CRANIUM: a quasi-experimental study to improve metabolic screening and HIV testing in community mental health clinics compared to usual care

Abstract

Background: Individuals with serious mental illness often do not receive guideline-concordant metabolic screening and human immunodeficiency virus (HIV) testing, contributing to increased morbidity and premature mortality. This study evaluates the effectiveness of CRANIUM (Cardiometabolic Risk Assessment and treatment through a Novel Integration model for Underserved populations with Mental illness), an intervention to increase metabolic screening and HIV testing among patients with serious mental illness in a community mental health clinic compared to usual care.

Methods: The study used a quasi-experimental design, prospectively comparing a preventive care screening intervention at one community mental health clinic (n = 536 patients) to usual care at the remaining clinics within an urban behavioural health system (n = 4,847 patients). Psychiatrists at the intervention site received training in preventive health screening and had access to a primary care consultant, screening and treatment algorithms, patient registries, and a peer support specialist. Outcomes were the change in screening rates of A1c, lipid, and HIV testing post-intervention at the intervention site compared to usual care sites.

Results: Rates of lipid screening and HIV testing increased significantly at the intervention site compared to usual care, with and without multivariable adjustment [Lipid: aOR 1.90, 95% CI 1.32-2.75, P = .001; HIV: aOR 23.42, 95% CI 5.94-92.41, P < .001]. While we observed a significant increase in A1c screening rates at the intervention site, this increase did not persist after multivariable adjustment (aOR 1.37, 95% CI .95-1.99, P = .09).

Conclusions: This low-cost, reverse integrated care model targeting community psychiatrist practices had modest effects on increasing preventive care screenings, with the biggest effect seen for HIV testing rates. Additional incentives and structural supports may be needed to further promote screening practices for individuals with serious mental illness.

Keywords: Collaborative care; Diabetes; Health care disparities; Integrated care; Serious mental illness.

Fig. 1

A1c screening, lipid screening, and HIV testing rates before and after intervention by intervention arm^a (Abbreviations: A1c: Haemoglobin A1c, HIV: Human Immunodeficiency Virus. (^aOnly the CRANIUM arm received the intervention.) (^bNo statistical significance by intervention arm for A1c screening before intervention (χ² = .04, P = .84).) CRANIUM lipid screening before intervention is significantly lower than usual care (χ² = 13.1, P < .001). CRANIUM HIV testing before intervention is significantly higher than usual care (χ² = 18.5, P < .001))