Table 4

Effectiveness of interventions on primary and secondary outcomes

Study (year)Quality ratingEffect on mental health outcomes
Welfare and advice services located in healthcare settings
Abbott and Hobby (2000)53 LowFor participants with increased income after 6 months (n=48), improved mean scores for some SF-36 domains: vitality (pre: 20.8, post: 28.5, t=3.3, p=0.002), emotional role functioning (pre: 36.8, post: 51.4, t=2.2, p=0.037) and mental health (pre: 45.9, post: 53.1, t=2.9, p=0.005), but not social functioning (pre: 29.4, post: 32.0, p>0.05). No improvements observed at 6 months in participants without increased income (n=20), or in either group after 12 months.
Caiels and Thurston (2005)56 LowFor those participants who complete preintervention and postintervention questionnaires (n=81), no change in SF-12 score (pre: mean=34.1, post: mean=35.6, p=0.335). Over the 12-month period, £356 754 gained on behalf of clients.
Abbott et al (2006)54 LowAt 6-month follow-up, no improvements in SF-36 domains (vitality, social functioning, emotional role and mental health) among those who saw an increase in income following participation (n=160) compared with those who did not (n=84). At 12-month follow-up, improvements observed in emotional role (adjusted mean difference: 16.37 (2.72–30.01), p=0.02) and mental health (adjusted mean difference: 6.85 (0.72–12.98), p=0.03) for participants with increased income (n=134), compared with those who did not (n=50). No change in vitality and social functioning scores between the two groups.
Harris (2013)57 LowImprovements in mean emotional well-being scores (preadvice: 47.8, postadvice: 61.3, t=3.3, p=0.001) and role limitation due to emotional problems (preadvice: 35.1, postadvice: 62.2, t=3.2, p=0.002) observed at case closure (n=65). 35% of participants achieved financial gains during the programme (mean gain £4686 per benefiting client). 17 weeks of staff time saved over the 3-year study period, resulting in an annual savings of £8700.
Ryan et al (2012)55 LowCompared with baseline, participants (n=67) reported decreased stress (mean difference: 8.1, p<0.001) and increased well-being (mean difference: 1.8, p<0.001) after participation in the programme.
Krska et al (2013)52 LowIn the 6-month period following participation, no changes observed in mean number of primary care mental health appointments, mental health referrals or antidepressant prescriptions per patient. However, a decrease was observed in mean prescriptions for hypnotics/anxiolytics per patient (−0.16, p<0.05).
Woodhead et al (2017)50 ModerateOverall, no evidence of effect on probable common mental disorder or well-being. However, relative to controls, probable common mental disorder reduced for female (rOR=0.37 (95% CI 0.20 to 0.70)) and Black advice recipients (rOR=0.09 (95% CI 0.03 to 0.28)). Well-being increased for participants who received a positive outcome from the advice service (β coefficient=1.29, 95% CI 0.25 to 2.32, p=0.015).
No evidence that the intervention impacted 3-month, self-reported consultation frequency. £2689 average financial gain per participant over the study period. £15 income gain per £1 provided by the funder.
Link worker social prescribing
Grant et al (2000)47 ModerateAfter adjustment for baseline scores, participants showed greater reductions in anxiety scores (−1.9, 95% CI −3.0 to −0.7, p=0.002), but not depression scores (−0.9, 95% CI −1.9 to 0.2, p=0.116). Participants showed greater improvement in functional health components relating to pain (−0.5, 95% CI −0.8 to −0.1, p=0.005), emotional feelings (−0.5, 95% CI −0.8 to −0.2, p=0.003), ability to carry out everyday activities (−0.5, 95% CI −0.8 to −0.2, p=0.001) and feelings about general health (−0.4, 95% CI −0.7 to −0.1, p=0.003). No difference was observed between groups in social support.
Grayer et al (2008)51 LowCompared with baseline, fewer participants experiencing psychological distress based on both the GHQ-12 (pre: 82.6%, post: 52.2%, difference: 30.4% (16.9–43.9)) and the CORE-OM (pre: 85.1%, post: 67.6, difference: 17.5% (7.4%–27.7%)). The proportion of patients prescribed psychotropic medication declined (pre: 34.7, post: 18.8%, difference: 15.8% (6.0–25.6)), while the proportion receiving onward mental health referrals increased (pre: 7.9%, post: 19.8%, difference: 11.9% (1.9–21.9)).
Telephone debt advice services
Jinhee Kim et al (2013)58 LowAt 18-month follow-up, small improvements in health scores were observed in participants (n=70) (pre: 10.60, post: 10.98, t=2.62, p<0.05) but not in controls (n=100) (pre: 10.68, post: 10.60, t=0.29, p>0.05).
Pleasence and Balmer (2007)48 LowTrial stopped early due to attrition. However, after 20 weeks, no changes were observed in anxiety or general health score for either the intervention group (n=119) or the control group (n=115).
Food insecurity interventions
Roncarolo et al (2016)61 LowImproved mental health scores were observed for participants in traditional food security interventions (pre: 58.1, post: 63.9, adjusted β coefficient: 5.3 (3.1–7.4)), while no changes were observed in participants of alternative interventions (pre: 66.1, post: 71.1, adjusted β coefficient: 4.2 (−1.3 to 9.7)).
Active labour market programmes
Rose (2019)59 LowLabour programmes that most closely replicate employment (wage subsidies and subsidised self-employment) had the largest effect on improving the well-being of participants. When results were disaggregated by sex, no differences in well-being were observed participating in the ALMP schemes, compared with non-participants.
Saloniemi et al (2014)60 LowOverall, no change in psychological distress, sense of coherence or stress at the end of the training course, compared with baseline. However, improvements in all three measures were seen among participants with a tertiary education and those who were previously employed in a ‘white-collar’ occupation.
Vinokur et al (2000)49 ModerateAt 2-year follow-up, compared with controls, participants had significantly higher role functioning and lower depressive symptoms, and were less likely to have experienced a probable major depressive episode.
  • CORE-OM, Core Outcome Measurement tools; GHQ-12, General Health Questionnaire; SF-12, 12-item Short Form Survey; SF-36, 36-item Short Form Survey.