Antipsychotic ADRs Affect 61.7% of Ugandan Psychiatric Outpatients, Study Finds
核心洞察
A cross-sectional study at Mbarara Regional Referral Hospital (搜索) in Uganda found 256 of 415 antipsychotic-treated outpatients (61.7%) experienced at least one adverse drug reaction.
Most reactions were moderate in severity (63.7%) and rated probable on the Naranjo causality scale, with sedation the most frequently reported adverse event.
Half of all identified ADRs were classified as definitely preventable, pointing to gaps in prescribing, monitoring and follow-up in the outpatient setting.
A cross-sectional study conducted at the psychiatry clinic of Mbarara Regional Referral Hospital (搜索) (MRRH) in southwestern Uganda has found that 61.7% of outpatients taking antipsychotic medication experienced at least one adverse drug reaction (ADR), with the majority of events rated moderate in severity and half judged definitely preventable.
The study, which enrolled 415 patients between March and May 2025, is the first to specifically examine antipsychotic-related ADRs among psychiatric outpatients at MRRH, a 600-bed tertiary referral hospital serving a population exceeding four million people across districts including Mbarara, Bushenyi, Ntungamo, Kiruhura, Ibanda, Buhweju, Rubirizi, Mitooma and Isingiro, as well as patients from Kabale, Masaka, Fort Portal and neighboring Rwanda. The clinic registers approximately 100 patients weekly and over 1,200 new patients annually, with more than 90% of outpatients prescribed antipsychotics.
Prevalence Exceeds Prior Reports From India
Of the 415 participants, 256 (61.7%) experienced at least one ADR. The authors note this prevalence is higher than the 51.9% previously documented in a study conducted in India among patients receiving antipsychotics, which they attribute to differences in patient characteristics, prescribing patterns and healthcare systems. Patients in the Ugandan setting may have been more likely to receive higher antipsychotic doses, combination therapy, or concurrent oral and intramuscular formulations, the researchers suggest, alongside variations in pharmacovigilance practices and availability of routine monitoring and follow-up services.
By contrast, the observed prevalence was markedly lower than the 91.8% documented among psychiatric patients prescribed psychotropic medications at Mizan Tepi University Teaching Hospital in Ethiopia. The authors note that the Ethiopian study examined ADRs across a wider spectrum of psychotropic agents, whereas the present study focused specifically on antipsychotics, a subset of psychotropic drugs.
Sedation emerged as the most common ADR recorded, a finding the authors describe as consistent with the pharmacological profile of first-generation antipsychotics (FGAs), the class predominantly prescribed among participants. High utilization of FGAs at the facility is driven by government supply and affordability constraints. A similar pattern of sedation among leading ADRs has been documented in studies conducted in Ethiopia.
Causality, Severity and Preventability
Causality was assessed using the Naranjo Adverse Drug Reaction Probability Scale. Of the 256 ADRs identified, the majority (68.0%) were classified as probable, 72 (28.1%) as mild and 10 (3.9%) as definite. The authors note these findings are consistent with previous studies reporting that most ADRs among patients receiving antipsychotics were rated probable, including 74.3% in Oman and 55% in India. They suggest many participants fulfilled key criteria for a "probable" classification, such as prior conclusive reports of the reaction, onset following administration of the suspected drug, and recurrence upon drug re-administration.
Severity was graded with the modified Hartwig and Siegel severity assessment scale. The majority of identified ADRs (163, 63.7%) were classified as moderate, while 93 (36.3%) were mild. The predominance of moderate reactions was comparable to findings from Oman, where 74.0% of reported ADRs were rated moderate. The authors attribute this in part to high use of typical (first-generation) antipsychotics, which are generally associated with a higher burden of clinically significant adverse effects and may necessitate dose adjustment, temporary withholding, discontinuation, switching of medication, or adjunctive treatment. They link this to strong dopamine D2 receptor (搜索) antagonism, associated with greater risk of extrapyramidal symptoms such as parkinsonism, akathisia (搜索), dystonia and tardive dyskinesia (搜索).
Preventability was evaluated using the Schumock and Thornton scale. A total of 130 ADRs (50.8%) were classified as definitely preventable, 89 (34.8%) as probably preventable and 37 (14.4%) as non-preventable. The authors state the predominance of definitely preventable reactions underscores gaps in prescribing practices, monitoring and follow-up care, and highlights the critical role of a multidisciplinary healthcare team in minimizing medication-related harm. The proportion of preventable ADRs was lower than the 97% reported in an Indian study where use of atypical antipsychotics and limited polypharmacy may have facilitated clearer attribution and prevention, but higher than proportions ranging from 21.6% to 45.9% in other Indian studies.
Three Factors Linked to Elevated Risk
Bivariate and multivariate logistic regression identified three independent factors associated with ADRs. Patients who had separated from their spouses were twice as likely to experience ADRs, which the authors attribute to inadequate social support, potentially resulting in poor adherence, worsening psychiatric symptoms, relapse and the need for higher antipsychotic doses. Poor adherence may also lead to incorrect medication use, suboptimal self-monitoring and underreporting of ADRs.
Patients diagnosed with bipolar disorder (搜索) were also twice as likely to develop ADRs following antipsychotic treatment. The authors note similar findings among 441 children and adolescents in Turkey, where those with bipolar disorders had a higher likelihood of experiencing ADRs while on antipsychotics. This association may reflect more aggressive antipsychotic dosing strategies, as bipolar patients often present with acute psychotic episodes requiring higher doses.
Co-administration of antipsychotics via both oral and intramuscular routes was associated with nearly double the odds of experiencing ADRs. The authors explain that additive pharmacologic effects can lead to higher plasma drug concentrations, and that intramuscular formulations typically have a long half-life, prolonging drug exposure in the blood. Insufficient follow-up and monitoring of these patients further amplify this risk.
Methodological Approach and Limitations
The study enrolled patients aged 18 years and above diagnosed with a psychotic disorder and taking antipsychotics for at least the past month, excluding those with active psychotic symptoms. Sample size was calculated using the Kish-Leslie formula, based on a 43.5% expected ADR prevalence from a prior Indian study, yielding a target of 415 participants after adding a 10% non-response rate. Consecutive sampling was used over three months.
Fifteen ADR types were assessed using standardized operational definitions derived from established pharmacovigilance guidelines and clinical assessment scales, including weight gain, akathisia (搜索), dry mouth, tremors, sedation, tardive dyskinesia (搜索), hyperglycemia (搜索), nausea and vomiting, increased appetite, slurred speech, constipation, skin rash, blurred vision, drowsiness and dystonia. Each ADR was confirmed by both the principal investigator and a resident psychiatrist before classification. Weight gain was defined as an increase of 7% or more from baseline body weight, and hyperglycemia as random blood sugar above 180 mg/dL or fasting blood sugar above 100 mg/dL occurring after antipsychotic initiation without prior documented hyperglycemia.
The authors acknowledge several limitations. The cross-sectional design limits the ability to establish causality between antipsychotic use and ADR occurrence, and future research should adopt longitudinal designs. As a single-center study, findings may have limited generalizability. The sample size calculation relied on a prevalence estimate from an Indian study because comparable local data were unavailable. Assessment of baseline weight and glycemic status relied on documented medical records and patient recall, potentially introducing recall bias, and laboratory testing for hyperglycemia (搜索) was not universally available due to resource constraints, potentially leading to under-detection of metabolic ADRs. The authors also note that assessment of subjective ADRs such as sedation, drowsiness and dry mouth relies on patient self-report, which may have introduced reporting bias, though clinical examination and investigator assessment were used to corroborate reports where possible.
Implications for Pharmacovigilance
The authors conclude that the findings highlight the need for active and coordinated pharmacovigilance systems, routine ADR screening and enhanced training of mental health professionals to improve early detection, documentation and management of antipsychotic-related ADRs. Across sub-Saharan Africa, where mental health services are often constrained by limited resources and specialist workforce shortages, integrating ADR monitoring into existing mental health programs could improve treatment safety, adherence and clinical outcomes.
They call for risk-based approaches to antipsychotic prescribing and monitoring, with particular attention to patients who may lack adequate social support, those diagnosed with bipolar disorder (搜索) and those receiving multiple antipsychotic formulations. Recommended practical measures include careful drug selection, individualized dosing, routine monitoring, structured follow-up, timely recognition and management of early symptoms, optimized adherence and comprehensive patient counseling. Mental health programs should incorporate routine assessment of social determinants of health, strengthen psychosocial support services and enhance patient and caregiver education.
The study was conducted in compliance with the Declaration of Helsinki, with ethical approval from the Mbarara University Research Ethics Committee (Reference No: MUST-2024-1960) and administrative approval from the director of Mbarara Regional Referral Hospital (搜索). Data were analyzed using STATA Version 17.