Disengagement from the Ribeirão Preto early intervention program for psychosis: A retrospective cohort study

Luis Felipe Scarabelot,Jéssica Morais Araújo,Livio Rodrigues Leal,Rebeca Mendes de Paula Pessoa,Fabiana Corsi-Zuelli,Camila Marcelino Loureiro,Gabriel Elias Corrêa-Oliveira,Cristina Marta Del-Ben
DOI: https://doi.org/10.1016/j.ajp.2024.104119
Abstract:Background: Treatment discontinuation within Early Intervention Services (EIS) for psychosis poses a significant challenge to achieving better outcomes in the early stages of psychotic disorders. Prevalence and predictors of early disengagement from EIS located in low- and middle-income countries (LMICs) remain poorly investigated. We aimed to examine the rates and predictors of disengagement from the Ribeirão Preto Early Intervention Program for Psychosis (Ribeirão Preto-EIP) in Brazil. Methods: We conducted a retrospective cohort study using data from patients referred to the Ribeirão Preto-EIP between January 01, 2015, and December 31, 2018. Exclusion criteria were individuals with a single consultation, a diagnosis other than a psychotic disorder, and documented cases of death. Results: Our sample comprised 234 patients, with an overall median follow-up time of 14.2 months. Early treatment disengagement was observed in 26.5 % (n=62), with a median time to disengagement of 5.25 months. Univariable analysis identified non-white skin color (HR=2.10, 95 %CI 1.26-3.49), positive THC screening (HR=2.22, 95 %CI 1.23-4.01), and substance-induced psychosis (HR=2.15, 95 %CI 1.10-4.21) as significant predictors. In multivariable analysis, only non-white skin color remained a significant predictor of early disengagement (HR=1.87, 95 %CI 1.08-3.27). Conclusions: The observed rates of early disengagement in our sample are similar to those reported in wealthy countries, but higher than previously reported for LMICs. Non-white skin color predicted early disengagement in our sample, probably due to social disadvantages. Our data highlights the need for enhanced research elucidating the specific features of EIS in LMICs.
What problem does this paper attempt to address?