The burden of serious adverse drug reactions in South Africa

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Background The burden of serious adverse drug reactions (ADRs) in South Africa (SA) is not well characterised, and may be influenced by the high HIV prevalence. This thesis aims to describe this burden, by reporting findings of three hospital-based surveys of (1) ADRs causing death among adult medical inpatients, (2) ADRs causing admission of adults to medical wards, and (3) serious ADRs causing admission, prolongation of existing admission, or death among paediatric medical inpatients. Survey findings are contextualised in a systematic review of studies characterising serious ADR burden in sub-Saharan Africa (SSA). Methods In each of the three surveys, folder review with the assistance of trigger tools was used to identify potential serious ADRs. Multidisciplinary teams assessed causality, preventability, and seriousness of these events using published criteria. The World Health Organization-Uppsala Monitoring Centre system for standardised case causality assessment was used. This choice was informed by a study using a sample of cases from one of our adult surveys, which demonstrated its higher interrater agreement compared with the Liverpool ADR Causality Assessment Tool. Multivariable logistic regression was used to explore associations between demographic and clinical factors and the occurrence of serious ADRs. Results Among adult medical inpatients 56 of 357 deaths (16%) and 164 of 1951 admissions (8.4%) were ADR-related, and 43% and 45% respectively of these serious ADRs were preventable. Drugs used in the management of HIV and tuberculosis were frequently implicated. Among paediatric inpatients 3.8 serious ADRs occurred per 100 drug-exposed admissions, of which 23% were preventable and 20% fatal or near-fatal. A broad range of drugs was implicated in these serious paediatric ADRs. HIV positive status was independently associated with an increased risk of ADRs in every survey. Fifteen studies contributed to the systematic review. The median proportion of admissions attributed to ADRs was 6.4% [IQR 4.0% to 8.4%] among nine active surveillance studies in adults in SSA, with antiretroviral and antituberculosis drugs often implicated in serious ADRs. Conclusions Fatal, near-fatal, and hospitalising ADRs occurred frequently in South African hospitals. Many of these serious ADRs, particularly in adults, were preventable. Serious ADRs should be addressed in patient and health worker education, and safety monitoring systems improved. In SA and SSA the high HIV prevalence appears to be a driver of serious ADRs.