Background
Mental disorders are highly prevalent and represent a major cause of disability worldwide [
1,
2]. Disability has been defined as functioning restrictions or activity limitations in multiple dimensions of life that results from the interaction between health determinants and contextual factors [
3‐
6]. Mental disorders represent a challenge to individuals’ quality of life, daily functioning and work performance, possibly contributing to reduction of income and standards of living [
7‐
14]. The labour force participation rate of people with mental disorders has been found to be lower when compared to the rest of the population due to higher unemployment rates, sickness absence and early retirement [
15‐
19]. Consequently, studies have shown productivity loss as a main contributor to the economic burden of these disorders [
8,
13,
20,
21].
Under a public health perspective, alongside prevention efforts and access to adequate healthcare, it is important to evaluate which factors may contribute to increased levels of disability among people with mental disorders. For instance, studies have suggested a higher risk of disability due to mental disorders among those socioeconomically more disadvantaged [
22‐
26]. Specifically, educational gradients in sickness absence and early retirement due to mental disorders have been found [
25,
26], as well as socioeconomic inequalities in onset, duration and recurrence of work related disability due to depression and other mental disorders [
22‐
26]. Moreover, the risk of exclusion from the labour market among people with mental disorders is likely to aggravate existing social inequalities [
27].
The identification of socioeconomic inequalities in the experience of disability among people with mental disorders may represent an opportunity to develop interventions to reduce its impact on well-being and associated personal and economic costs [
27,
28]. However, the use of different indicators, particularly those assessing disability, limits comparisons across settings. This also represents an important public health challenge in Portugal since the results from the World Mental Health Survey (WMHS) Initiative (2008/9) have shown a high prevalence rate of 12-month mental disorders (22.9%), associated with substantial societal costs, particularly relevant in relation to other countries [
8,
29]. Studies using days out of role as an indicator of disability, corresponding to the number of days in the last 30 that individuals were unable to work or carry out their normal activities due to health-related problems, have found mental disorders to be responsible for 20.2% of days out of role in Portugal, in comparison to 16.0% among high-income WMHS countries, assessed through its population attributable risk proportion [
8,
13].
This study aimed to examine the role of socioeconomic position in the experience of disability among people with mental disorders. It was hypothesized that the odds of reporting disability vary according to socioeconomic position, affecting disproportionately those more disadvantaged. Socioeconomic position is a concept widely used in epidemiological research and refers to social and economic factors that contribute for an individuals’ position within society [
30]. Indicators of socioeconomic position may not be inter-changeable [
31] and be differently associated with health outcomes across the life course [
30]. Therefore, the indicators included in this study, namely education, employment status, self-reported financial deprivation and subjective social status, were evaluated independently, integrating both the assessment of objective and subjective aspects of socioeconomic position.
Portugal is among the most unequal European countries and an absence of research and policy efforts to effectively tackle health inequalities have been reported [
32]. The results of this study may contribute to a better understanding on the effect of social inequalities in the experience of disability among people with mental disorders, for which current knowledge is still scarce, providing valuable insights for policy making.
Results
Table
1 presents the demographic, socioeconomic and clinical characteristics of the sample. Of the 3849 participants interviewed, 51.6% (
n = 2217) were women. The mean age of the participants was 46.38 (SD = 16.88) and the mean years of education were 8.76 (SD = 4.79). The majority of the participants were working at the time of interview (65.1%;
n = 1362). Financial deprivation was reported by 33.2% (
n = 732) of the participants and 34.9% (
n = 1344) perceive themselves to have a low social status in comparison to others in society. The prevalence rate of any 12-month mental disorder was 21.0% (
n = 788) and the prevalence of disability was 8.6% (
n = 212). Among people with any mental disorder, higher levels of unemployment (8.3%;
n = 40), financial deprivation (41.1%;
n = 303) and low subjective social status (37.6%;
n = 300) were found. Moreover, 14.7% (
n = 115) of these participants reported disability.
Table 1
Descriptive statistics of the demographic, socioeconomic and clinical characteristics of the WMHS Portugal sample and sub-sample of participants with any mental disorder
Demographic and socioeconomic characteristics | n | % | n | % |
Gendera |
Female | 2217 | 51.6 | 596 | 70.1 |
Male | 1632 | 48.4 | 192 | 29.9 |
Employment statusb |
Working | 1362 | 65.1 | 430 | 69.6 |
Unemployed | 172 | 6.8 | 40 | 8.3 |
Retired and others | 526 | 28.1 | 133 | 22.2 |
Financial deprivationb |
No | 1311 | 66.8 | 410 | 58.9 |
Yes | 732 | 33.2 | 303 | 41.1 |
Subjective social statusa |
High | 2463 | 65.1 | 482 | 62.4 |
Low | 1344 | 34.9 | 300 | 37.6 |
| Mean | SD | Mean | SD |
Agea | 46.38 | 16.88 | 42.82 | 15.19 |
Educationa | 8.76 | 4.79 | 9.50 | 4.64 |
Clinical characteristics | n | % | n | % |
12-month mental disordersb |
Any mental disorder | 788 | 21.0 | – | – |
Physical disordersb |
Any physical disorder | 1513 | 68.7 | 588 | 82.2 |
Disabilityb |
Presence of substantial disability | 212 | 8.6 | 115 | 14.7 |
Table
2 shows the association between the presence of any mental disorder and disability. After adjusting for age, gender and presence of any physical disorder, people with any mental disorder were almost 3 times more likely to report disability when compared to those without any mental disorder (OR = 2.82; 95%CI: 1.95–4.09).
Table 2
Odds ratio (OR) and respective 95% confidence interval (95%CI) of the association between the presence of any 12-month mental disorder and disability
Yes | 2.82 (1.95–4.09) *** |
No | Ref. |
Table
3 presents the odds ratios of the interaction effects between the presence of any mental disorder and each category of the indicators of socioeconomic position on disability. The results indicate that, after adjusting for age, gender and presence of any physical disorder, the association between disability and presence of any mental disorder varies significantly according to the category of the indicators evaluated, namely employment status (being “retired or others”) and perceived financial deprivation (being financially deprived). Among people with any mental disorder, those classified as “retired or others” were found to be 2.19 times more likely to report disability when compared to those in the working group (OR = 2.19; 95%CI: 1.06–4.48). Likewise, individuals financially deprived were 2.36 times more likely to report disability when compared to those who did not report this situation (OR = 2.36; 95%CI: 1.31–4.24). The same pattern was found regarding unemployment and low subjective social status, although not statistically significant. The results obtained in the specific years of education selected to report data were not statistically significant as well. However, among participants with any mental disorder, those with lower levels of education appeared to be more likely to report disability when compared to the highest level of education and a gradient was suggest by the results (e.g. no education: OR = 1.81, 95%CI: 0.57–5.82; 4 years of education: OR = 1.58, 95%CI: 0.65–3.84; and 12 years of education: OR = 1.19, 95%CI: 0.85–1.68).
Table 3
Odds ratios (OR) and respective 95% confidence intervals (95%CI) for disability, considering participants with any 12-month mental disorder, based on the interaction terms with education, employment status, self-perceived financial deprivation and subjective social status
Presence of any mental disorder * Education |
No education | 1.81 (0.57–5.82) |
4 years | 1.58 (0.65–3.84) |
9 years | 1.32 (0.76–2.29) |
12 years | 1.19 (0.85–1.68) |
17 years | Ref. |
Presence of any mental disorder * Employment status |
Working or students | Ref. |
Unemployed | 1.87 (0.78–4.53) |
Retired or others | 2.19 (1.06–4.48) * |
Presence of any mental disorder * Financial deprivation |
No | Ref. |
Yes | 2.36 (1.31–4.24) * |
Presence of any mental disorder * Subjective social status |
High | Ref. |
Low | 1.45 (0.81, 2.60) |
Discussion
The objective of this study was to evaluate the effect of socioeconomic position on the disability experienced by people with mental disorders. As hypothesised, the findings suggest that the likelihood of reporting disability varies according to socioeconomic position, in particular employment status and perceived financial deprivation. Participants with any mental disorders in the category of “retired or others” and who perceived themselves as financially deprived had two times higher odds of reporting disability, when compared to those working and not financially deprived, respectively. Moreover, despite not reaching statistical significance, an education gradient seems to be suggested by the results, given that among people with any mental disorder, those with the lowest years of education were almost two times more likely to report disability, with the odds decreasing alongside the number of years of education.
The results are in line with previous research. Studies have shown a lower labour force participation of individuals with mental health problems due to early retirement and sickness absence, among other factors [
15,
18,
19,
27]. It is important to highlight that in this study early retirement was not evaluated separately but the association was adjusted for age. The findings on perceived financial deprivation are aligned with research suggesting higher levels of economic disadvantage among individuals with disability due to mental disorders [
17,
18,
27]. Regarding education, despite the absence of statistical significance, the results are in line with studies that found an education gradient in the risk of early retirement and long term sickness absence due to mental health problems [
25,
26]. It has been suggested that individuals with lower socioeconomic position are more likely to have demanding occupations, both physically and psychosocially, or may not have the same opportunities to accommodate their ill-health on their task requirements and working conditions [
25]. This may be particularly relevant within the context of the Portuguese welfare system, characterized by providing the smallest public expenditure per capita in social protection in Western Europe, alongside other Southern European countries [
44].
The findings of this study should be interpreted within several limitations. The cross-sectional design limits causal inference, namely to understand if the differences in the experience of disability among people with mental disorders according to socioeconomic position are related to factors existing before the onset of disability, to the onset itself or its duration over time [
45]. However, two main pathways may operate co-currently: Among people with mental disorders, those with lower socioeconomic position may be more likely to experience disability. Low socioeconomic position has been associated with higher prevalence of mental disorders and disability, due to risk factors such as low educational level, unemployment, precarious working conditions and lower standards of living [
18,
27,
28]. Also, the experience of disability among individuals with mental disorders may further aggravate socioeconomic inequalities. The onset and duration of mental disorders and disability have been associated with more adverse economic outcomes such as job loss, reduced income and healthcare expenditure [
17,
27,
45].
Furthermore, similar to other research in this area, disability was evaluated in the previous month, whereas mental disorders are 12-month based. For episodic conditions, the past month disability may not include the time period of the disorder, while using a 12-month diagnosis allows the inclusion of remitted disorders that may have residual adverse effects on disability [
10,
13]. Besides, the changes made to reduce respondent burden in the WHODAS-II in the WMHS, such as the use of filter questions, impaired the measurement properties with scores having highly skewed distributions with low mean scores and large proportions of zero scores [
37]. To address this issue, the cut-off for defining substantial disability (percentile 90th) has been recommended. However, this procedure may mask cross-national differences and caution is needed when comparing the results obtained in this study with those from other countries [
37]. Another possible limitation is the use of a broad category of any mental disorder, which does not consider differences that may occur in the experience of disability associated with specific conditions. Finally, this study fails to account for the recent macroeconomic changes in Portugal, one of the European countries most affected by the global financial crisis [
46]. Mental health and well-being are likely to deteriorate more immediately and severely than other health outcomes during periods of economic recession [
47], contributing to wider health and social inequalities that may not be represented in the results. This scenario may be further aggravated by the absence of effective policies to address social and health inequalities in Portugal [
32]. In spite of these limitations, to our knowledge, this study was the first to assess the effect of socioeconomic position in the experience of disability among people with mental disorders in Portugal. Different indicators of socioeconomic position were used, complementing research in this area and drawing attention to the need to conduct longitudinal studies to ascertain the causal pathways involved in these associations. Furthermore, a nationally representative of the Portuguese population and robust instruments to access disability and mental disorders were used. The use of a multi-dimensional instrument to assess disability also represents a major strength of this study since most research uses indicators of productivity loss, which are difficult to compare and only partially assess the experience of disability.
Acknowledgements
The authors wish to thank the World Mental Health Survey Initiative staff for their assistance with instrumentation, fieldwork, and data analysis. A complete list of funding support and publications can be found at:
http://www.hcp.med.harvard.edu/wmh.