Socioeconomic status is not associated with health-related quality of life in a group of overweight middle-aged men

Introduction

Socioeconomic status (SES) is a major determinant of health. Across the socioeconomic spectrum there are stepwise improvements in mortality and morbidity with increasing wealth (Adler & Ostrove, 1999; Braveman & Gottlieb, 2014). SES is likely to affect health through complex direct and indirect pathways (Adler & Newman, 2002; Braveman & Gottlieb, 2014). For example, people with lower incomes report greater financial obstacles to effective treatment, often leading to delayed diagnosis and suboptimal management of health conditions (Osborn et al., 2016).

Health-related quality of life (HRQL) refers to the effects of health, illness and treatment on perceived quality of life (Ferrans et al., 2005). SES has previously been described to affect HRQL in a nationally representative cohort of Canadian adults (Ross et al., 2012), in adults with a chronic disease in Germany (Mielck, Vogelmann & Leidl, 2014), and among males (but not females) in Japan (Yamazaki, Fukuhara & Suzukamo, 2005). Among patients with rheumatoid arthritis and other chronic conditions, there was evidence that lower income was associated with poorer HRQL (Alishiri et al., 2008; Ovayolu, Ovayolu & Karadag, 2011), but females made up the vast majority of both study populations. Of note, in the USA, disparities in self-rated health when stratified by income group were widest in adults aged 45 to 54 years (Robert et al., 2009). Previous studies have also indicated that poorer SES is associated with adverse metabolic outcomes, but two of these studies observed this association only among women (Chichlowska et al., 2008; Lim et al., 2012), with a third investigation showing only a week association among male participants (Loucks et al., 2007).

We are not aware of any previous study in New Zealand into the association between SES and HRQL in middle-aged males. Importantly, the prevalence of obesity among adult males in New Zealand increased from 26% in 2006/07 to 30.5% in 2015/16, and it was highest in the most deprived areas at 40.7% (Ministry of Health, 2016). Thus, in light of previous evidence and given the increasing prevalence of obesity in New Zealand men (and its particularly high prevalence in areas of low socioeconomic status), we aimed to investigate whether lower SES would also be associated with HRQL in a phenotypically homogenous group of overweight middle-aged men.

Methods

Results

Forty-five subjects participated in the clinical trial (De Bock et al., 2013), but seven had incomplete HQRL data and were excluded. Thus, we studied 38 overweight men (BMI 27.3 ±1.4 kg/m²) aged 45.9 ±5.4 years (range 34.5–55.6 years), who were mostly of New Zealand European ethnicity (89%). Three participants were on antihypertensive medication, three were on lipid-lowering medications, and two participants were on both. No participants had any other physical or mental health co-morbidities.

Comparisons between our group of participants and New Zealand normative data have been previously reported (Derraik et al., 2014). Briefly, our cohort had similar scores in all mental health domains when compared to normative data, but our participants displayed better role physical (p < 0.001), physical functioning (p < 0.001), bodily pain (p = 0.012), and general health (p = 0.009) scores (Derraik et al., 2014).

Discussion

We did not observe any associations between SES and HRQL in our group of overweight middle-aged men, whether assessed by NZDep2006 or capital value of residence. Our findings contrast to the results of Minet Kinge & Morris (2010) who observed lower HRQL scores among obese and overweight individuals with a lower SES than those of the same weight with a higher SES. Studies in Canada (Ross et al., 2012) and Greece (Pappa et al., 2009) also observed that greater affluence was associated with higher HRQL.

However, the existing evidence for the association between HRQL and SES is not consistent. In Japan, Yamazaki, Fukuhara & Suzukamo (2005) observed a strong association between income and HRQL amongst Japanese men but not women, while an older study in that country found a very modest effect of income on HRQL (Asada & Ohkusa, 2004). Pappa et al. (2015) observed that among Roma adults in Greece material deprivation was associated with lower HRQL scores on certain domains (general health and vitality), but with higher HRQL scores on others (role physical and role emotional), findings that the authors were unable to explain. Other studies focusing on physical health (more specifically on prevalence and characteristics of the metabolic syndrome) also failed to observe any association with HRQL among men in the United States (Chichlowska et al., 2008) and South Korea (Lim et al., 2012). It is possible that some of these inconsistencies may be associated with for example, varying levels of education, as previous studies have shown a strong relationship between education levels and HRQL (Devlin, Hansen & Herbison, 2000; Lacey & Walters, 2003; Mielck, Vogelmann & Leidl, 2014). Of note, Miravitlles et al. (2011) observed that lower levels of education were associated with lower HRQL, but no association was observed for occupational status. Thus, it is possible that while we did not observe an association between area-level deprivation or capital value of residence with HRQL, the results might have been different if it had been possible to account for the participants’ levels of education.

Nonetheless, while SES did not appear to affect HRQL in our participants at the time of assessment, it is unlikely to remain unimportant over time. It has been proposed that people of lower SES face a ’double burden’, namely increased health complications as well as lower HRQL when their health declines (Mielck, Vogelmann & Leidl, 2014). Participants in our study were middle-aged, overweight but not obese, did not smoke, and had no significant comorbid conditions. Over time, these men may put on more weight, and adult weight gain is associated with increased risk of chronic diseases in men (De Mutsert et al., 2014). This progression from overweight to a diseased state is more likely to occur in those of lower SES. Thus, even though our results covered a period of approximately 30 weeks, our findings may not reflect the trend over years, and over time it is likely that those of lower SES would have a larger reduction in HRQL than those who are wealthier.

Lower SES is associated with increased morbidity and mortality. A recent review of the association between SES and cardiovascular disease and/or cardiovascular risk factors, reported a clear inverse gradient in high income countries (De Mestral & Stringhini, 2017). Several studies have also identified a link between SES and type 2 diabetes (Bird et al., 2015; Espelt et al., 2013; Hwang & Shon, 2014). Early identification of modifiable risk factors and appropriate lifestyle changes can improve health outcomes and reduce the onset of chronic disease. However, if the health of at-risk middle-age men is not affecting their quality of life, they are less likely to meet a health professional. Health promotion efforts are also more likely to be acted upon by those of higher SES (Adler & Newman, 2002). Thus, our findings are of relevance as they underscore the importance of health promotion that focuses on disease prevention and identification of risk factors in men who feel well, particularly targeted to those of low SES.

The main limitations of our study were the secondary analyses of our clinical trial data, our relatively small number of participants (n = 38), and a selected group of individuals (overweight males living in a large urban centre), which may limit wider applicability of our findings. Particularly, since our cohort was relatively narrow and recruited from a single centre in New Zealand. Further, as all participants were volunteers in a clinical trial examining the effects of a nutritional supplement on metabolism, they are potentially more motivated than equivalent groups in the wider population. We also acknowledge that other factors that have been shown to affect HRQL (such as the education levels of the individuals and their parents (Ross et al., 2012)) have not been controlled for, and could have affected study outcomes.

Nevertheless, our findings are likely to be robust as each participant underwent three assessments over a 30-week period, minimizing the potential effects of temporal variations in HRQL that individual participants might have experienced. In addition, the lack of associations in our study remained irrespective of whether SES was assessed using a measure of neighbourhood wealth (NZDep2006) or family wealth (capital value of residence). Lastly, we studied a relatively homogeneous group of men (mostly of New Zealand European ethnicity), which likely mitigated the confounding effects of ethnicity on study outcomes.

Conclusion

Our findings show that SES per se was not associated with HRQL in middle-aged men who are overweight but otherwise healthy, an important and increasing group. While the equality of HRQL across the socioeconomic spectrum in this group would generally be considered favourable, overweight men of lower SES are known to have increased cardiometabolic risk. That this risk is not associated with impaired quality of life potentially reduces the likelihood that such men will to seek to improve their health. This finding underscores the fact that health promotion interventions aimed at improving the health of at-risk men should also aim to reach asymptomatic people who feel generally well.

Supplemental Information