively related to life satisfaction in middle-aged men. Substituting PA types for one another did not relate to life satisfaction for either middle-aged or older men.

Life worth

When assessing the relationship between PA and life worth, among older men, a 30 min/week increase in dancing was associated with lower life worth, while for middle-aged men, increasing weekly walking significantly correlated with higher life worth. Reallocating 30 min/week of dance with any of the other six PA types was associated with greater life worth in older men, while PA substitution did not relate to life worth in middle-aged men.

Previous day happiness and anxiety

The odds of better happiness or anxiety were not related to PA types within partition and substitution models in middle-aged or older men.

Discussion

This study investigated whether PA type was associated with well-being and self-rated health in middle-aged and older men. Different types of PA were related differently to self-rated health and well-being outcomes, while the reallocation of PA types was primarily associated with self-rated health and not well-being in middle-aged and older men.

Overall PA, well-being and self-rated health

Consistent with prior research, overall weekly PA was positively associated with self-rated health9 and life satisfaction10 of middle-aged and older men, as well as with a sense of life worth of middle-aged men.33 The positive relationship between PA and both life satisfaction and self-rated health is consistent with findings that improvements in life satisfaction from PA are mediated by improvements in health status among middle-aged and older men.10 Notably, greater weekly PA was related to increased odds of better previous day happiness and anxiety, but in middle-aged men only. While the positive relationship between PA and positive affect aligns with a recent review, the relationship between PA and negative affect identified does not,8 although research has identified sufficiently active adults do experience reduced negative affect when stressful events occur.34 Taken together, the attenuating influence of PA on daily stressors may be particularly important for middle-aged men navigating and balancing competing stressors such as work and family. In contrast, the absence of a relationship between PA and previous day happiness and anxiety in older men may be linked to their better emotional regulation, in part due to the selection of more meaningful social connections that lead to fewer social conflicts.35 Although PA could contribute to the social well-being of older men, other more sedentary social activities may afford these benefits (eg, men’s community groups36).

PA type, well-being and self-rated health

Overall, through both the partition and isotemporal analysis, this study identified limited relationships between the type of PA, self-rated health and well-being, particularly concerning subjective and eudaimonic well-being. Recent work has identified certain types of PA in isolation, such as team sport and jogging/running, to be related to the well-being of middle-aged men.18 However, the findings of the current study may have differed, as the substitution analysis enabled statistical comparison between PA types rather than assessing each in isolation. Activity substitution did, however, demonstrate a relationship with middle-aged men’s self-rated health. Specifically, in middle-aged men, replacing 30 min of team sport with 30 min of cycling, fitness/gym-based activities or other recreational activities was positively correlated with self-rated health. This finding contrasts with the typical perception of team sport as a socially enjoyable and health-promoting activity.37 Further, a recent review identified a mostly positive relationship between sport participation and the self-rated health of middle-aged men, although it also highlighted, the negative role of sport-related injury.38 While the current study controlled for self-reported physical disability; the influence of short-term injury and pain, common from sports, could not be accounted for, important to consider given the negative influence that pain39 and sport-related injury40 may have on the perceived health of middle-aged adults.

Walking, cycling or participating in gym/fitness-based activities emerged as potentially more influential for the positive self-rated health of older men, as increasing time spent in these activities, either in addition to all other PA (partition analysis) or as a replacement to other PA types (isotemporal substitution analysis) were positively related to self-rated health in older men. The functional capacity and independence of older adults are linked to self-rated health,41 and aerobic exercise alone, such as walking and cycling, or in combination with resistance-based exercise, slows the onset of functional impairments.42 The older men in the current study may gain a sense of independence and physical capability through participation in these types of PA, contributing to more positive self-rated health. However, it cannot be confirmed whether PA directly influences self-rated health, as initial health status may influence the type of PA they engage in. For example, older men reporting the presence of a disease or medical risk factor are less likely to cycle regularly.43 Unfortunately, the study could not control for disease status beyond self-reported physical disability.

As older adults with low life worth are more likely to report lower self-rated health,12 it is unsurprising that both self-rated health and life worth shared a similar relationship with dance in the older men of this study; however, the negative direction of this relationship raises questions. As previously discussed, lower self-rated health, and here, lower life worth, may precede participation in dance. However, the observational nature of the present study did not enable this to be confirmed, nor was it possible to rule out a direct negative effect of dance. Limited research has examined the influence of dance on older men’s self-rated health and well-being. One large-scale cross-sectional analysis found no relation between dance and self-rated health, depressive symptoms or daily laughter in older men.44 Caution is warranted when interpreting the relation with dance in the current study, given the low participation rate for dancing, potentially contributing to the wide CIs, and the adjusted r2 indicating the model to only account for ~25% and ~7% of the variance in self-rated health and life worth of older men, respectively.

Rather than a single type of PA being universally ‘better’, when considering the PA well-being relationship, it may be more important to consider how types of PA align with individuals needs or PA preferences. The literature has suggested that PA may only increase well-being when people also experience satisfaction of their psychological needs,45 which may not be universally achieved by engaging in one specific PA type, and could even require a combination of PA types. As discussed by Biddle,46 if people prefer an exercise type, they will inevitably ‘feel good’ completing it, while they will feel ‘less good’ after partaking in exercise not to their liking.

Limitations

While this study offers a novel analysis of the relationship between PA types and health and well-being, it has limitations. The correlative nature of the dataset prevents establishing causation and may in part explain the small beta coefficients and ORs identified.8 As the small relationships reported reflect a 30 min change in PA, this suggests that large increases in PA would be warranted to near a practically significant change in this sample’s self-rated health and well-being, which is important for health practitioners to consider. While several PA types were examined, not all possible PA types were assessed, and the participation rates of some PAs were low. Further, PA participation was based on minutes per week, without consideration for the intensity of the PA, as this was not asked for all PA types consistently. As PA intensities can relate to well-being differently,20 this could be included in similar future studies. In addition, time spent participating in cycling and walking was assessed with a shorter recall period (last 7 days) than all other PA types (over last 28 days/4 weeks), which could influence recall accuracy. With consideration for the analyses used, it is important to consider that partition and isotemporal substitution analyses provided a group-level comparison between activities and health outcomes and did not consider the effect at the individual participant level. Despite including covariate variables, the models accounted for approximately ≤30% of the variance in the outcomes, highlighting the weight of other influences not captured in this study, such as mental (eg, depression) and physical (eg, hypertension) chronic diseases. Lastly, as this study was a secondary analysis, it is, by definition, exploratory, and therefore adjustment for multiple comparisons was not undertaken. While p value adjustment reduces the likelihood of type 1 error (false positive), it, in turn, increases the likelihood of type 2 error (false negative). Given this, the certainty of findings is impaired, and it is important for the reader to interpret relationships reaching significance with caution.

Conclusion

This study supports the continued promotion of PA to improve the self-rated health of middle-aged and older men as well as the well-being of middle-aged men. The substitution of certain PA types was associated with self-rated health of middle-aged and older men, and the life worth of older men. Given the positive relationships between weekly PA and well-being found in this study, particularly among middle-aged men, and the large number of non-significant PA type substitution effects, the type of PA that is engaged in may not be the most important consideration when promoting PA to men (ie, one specific type of PA may not be universally better than another). Therefore, practitioners should encourage the men they work with to choose the type of PA they prefer and would enjoy, promoting engagement for greater health and well-being.

supplementary material

10.1136/bmjsem-2023-001875 online supplemental file 1

Acknowledgements

The authors would like to gratefully acknowledge the South Australian Office for Recreation, Sport, and Racing for providing the dataset used within this study. No individual from the South Australian Office for Recreation, Sport, and Racing provided input into the design of this manuscript nor the analysis and interpretation of the data.

Data availability statement

Data may be obtained from a third party and are not publicly available.

Funding: This study was funded by the Australian Government Research Training Program Scholarship and Norwood Football Club (Top-up scholarship).

Provenance and peer review: Not commissioned; externally peer reviewed.

Patient consent for publication: Not applicable.

Ethics approval: This study involves human participants and was approved by the SA Health Human Research Ethics Committee (HRE00418) and the University of South Australia Human Research Ethics Committee (204456). Participants gave informed consent to participate in the study before taking part.

Data availability free text: The dataset that supports the findings of this study is not publicly available and was provided by the South Australian Office for Recreation, Sport and Racing for the current analysis, and therefore restrictions apply to the availability of these data. Data are however available from the authors on reasonable request and with permission of the South Australian Office for Recreation, Sport and Racing.

Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
==== Refs
References

1 Forgeard M Jayawickreme E Kern M et al Doing the right thing: Measuring well-being for public policy Intnl J Wellbeing 2011 1 10.5502/ijw.v1i1.15
2 Martela F Sheldon KM Clarifying the concept of well-being: Psychological need satisfaction as the common core connecting eudaimonic and subjective well-being Rev Gen Psychol 2019 23 458 474 10.1177/1089268019880886
3 Steptoe A Deaton A Stone A Psychological wellbeing, health and ageing Lancet 2015 385 640 648 10.1016/S0140-6736(13)61489-0 25468152
4 De Neve J Diener E Tay L et al The objective benefits of subjective well-being World Happiness Report 2013 New York UN Sustainable Development Solutions Network 2013
5 Wood AM Joseph S The absence of positive psychological (eudemonic) well-being as a risk factor for depression: a ten year cohort study J Affect Disord 2010 122 213 217 10.1016/j.jad.2009.06.032 19706357
6 Adler MD Dolan P Kavetsos G Would you choose to be happy? Tradeoffs between happiness and the other dimensions of life in a large population survey J Econ Behav Organ 2017 139 60 73 10.1016/j.jebo.2017.05.006
7 Wiese CW Kuykendall L Tay L Get active? A meta-analysis of leisure-time physical activity and subjective well-being J Posit Psychol 2018 13 57 66 10.1080/17439760.2017.1374436
8 Buecker S Simacek T Ingwersen B et al Physical activity and subjective well-being in healthy individuals: a meta-analytic review Health Psychol Rev 2021 15 574 592 10.1080/17437199.2020.1760728 32452716
9 Engberg E Liira H Kukkonen-Harjula K et al Associations of physical activity with self-rated health and well-being in middle-aged Finnish men Scand J Public Health 2015 43 190 196 10.1177/1403494814561820 25504655
10 Maher JP Pincus AL Ram N et al Daily physical activity and life satisfaction across adulthood Dev Psychol 2015 51 1407 1419 10.1037/dev0000037 26280838
11 Park J Joshanloo M Scheifinger H Predictors of life satisfaction in Australia: A study drawing upon annual data from the Gallup World Poll Aust Psychol 2020 55 375 388 10.1111/ap.12441
12 AshaRani PV Lai D Koh J et al Purpose in life in older adults: A systematic review on conceptualization, measures, and determinants Int J Environ Res Public Health 2022 19 5860 10.3390/ijerph19105860 35627396
13 Yousaf O Grunfeld EA Hunter MS A systematic review of the factors associated with delays in medical and psychological help-seeking among men Health Psychol Rev 2015 9 264 276 10.1080/17437199.2013.840954 26209212
14 Deeks A Lombard C Michelmore J et al The effects of gender and age on health related behaviors BMC Public Health 2009 9 213 10.1186/1471-2458-9-213 19563685
15 Bottorff JL Seaton CL Johnson ST et al An updated review of interventions that include promotion of physical activity for adult men Sports Med 2015 45 775 800 10.1007/s40279-014-0286-3 25430599
16 Ekkekakis P Backhouse S Exercise and psychological well-being Olympic Textbook of Science in Sport UK Blackwell Publishing 2008
17 Chekroud SR Gueorguieva R Zheutlin AB et al Association between physical exercise and mental health in 1·2 million individuals in the USA between 2011 and 2015: a cross-sectional study Lancet Psychiatry 2018 5 739 746 10.1016/S2215-0366(18)30227-X 30099000
18 Werneck AO Stubbs B Kandola A et al Prospective associations of different contexts of physical activity with psychological distress and well-being among middle-aged adults: An analysis of the 1970 British Cohort Study J Psychiatr Res 2021 140 15 21 10.1016/j.jpsychires.2021.05.049 34087751
19 Kekäläinen T Freund AM Sipilä S et al Cross-sectional and longitudinal associations between leisure time physical activity, mental well-being and subjective health in middle adulthood Applied Research Quality Life 2020 15 1099 1116 10.1007/s11482-019-09721-4
20 Klussman K Langer J Nichols AL The relationship between physical activity, health, and well-being Eur J Health Psychol 2021 28 59 70 10.1027/2512-8442/a000070
21 Grgic J Dumuid D Bengoechea EG et al Health outcomes associated with reallocations of time between sleep, sedentary behaviour, and physical activity: a systematic scoping review of isotemporal substitution studies Int J Behav Nutr Phys Act 2018 15 69 10.1186/s12966-018-0691-3
22 Mekary RA Willett WC Hu FB et al Isotemporal substitution paradigm for physical activity epidemiology and weight change Am J Epidemiol 2009 170 519 527 10.1093/aje/kwp163 19584129
23 Ortolá R García-Esquinas E Cabanas-Sánchez V et al Association of physical activity, sedentary behavior, and sleep with unhealthy aging: consistent results for device-measured and self-reported behaviors using isotemporal substitution models J Gerontol A Biol Sci Med Sci 2021 76 85 94 10.1093/gerona/glaa177 32701141
24 Ipsos MORI Active Lives Survey: Year 1 Technical Report Ipsos MORI Social Research Institute 2017
25 von Elm E Altman DG Egger M et al The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies PLoS Med 2007 4 296 10.1371/journal.pmed.0040296
26 Department of Health The victorian happiness report: the subjective wellbeing of victorians Victoria State Government of Victoria 2015
27 National Research Council Subjective well-being: measuring happiness, suffering, and other dimensions of experience Washington DC The National Academies Press 2013
28 Bamia C Orfanos P Juerges H et al Self-rated health and all-cause and cause-specific mortality of older adults: Individual data meta-analysis of prospective cohort studies in the CHANCES Consortium Maturitas 2017 103 37 44 10.1016/j.maturitas.2017.06.023 28778331
29 Baćak V Ólafsdóttir S Gender and validity of self-rated health in nineteen European countries Scand J Public Health 2017 45 647 653 10.1177/1403494817717405 28673121
30 Dal Grande E Fullerton S Taylor A Reliability of self-reported health risk factors and chronic conditions questions collected using the telephone in South Australia, Australia BMC Med Res Methodol 2012 12 10.1186/1471-2288-12-108
31 Das KV Jones-Harrell C Fan Y et al Understanding subjective well-being: perspectives from psychology and public health Public Health Rev 2020 41 25 10.1186/s40985-020-00142-5 33292677
32 Gareth J Daniela W Trevor H et al Linear regression An Introduction to Statistical Learning with Applications in R 1st edn New York Springer 2013
33 Rector JL Christ SL Friedman EM Well-being and long-term physical activity participation in midlife adults: A latent class analysis Ann Behav Med 2019 53 53 64 10.1093/abm/kay016 29878042
34 Puterman E Weiss J Beauchamp MR et al Physical activity and negative affective reactivity in daily life Health Psychol 2017 36 1186 1194 10.1037/hea0000532 29016149
35 Charles ST Carstensen LL Social and emotional aging Annu Rev Psychol 2010 61 383 409 10.1146/annurev.psych.093008.100448 19575618
36 Foettinger L Albrecht BM Altgeld T et al The role of community-based men’s sheds in health promotion for older men: a mixed-methods systematic review Am J Mens Health 2022 16 10.1177/15579883221084490
37 Andersen MH Ottesen L Thing LF The social and psychological health outcomes of team sport participation in adults: An integrative review of research Scand J Public Health 2019 47 832 50 10.1177/1403494818791405 30113260
38 Blake HT Buckley JD Stenner BJ et al Sport participation and subjective uutcomes of health in middle-aged men: A scoping review Am J Mens Health 2022 16 10.1177/15579883221084493
39 Chireh B D’Arcy C Pain and self-rated health among middle-aged and older Canadians: an analysis of the Canadian community health survey-healthy aging BMC Public Health 2018 18 1006 10.1186/s12889-018-5912-9 30103705
40 Moreira NB Mazzardo O Vagetti GC et al Quality of life perception of basketball master athletes: association with physical activity level and sports injuries J Sports Sci 2016 34 988 96 10.1080/02640414.2015.1082615 26323316
41 Rooney B Havens S Functional fitness, disease and independence in community-dwelling older adults in Western Wisconsin Wis Med J 2004 103 42 8
42 Paterson DH Warburton DE Physical activity and functional limitations in older adults: a systematic review related to Canada’s Physical Activity Guidelines Int J Behav Nutr Phys Act 2010 7 38 10.1186/1479-5868-7-38 20459782
43 Huy C Becker S Gomolinsky U et al Health, medical risk factors, and bicycle use in everyday life in the over-50 population J Aging Phys Act 2008 16 454 464 10.1123/japa.16.4.454 19033605
44 Tsuji T Kanamori S Saito M et al Specific types of sports and exercise group participation and socio-psychological health in older people J Sports Sci 2020 38 422 9 10.1080/02640414.2019.1705541 31876442
45 Teas E Friedman E Amireault S Purpose in life and personal growth: The unique and joint contribution of physical activity and basic psychological needs Appl Psychol Health Well Being 2022 14 795 818 10.1111/aphw.12347 35107871
46 Biddle S Emotion, mood and physical activity Biddle S Fox K Boutcher S Physical activity and psychological well-being London Routledge 2000
