Arts on prescription intervention for primary healthcare patients with poor mental health or social isolation: a mixed-method study.

BACKGROUND

Primary healthcare providers are increasingly challenged in supporting patients with psychosocial needs. Arts on Prescription (AoP) has been shown to improve primary healthcare patients' mental health wellbeing. The aim of the current study is to understand the psychosocial effect of participating in an Arts on Prescription programme.

METHODS

A total of 112 primary healthcare patients from 18 primary healthcare centres in Scania with mental health diagnoses depression and anxiety or social isolation participated in a 10-week group-based arts programme, twice a week for 2 h. A questionnaire with the Short Warwick Edinburgh Mental Wellbeing Scale (SWEMWBS), the Salutogenic Health Indicator Scale (SHIS) (for baseline and follow-up) and 14 sociodemographic and self-rated health covariates were collected as baseline. We also conducted 28 semi-structured interviews. We analysed data using paired t-test and a general linear regression model for change in SWEMWBS and SHIS. Qualitative data was analysed using a thematic approach.

RESULTS

The paired t-test showed highly significant results (p < 0.001) for increase in both SWEMWBS and SHIS. The general linear regression. models show that women and participants with poorer self-rated health (SRH), more contacts with the healthcare system, other referrals from the primary healthcare centre, and no previous arts and culture engagement displayed significantly stronger associations with increase in SWEMWBS but not SHIS. Qualitative results highlight use of other interventions and difficulties navigating the health system.

CONCLUSIONS

Our findings support a proportionate universalism (scale and intensity proportionate to the degree of need) approach indicating that AoP programmes could be valuable additions to healthcare pathways enhancing wellbeing for vulnerable populations. Findings should be interpreted with caution due to small sample size.