Commonly prescribed medicines, core prescribing indicators, and prescription errors in a tertiary healthcare setting in Sri Lanka: a descriptive cross-sectional study

Abstract

Abstract Objectives To describe the commonly prescribed medicines, WHO/INRUD core prescribing indicators, and prescription errors from a tertiary healthcare setting in Sri Lanka. Methods This study was conducted at clinics and wards in a group of tertiary hospitals from Colombo, Sri Lanka. A total of 1600 prescriptions (800 in-ward [IW] and 800 out-patient clinic [OP]) were analysed and evaluated against WHO/INRUD core prescribing indicators. Prescription errors and medication interactions were also analysed. Key findings Number of medicines per encounter on average was 4.2 ± 2.9 (median: 4.0; range: 1–18) (IW vs. OP: 5.0 ± 3.1 vs. 3.4 ± 2.1; P < 0.001). Overall, the percentage of prescriptions with an antibiotic was 25.9% (n = 415) (IW vs. OP: 45.0% vs. 6.9%, P < 0.001). Percentage of encounters with an injection was 20.4% (n = 326) (IW vs. OP: 36.9% vs. 3.9%, P < 0.001) and percentage of medicine prescribed by generic name was 90.7% (IW 92.9% vs. OP 87.3%, P < 0.001). Prescribing from the essential medicines list (EML) was 76.9% (IW 77.7% vs. OP 75.7%, P-NS). Among IW prescriptions details regarding allergies were not mentioned in 643 (80.4%), while in paediatric prescriptions weight was not specified in 242 (75.6%). At least a single abbreviation was used in 638 prescriptions (39.9%). One or more medication interactions were identified in 744 (46.5%) prescriptions (major: 86 [5.4%]). Conclusions We identified a relatively high number of average medicines per encounter, indicating a degree of polypharmacy. In addition, generic prescribing and usage of EML medicines are areas recommended for further improvement. Prescription errors were common, and several commission/omission errors had potential for preventable patient harm.