Background Job loss, debt and financial difficulties are associated with increased risk of mental illness and suicide. During economic recessions the incidence of these rises. Interventions targeting people in debt or unemployed might help mitigate these effects. We systematically reviewed randomised controlled trials (RCTs) for evidence of effects on mental health outcomes of interventions for the unemployed, people in debt or on low incomes in the general population.

Methods We searched MEDLINE, PsycINFO, EMBASE, CENTRAL, Science, Social Science, and Arts and Humanities Citation Index on Web-of-Science (January 16^th 2016) and citations and reference lists of included studies. We included randomised controlled trials (RCTs) of public health or health service interventions designed to mitigate the effect of unemployment, debt or austerity measures in the general population with mental health outcomes. Studies on people with serious mental health problems, ill health or rehabilitation interventions were excluded. We screened abstracts for relevance, assessed eligibility, and extracted data in duplicate. Bias was assessed with the Cochrane risk-of-bias tool. We prepared a structured narrative synthesis. Data were too sparse and interventions too heterogeneous for meta-analysis.

Results Database searching identified 2389 records. We read 136 full-text papers and included eleven RCTs (n = 5303 participants) for our review. Seven were published before the year 2000. Five assessed ‘job-club’ interventions, two group cognitive behavioural therapy (CBT) and single RCTs assessed emotional competency training, expressive writing, guided imagery and debt advice. Reports omitted much detail necessary to assess bias but all studies were at high risk-of-bias. ‘Job-club’ interventions (n = 4222) were associated with lower levels of depression than control interventions up to 2 years post-intervention; effects were strongest amongst those identified at baseline at increased risk of depression (improvements of up to 0.2 to 0.3 SD in depression scores). In one RCT (n = 289), compared to social support CBT was associated with reduced GHQ-30 scores at 3 months (ANOVA, difference in means −1.44 F = 3.91 p < 0.05). A second RCT of CBT versus first-aid training (n = 195) showed no effect on the mental health (SF-36 MCS difference in means −2.17 95% CI −7.13 to 2.79). Single RCTs of four other interventions showed no evidence of benefit.

Conclusion ‘Job club’ interventions may be effective in reducing the incidence of depression in unemployed people, particularly those at high risk of depression. Evidence for CBT-type interventions is mixed; further trials are needed. The studies are dated and at high risk-of-bias and future studies should follow CONSORT guidelines.