People with elevated psychological distress had much higher odds of being unemployed compared with people with mild or no psychological distress. This effect of psychological distress on unemployment was significantly larger for men than for women. The lost productivity associated with moderate and severe psychological distress represented a loss of nearly 7% of GDP, far higher than estimated GDP reportedly lost due to malaria
Our study represents one of the first analysis of this relationship between mental health and employment using a nationally representative sample of adult men and women in a low-income country. Additionally, our analysis examines severe psychological distress but also identifies an association between unemployment and moderate psychological distress, both of which have been shown to be associated with negative health outcomes in previous research in high-income countries
]. The results from this study demonstrate the importance of addressing mental health in the adult population. Previous literature has shown positive results when high-income countries prioritize mental health. Beginning in 2007, the United Kingdom began a large scale financial investment in improving access and treatment for mental health, specifically addressing depression and anxiety
]. In pilot data from two locations, Doncaster and Newham, the increased investment showed a significant reduction in the number of clinical cases of depression and anxiety as well as a significant increase in the number of patients who returned to work
]. Additionally, after 2 years of prioritized national focus on mental health, the program has nearly met targets for number of individuals seen and for recovery rates and has exceeded targeted numbers for moving individuals off of sick pay and state benefits
At present, the country of Ghana has shown a commitment to addressing the mental health burden with its recent passage of the Mental Health Bill, and thus may serve as an example for other low-income countries in the region. Legislation appropriating finances and other resources is a first step for ensuring better care among people suffering from poor mental health and for reducing its associated stigma. Our analysis suggests that there is a need for investment of resources to address lost productivity associated with mental health.
Despite the strong statistical association between psychological distress and unemployment, causality cannot be established as the analysis is based on cross-sectional data. Studies assessing mental health and employment in high-income countries
] demonstrate that causality is complex. An inability to find work may result in higher levels of psychological distress or, conversely, those with higher levels of psychological distress may be less able to find work; in some cases, both causal paths may be present. Additionally, the direction of the association between employment and psychological distress may not be unilateral. Higher levels of psychological distress among an employed population have been observed due to difficult working conditions
] or problems associated with underemployment
], thus we could have underestimated the association between unemployment and psychological distress. Although it is important not to infer causality from our findings, this is nonetheless one of the first large-scale studies establishing a robust association between psychological distress and employment in sub-Saharan Africa. Additionally, the magnitudes of the estimated associations are large, and should serve as motivation for prospective studies to evaluate the gains in productivity that might be achieved when mental illness is adequately managed or treated.
Our findings should be interpreted in light of additional limitations. First, the survey did not measure average number of hours worked per day. If this is lower for people with moderate or severe psychological distress, then we may have underestimated employment associations. Second, estimates of GDP can vary based on assumptions made during its computation. We calculated the percent of GDP represented by the lost productivity among people with moderate and severe psychological distress with methods that have been commonly used to study the indirect financial costs associated with disease
], and we applied a conservative estimate of all working age individuals as the population denominator. Although exact numbers vary based on how GDP association is calculated
], WHO estimates show that the malaria accounts for less than 4% of the overall GDP in Ghana
]. Highlighting the GDP cost of lost productivity associated with psychological distress can be useful for prioritizing mental health spending over other illnesses. It is important to note that our estimates do not provide casual evidence; however, the magnitudes of the estimated associations are large and should motivate prospective cohort studies and randomized controlled trials to rigorously evaluate the gains in productivity that might be achieved by addressing mental health and other barrier to full employment. Because our estimates depend on extrapolations to the full population, they are speculative, not definitive, and additional studies to replicate these findings would be helpful. Last, our response rate was 85%, and respondents differed significantly from non-respondents in marital status, education, region, religion and wealth, giving rise to potential for response rate bias; however, we did adjust for these variables in the multivariate analysis to mitigate this concern.