Understanding Stigma: A Pooled Analysis of a National Program Aimed at Health Care Providers to Reduce Stigma towards Patients with a Mental Illness

Abstract

Background and Objective: The problem of mental illness-related stigma within healthcare is an area of increasing attention and concern. Understanding Stigma is an anti-stigma workshop for healthcare providers that uses social contact as a core teaching element, along with educational and action-oriented components. The objective of our study was to determine the impact of this program on healthcare providers’ attitudes and behavioural intentions towards patients with a mental illness, and also to ascertain whether various participant and program characteristics affected program outcomes. Our paper reports the results of a pooled analysis from multiple replications of this program in different Canadian jurisdictions between 2013 and 2015. Material and Methods: We undertook a pooled analysis of six separate replications of the Understanding Stigma program. All program replications were evaluated using a non-randomized quasi experimental pre- post- follow-up design. The Opening Minds Scale for Health Providers (OMS-HC) was used as the main assessment tool. Study-level and individual-level meta-analysis methods were used to synthesize the data. First, the ‘metan’ command was used to show outcomes by study, using a forest plot. Then, a pooled dataset was produced and analyzed using a random intercept linear mixed model approach with each program being modelled as a random effect. Program and participant characteristics were examined as independent variables using this approach. These were each entered individually. Individual tests included pre to post change by program version (original or condensed), by occupation (nurses versus other healthcare providers), by gender, age, and previous diagnosis of a mental illness. Results: Program effect sizes ranged from .19 to .51 (Cohen’s d), with an overall combined effect size of .30. The results of the mixed model analysis showed the improvement from pre to post intervention was statistically significant for the total scale and subscales. Analysis of program and participant factors found that version type, healthcare provider type, gender, and previous diagnosis of a mental illness were all non-significant factors on program outcomes. A significant inverse association was revealed between increasing age and score change. Results also showed a significant positive linear relationship between baseline score and improvement from pre to post intervention. Maintenance of scores at follow-up was observed for participants who attended a booster session. Conclusion: The results are promising for the effectiveness of this brief intervention model for reducing stigmatizing attitudes and improving behavioural intentions among nurses and other healthcare providers.