Table 5.

Studies investigating PA in adults across SES. Mixed relationships.

Study	Aim and Study Design	Sample	Measures of SES 1	Outcome/Conclusion
Borrell et al., 2000	Describe social class inequalities in health related behaviors. Interview survey	n = 4171 M: 1942 F: 2229 Age: 14–65+ SPAIN	Education: − Income: − Occupation: + Neighborhood: − Other: −	Less than 5% of M and W in class 1 (highest SES) declared that they usually performed intense PA in contrast with 11.5% of M and 8.6% of W in class 5, an association that persisted in the multivariate analysis. People of classes 1&2 were more likely to engage in usual PA classified as “light or none” than lower classes. For LTPA the situation was reversed, particularly in M, as a greater proportion of the lower classes did not engage in PA three or more times per week. In the multivariate analysis, the association was not significant. In W, there was no clear trend. Engaging in usual PA as “light or none” in M decreased with lowering class [53]
Livingstone et al., 2001	Evaluate habitual levels of PA. Questionnaire	n = 1369 M: 655 F: 714 Age: 18–64 IRELAND	Education: − Income: − Occupation: + Neighborhood: − Other: −	Professional/skilled non-manual M engage in less total and OPA than M from other social groups. Reverse in W. HPA by M were broadly similar across social class groupings but W in skilled manual/partly skilled/unskilled occupations spent more time in these HPA than W from other social groups. Differences in time spent in vigorous active recreation by M were reported, but none was significant. Approximately 2× difference in the range of time spent in vigorous active recreation by the W (0.7 ± 0.9 h·week−1 skilled manual vs. 1.2 ± 2.0 h·week−1 skilled non-manual). W in professional/skilled non-manual groups spent significantly more time in these pursuits than W in other social class groupings [54]
Popham & Mitchell, 2007	Investigate further associations between SES position and overall PA levels and specific types of PA. To investigate the role of employment status and health. Questionnaire	n = 5287 M: 2346 F: 2941 Age: 25–64 SCOTLAND	Education: + Income: − Occupation: + (parent’s and own) Neighborhood: − Other: housing tenure	Increasing accumulated socioeconomic disadvantage was associated with higher rates of low or no PA. For M, this association largely disappeared after adjustment for employment status and health, while among W the differences were reduced. Although low SES was associated with higher rates of OPA, the most disadvantaged did not have the highest rate. However, after adjustment for employment status (especially) and health, a clearer social gradient in OPA emerged in which relative rates of OPA increased with increasing disadvantage. Low SES was associated with low rates of participation in brisk walking, sport and exercise and heavy manual leisure. SES differences in these PA were not greatly changed after adjustment for health and employment status [55]
Allender et al., 2008	Examine relative contribution of OPA to English adults’ meeting PA recommendations. Cross-sectional survey, individual interviews	n = 13,974 M: 6237 F: 7737 Age: 16–75+ ENGLAND	Education: + Income: + Occupation: + Neighborhood: − Other: −	Education: OPA included, M w/any qualification were more likely to meet the PA guideline than those w/a degree or higher or the no qualification group. OPA removed; those w/any qualification or a degree qualification or higher were more likely to meet the guideline than the no qualifications group. Occupation: OPA included; unskilled manual, semiskilled manual and skilled manual W were more likely to meet the PA guideline than the professional group. Not significant when OPA was removed from the analysis [56]
Jurakic et al., 2009	Determine the PA level in different domains of everyday life. Questionnaire	n = 1032 M: 500 F: 532 Age: 15+ CROATIA	Education: + Income: + Occupation: − Neighborhood: − Other: settlement	Total PA was inversely related to the size of settlements. OPA domain was also inversely related to the size of settlements. Furthermore, TPA was inversely related to household income, while PA in HPA was positively related to age and inversely related to the size of settlements and educational level. Finally, LTPA was positively related to the size of settlements and to household income [57]
Molina-García et al., 2014	Describe differences in energy exposure in active commuting to university by transport mode in students and examine sociodemographic associations with energy exposure. Questionnaire	n = 518 M: 209 F: 309 Age: mean 22.4 SPAIN	Education: − Income: − Occupation: − Neighborhood: − Other: subjective definition	Low SES-students walked more but biking was significantly higher in the high SES group than the medium SES group [18]
Golubic et al., 2014	Describe PA and sedentary behavior and examine the variation of PA-sub-components by key health-related, anthropometric, and socio-demographic factors as well as prior PA. Questionnaire, heart rate and move sensing	n = 1787 M: 862 F: 925 Age: 60–64 GREAT BRITAIN	Education: + Income: − Occupation: + Neighborhood: − Other: employment status	For those still working, M in manual work had higher PA energy expenditure (14%), than non-manual workers, but values for W did not differ. In W, PA energy expenditure were greater with higher education. PA energy expenditure from questionnaire was higher in full-time employed than in those who were employed part time or retired. PA energy expenditure were greater in those in manual than non-manual occupations in M, but not significant in W. W with higher education had higher PA energy expenditure than those with lower, but the opposite patterns were observed in M [21]
Hawkins et al., 2004	Describe the prevalence of self-reported moderate/vigorous PA. Questionnaire	n = 40,261 M: 18,375 F: 21,406 Age: 20–55+ USA	Education: + Income: − Occupation: − Neighborhood: − Other: −	Subjects with education beyond high school were less likely to meet the moderate PA guideline than those with less education. The younger, M, and better educated were most likely to achieve the vigorous PA guideline before and after adjustment for potential confounding variables [58]
Berrigan et al., 2006	Explore inclusion of non-leisure-time walking and bicycling (NLTWB) used for transportation on the prevalence of adherence to PA recommendations and the magnitude of apparent disparities in adherence for Californian adults. Phone survey	n = 55,151 M: 22,930 F: 32,221 Age: 18–≥70 USA	Education: + Income: + Occupation: − Neighborhood: − Other: −	Adherence based on LTPA increased with education and income level. By contrast, adherence based on NLTWB decreased with education and income. Logistic regression confirmed the presence of significant effects of(…) education, and income on adherence based on LTPA and the prevalence of adherence based on LTPA and NLTWB combined. In multivariate models, (…) education, and income, were associated with adherence based on NLTWB alone. LTPA increases as education and income levels increase but NLTWB decreases [16]
Lee & Levy, 2011	Examine PA in multiple contexts and blood pressure across gender and income among older adults living independently. Questionnaire	n= 372 M: 128 F: 244 Age: 60+ USA	Education: + Income: + Occupation: − Neighborhood: − Other: −	M at low income levels reported greater HPA than M at high income levels. For W, no differences by income level in HPA were seen. Income level alone also made a significant contribution to differences seen in HPA however these effects appear to have been overridden by the significant interaction. Income level significantly contributed to differences seen in total LTPA with those at low income levels reporting less LTPA than those with higher income levels [59]
Florindo et al., 2009	Estimate the prevalence of and identify factors associated with LTPA, TPA, OPA and HPA. Questionnaire	n = 1318 M: 652 F: 666 Age: 18–65 BRAZIL	Education: + Income: − Occupation: − Neighborhood: − Other: −	Higher education level was negatively associated with low level of LTPA, while it was positively associated with low activity in both OPA and HPA [60]
Bicalho et al., 2010	Estimate the PA level and its association with SES factors in adults living in rural areas. Questionnaire	n = 567 M: 275 F: 292 Age:18–60 BRAZIL	Education: + Income: − Occupation: − Neighborhood: − Other: marital status	There was an inverse relationship between education and the percentage of participants performing 150 min. at work. Education had an inverted U-shaped association with the practice of HPA, in the total population and in W. M (p = 0.133). LTPA was more frequent in individuals with greater education for total, M, and W. M with higher education were the least active in the TPA domain. (W and total: not significant) [61]
Del Duca et al., 2013	Estimate the prevalence and sociodemographic indicators associated with PIA LTPA, TPA, OPA and HPA in adults. Questionnaire	n = 1720 M: 769 F: 951 Age: 20–59 BRAZIL	Education: + Income: + Occupation: − Neighborhood: − Other: −	LTPA: those with lower education and lower income had higher probability of PIA. TPA: higher income presented higher prevalence of PIA. OPA: higher education and income, more PIA. HPA: Higher education and higher income; higher prevalence of PIA [62]
Linetzky et al., 2013	Evaluate how SES gradients in non-communicable diseases and non-communicable disease-related risk factors change over time (2005–2009). Questionnaire	n = 41,392/34,732 M:17,827/15,028 F:23,565/19,704 Age: 43.3/43.6 ARGENTINE	Education: + Income: + Occupation: − Neighborhood: − Other: −	In 2005, M with low education (OR = 0.65, 95% CI = 0.50–0.85) and medium education (OR = 0.79, 95% CI = 0.67–0.93) were less likely than males with high education to be physically inactive. In 2009, the direction of the gradient switched direction. By 2009, W with low education (OR = 1.57, 95% CI = 1.34–1.84) and medium education (OR = 1.18, 95% CI = 1.06–1.32) were more likely than women with high education to be physically inactive [63]
Giles-Corti et al., 2002	Examine spatial access to recreational facilities and perceptions of the neighborhood environment and PA levels by the SES of area of residence. Survey	n = 1803 M: 580 F: 1223 Age: 18–59 AUSTRALIA	Education: + Income: + Occupation: − Neighborhood: + Other: work outside home, access to motor vehicle	No difference between the two SES areas in walking overall, but the types of walking differed significantly. Compared with high SES areas, walking for transport was 33% more prevalent in walkers from low SES areas and walking for recreation was 21% lower. Participation in vigorous PA was 24% lower for those living in low SES areas compared with those in high SES areas and participation in light to moderate PA was 16% lower. On average, compared with those living in high SES areas, those living in low SES areas who walked for transport did so for nearly 1 more hour per fortnight more. Although the difference in walking occasions did not reach significance, in low SES areas transport walkers walked on nearly two more occasions per fortnight. Respondents living in low SES areas were 26% less likely to do sufficient PA compared w/those living in high SES areas. The odds of reaching high levels of vigorous PA were also near 50% lower [64]
Proper et al., 2006	Examine the influence of neighborhood and individual SES on OPA. Questionnaire	n = 1236 M: 470 F: 766 Age: 20–65 AUSTRALIA	Education: + Income: + Occupation: − Neighborhood: + Other: −	Neighborhood SES and individual SES were independently inversely related to absolute and relative amount of OPA. Significant interactions between neighborhood SES and level of educational attainment in the contribution of total and vigorous OPA to total PA were found [65]
Kahan, et al., 2005	Evaluate levels of LTPA, OPA, sports PA and correlate them with SES and health factors Questionnaire	n = 406 M: 173 F: 211 Age: 20–65 ISRAEL	Education: + Income: + Occupation: − Neighborhood: − Other: −	OPA level decreased with level of education, whereas sports PA increased. The sports index was also directly correlated with monthly income status: income 5000< NIS (4 NIS equaled U.S. $1.00) was associated with a significantly higher sports PA index and lower OPA index. Regression models showed that the lower the level of education, the greater the degree of OPA and the lower the degree of sports PA. The higher the income, the greater tendency to less OPA but more at sports PA [66]
Khaing Nang et al., 2010	Evaluate the characteristics of individuals participating in different PA domains. Questionnaire	n = 4750 M: 2280 F: 2470 Age: 18–60 SINGAPORE	Education: + Income: + Occupation: − Neighborhood: − Other: work, house	A higher SES was associated with a higher likelihood of participating in LTPA. OPA was higher in those with low SES. TPA was lower for those with higher SES. HPA was lowest for those with higher SES. Participants with a higher SES had more LTPA, but less OPA, TPA and HPA resulting in lower overall PA [67]
Saito et al., 2013	Examine the association of 3 types of PA and their associations with individual and neighborhood environmental factors among middle-aged and elderly Japanese. Questionnaire	n = 1940 M: 943 F: 997 Age: 40–69 JAPAN	Education: + Income: + Occupation: − Neighborhood: + Other: employment, number of children	Not working increased and number of children in the household decreased the odds of all three types of PA (not all significant). Economic status increased the odds of moderate-to-vigorous LTPA but decreased the odds of transport-related walking. High education increased the odds of moderate-to-vigorous LTPA. Owing motor vehicles increased the odds of engaging in moderate-to-vigorous LTPA other than walking [68]
Talaei et al., 2013	Investigate PA by SES and sex in an Iranian adult population. Questionnaire	n = 6622 M: 3221 F: 3401 Age: mean 45.2 IRAN	Education: + Income: + Occupation: + Neighborhood: − Other: employment	LTPA: higher for high SES participants than middle and low SES for both M and W. OPA: W: no significant difference between low and high SES. M: less in high than middle and low SES. HPA: W: significantly different (p < 0.001) higher in middle than high and low SES: it was lower in high than low SES. M: No significant differences found. TPA: M and W: No significant differences [69]
Ying Chan et al., 2014	Examine the association between socio-demographic factors and PIA by gender. Questionnaire	n = 33,949 M: 15,205 F: 18,744 Age: 18–65+ MALAYSIA	Education: + Income: + Occupation: − Neighborhood: − Other: employment	PIA in M increased with increasing income, but not in W. The widow/widower/divorcee, non-working group, and those with no formal education were found to have high PIA in both M and W [70]

PA: physical activity; LTPA: leisure time physical activity; OPA: occupational physical activity; HPA: household physical activity; TPA: transporting physical activity; PIA: physical inactivity; SES: socioeconomic status; SEP: Socioeconomic position; M: men; W: women. ¹ The symbols +/− indicates whether the particular SES measure is used in the study (+) or not (−).