Examining Trends in Research Retention and Attrition Among Individuals with First-Episode Psychosis in Kwazulu-Natal, South Africa

Background and Purpose: Retention is a significant methodological challenge in mental health research, particularly in low- and middle-income countries (LMICs), where infrastructural and resource limitations exacerbate participant attrition. Among individuals with psychotic disorders, high dropout rates are often driven by symptom severity, logistical barriers, and psychosocial factors such as mental illness stigma and distrust of researchers. Despite the growing burden of mental illness in LMICs, limited attention has been paid to patterns of study attrition in mental health research within these settings. This study aimed to examine trends in research retention and attrition among adults with first-episode psychosis (FEP) in KwaZulu-Natal, South Africa. We sought to describe rates of follow-up participation and identify sociodemographic and clinical correlates of study retention.

Methods: This study used data from a longitudinal cohort project, The HIV in FEP Study, involving adults aged 18–45 years diagnosed with FEP and receiving outpatient or inpatient psychiatric care at five regional hospitals in KwaZulu-Natal province. The analytic sample (n=97) included participants enrolled from October 2020 to December 2022, following the suspension of data collection due to the COVID-19 pandemic. Participants were assessed at four time points: baseline, and at 3-, 6-, and 12-months post-baseline. Measures included the Positive and Negative Syndrome Scale (PANSS), Patient Health Questionnaire (PHQ-9), Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST), WHO Quality of Life scale, and International HIV Dementia Scale. Sociodemographic data were also collected. Descriptive statistics, bivariate analyses (Spearman correlation and chi-square tests), and negative binomial regression models were used to examine associations between participant characteristics and the number of completed follow-up assessments.

Results: Of the 97 participants, 41.2% (n=40) completed all follow-up assessments, 36.1% (n=35) completed none, and the remaining 22.7% completed one or two follow-ups. Cross-sectional completion rates declined at each time point: 56.7% at 3 months, 52.6% at 6 months, and 47.4% at 12 months. Bivariate analyses identified two significant correlates of higher retention: tertiary education and receipt of long-acting injectable (LAI) antipsychotic medication. Due to the small number of LAI cases (n=4), only education was further analyzed. Negative binomial regression indicated that participants with tertiary education had a 59% higher rate of follow-up completion (Incident Rate Ratio=1.59) compared to those without tertiary education.

Conclusions and Implications: Retention in longitudinal research among individuals with FEP in South Africa is a major challenge, with over one-third of participants lost to follow-up after baseline. Higher educational attainment was significantly associated with better study retention. These findings underscore the need for contextually appropriate and resource-sensitive retention strategies in LMICs. Enhancing retention may involve expanding follow-up methods (e.g., phone or digital platforms), integrating community-based support, or developing centralized health information systems to improve patient tracking across healthcare settings and services. Future research should explore participant-informed approaches to engagement and examine the intersection between research participation and service retention in mental health care.