Feasibility study to evaluate capabilities for conducting psychiatric clinical research within the Rwandan mental healthcare system

{"Background": "Schizophrenia outcomes remain poor in resource-limited settings, where treatment is often limited to poorly tolerated first-generation antipsychotics. Rwanda has made significant strides in mental healthcare but still faces severe resource constraints\u2014only 12 practicing psychiatrists for a population of nearly 13 million as of 2021. Long-acting injectable (LAI) antipsychotics such as paliperidone palmitate once-monthly (PP1M) and once-every-3-months (PP3M) have demonstrated benefits in high-resource settings, but their value in resource-limited settings requires local testing due to differences in delivery challenges, storage requirements, administration procedures, and cultural attitudes. This feasibility study was designed to evaluate whether a comprehensive clinical trial of PP1M/PP3M could be successfully conducted within the Rwandan mental healthcare system.", "Methods": "This was an open-label, uncontrolled, multicentre, interventional feasibility study conducted at three sites in Rwanda: CARAES Ndera Neuropsychiatric Hospital, Ruhengeri Referral Hospital, and Kibungo Referral Hospital. Eligible participants were men and women of non-childbearing potential aged 18\u201370 years with a DSM-IV diagnosis of schizophrenia confirmed by the Mini International Neuropsychiatric Interview. The study included four phases: screening, a 1-week oral risperidone run-in (3 mg) to test tolerability, a 17-week lead-in with flexibly dosed PP1M (50\u2013150 mg eq), and a 24-week maintenance phase with PP3M (175\u2013525 mg eq). Feasibility endpoints included compliance with governmental and institutional requirements, acceptable supply chain delivery, adequate site infrastructure, adequate staff training, proper medication administration, and successful completion of study procedures and scales. Nine assessment scales were administered including the SMILE (patient and caregiver versions), CGI-S, SDS, WHO QoL BREF, CSRI, CAQ, and satisfaction scales. A sample size of 30 participants was considered adequate for feasibility assessment.", "Key Results": "33 participants were enrolled (mean age 42.0 years, range 22\u201366; 51.5% male, 48.5% female; 100% Black African). The study was terminated early by the sponsor because certain aspects of study conduct needed to be addressed to maintain Good Clinical Practice (GCP) requirements. Prior to termination, 97% (32/33) of participants had received PP3M and tolerated it well. One participant was discontinued due to disease relapse, one withdrew consent, and two completed the study; the remaining 29 were discontinued at early termination. All 31 participants who completed or were discontinued were offered treatment through a post-trial access programme. Key feasibility findings included: significant protocol deviations related to timely data entry and review; rater exhaustion and burnout due to study demands on top of clinical workloads; inadequate site infrastructure (computers, internet access, laboratory assays); insufficient GCP training for investigators without prior interventional trial experience; language barriers in training; and study scales that were poorly translated into Kinyarwanda and poorly adapted to Rwandan economic realities. Despite these challenges, risperidone/PP1M/PP3M were reliably supplied and administered without difficulty, and eligible patients were readily identified. Exploratory efficacy data showed significant improvement: SMILE patient version improved from 7.9 (SD 8.54) at baseline to 2.8 (3.09) at week 21 (change \u22125.1, SD 7.33, p<0.001); SMILE caregiver version improved from 8.9 (9.85) to 2.9 (3.20) (change \u22126.0, SD 8.90, p<0.001); CGI-S improved from 3.1 (0.90) to 2.7 (0.65) (change \u22120.4, SD 1.12); SDS improved from 7.8 (7.13) to 4.6 (4.01) (change \u22123.1, SD 6.86, p<0.01). Safety data showed 12 participants (36.4%) reported at least one treatment-emergent AE: gastrointestinal disorders (12.1%, n=4), general disorders (9.1%, n=3), infections (6.1%, n=2), metabolism disorders (6.1%, n=2), nervous system disorders (6.1%, n=2), and psychiatric disorders (6.1%, n=2). One participant (3.0%) reported a serious AE of hyperglycaemia deemed related to study treatment. Clinician satisfaction was high (mean 4.8/5 at baseline, 5.0/5 at week 21).", "Clinical Implications": "This feasibility study demonstrated that conducting a GCP-compliant psychiatric clinical trial in Rwanda is achievable but requires substantially more preparation than initially anticipated. The identified limitations\u2014including inadequate staff training, insufficient site infrastructure, poor scale translation, and lack of prior trial experience\u2014were all considered surmountable with additional resources and planning. The study directly informed six key modifications for a larger ongoing interventional trial: inclusion of a national principal investigator to improve communication; comprehensive GCP and research methodology training including e-learning and in-person sessions; Rwanda-based site monitors; health economic scales adapted to the Rwandan context; improved prestudy infrastructure assessment; and clinical team building that accounts for investigators' regular hospital schedules. The study underscores that small preliminary feasibility studies are essential in resource-limited settings to identify and address logistical and design problems before committing to large trials, thereby improving the likelihood of success and ultimately bringing better-tolerated treatments to underserved populations."}