Is any job better than no job? A systematic review

ObjectivesThis systematic review addresses the question "Is any job better than no job?" Specifically, it compares health and well-being outcomes in those who are unemployed with those who are in jobs that could be considered poor or low quality and the impact of any movement between them. MethodWe conducted a systematic review following a PROSPERO-registered protocol (CRD42020182794). Medline, Embase, PsycINFO, HMIC, ASSIA, TRIP, Google Scholar and 10 websites were searched in April 2020 and again in May 2021 without date limits. Two reviewers working independently screened search results against the inclusion/exclusion criteria. A checklist for quantitative studies reporting correlations was used to critically appraise articles included at full text. We undertook synthesis without meta-analysis (narrative synthesis) and explored a range of variables (for example, study design and quality, type of outcome measure) that we considered might have an impact on the association between exposure and outcome. ResultsWe included 25 studies reported in 30 journal articles. All 25 studies involved secondary analysis of data from national cohorts, including six from the UK. The most frequent outcomes reported were measures of mental well-being. There was considerable heterogeneity across included studies in terms of design, population, definition of poor/bad or low quality job and outcome types and measures. Overall the quality of the included studies was moderate. The evidence base is inconsistent. There are studies that suggested either labour market position might be preferable, but a number of studies found no statistically significant difference. Cohort and case- control studies looking at mental well-being outcomes showed some support for a poor job being better than unemployment. However, we did not find sufficient numbers of well-designed studies showing a strong association to support a causal relationship. Most included study designs were unable to distinguish whether changes in employment status occurred before a change in outcome. Three studies looking at employment transitions found that moving to a poor job from unemployment was not associated with improved mental health, but moving from a poor job to unemployment was associated with a deterioration. ConclusionEvidence that better health and well-being outcomes are more likely to be associated with a poor/bad or low quality job than with unemployment is inconsistent. Studies conducted in the UK suggest that a poor job is not significantly associated with better health and well-being outcomes than unemployment. The studies we identified do not allow us to distinguish whether this lack of association is the result of a state welfare regime preventing some of the worst ills associated with unemployment, or a reflection of job quality. The evidence base has significant limitations in study design and conduct. In summary, the evidence we found suggests it is not safe to assume that, in the UK, any job will lead to better health and well-being outcomes than unemployment.