Health provider factors associated with reporting of adverse drug reactions in Kenyatta national hospital

Abstract

Background: Medicines are an important component of public health because they are used to prevent and treat disease; however the use of medicines may cause harm to the patient. An adverse drug reaction is harm that arises from the use of a drug. Adverse drug reactions have a huge impact on the health system since they result in drug-related morbidity and mortality as well as indirect costs such as loss of productivity at work. Reporting of adverse drug reactions to a relevant authority is one of the methods of enhancing medication safety; however underreporting of adverse drug reactions by health workers is a major challenge in enhancing medicines safety. The factors that affect reporting of adverse drug reactions include health worker factors such as age, specialty, location of practice, knowledge and attitudes about adverse drug reaction reporting; health systems factors such as availability of reporting tools, training and workload; patient factors such as the knowledge of adverse reactions and willingness to discuss them with the health worker and the severity of the adverse reaction. Objectives: The broad objective of the study was to determine the health provider factors that are associated with reporting of adverse drug reactions in Kenyatta National Hospital. The specific objectives were to determine the socio-demographic factors of the health workers in Kenyatta National Hospital and to determine the knowledge and attitudes of health workers in Kenyatta National Hospital on adverse drug reaction reporting. Methodology: The study used both quantitative and qualitative techniques. The quantitative component of the study involved the use of a cross-sectional study design. The study population was health workers in Kenyatta National Hospital who were in a position to detect adverse drug reactions. The study population was clustered by cadre then systematic sampling was used to select the participants. A closed-ended questionnaire was administered to the study participants who consented to participate in the study. The qualitative component of the study involved the use of key informant interviews using open-ended questions. Key informants were selected from the Pharmacovigilance Department of the Pharmacy and Poisons Board. Analysis: Data from the questionnaire was coded and entered into Stata version 12. Descriptive statistics were used to summarize the socio-demographic characteristics of the study participants. Categorical predictor variables were summarized using frequencies and percentages while the continuous predictor variables were summarized using means and medians. Contingency table analysis using Fischer’s exact test was used to determine if there was a statistically significant association between reporting an adverse drug reaction and the categorical predictor variables which are sex, department of the health worker and previous pharmacovigilance training. Student’s t test was used to determine if there was a statistically significant association between reporting an adverse drug reaction and the continuous predictor variables age, duration of practice and the knowledge score. Wilcoxon rank sum test was used to determine if there was a statistically significant association between reporting an adverse drug reaction and the overall attitude score. Logistic regression analysis was carried out to determine the effects of all the predictor variables, taken together, on reporting of adverse drug reactions. Analysis of the qualitative data was done by sorting the data into categories and examining the emerging themes. Results: Previous pharmacovigilance training was found to be significantly associated with reporting of adverse drug reactions (p=0.000). Health workers were more likely to report adverse drug reactions if they had been trained. Health workers’ knowledge on adverse drug reactions was significantly associated with reporting of adverse drug reactions (p=0.0021) with reporters having higher mean knowledge scores than the non-reporters. Previous pharmacovigilance training (p=0.000, Odds Ratio 14.04, 95% CI: 3.19-61.76) and knowledge (p=0.033, Odds Ratio 1.19, 95% CI: 1.01-1.40) were found to be the strongest predictors of reporting adverse drug reactions when logistic regression was carried out. The key informants identified several health provider and health systems factors that affect reporting of adverse drug reactions. Lack of knowledge about the adverse drug reaction reporting scheme and poor attitudes were identified as health worker factors that hindered adverse drug reaction reporting. Health systems factors that hindered the reporting of adverse drug reactions were the unavailability of reporting tools, high workloads and the costs incurred when sending a hard copy report to the Pharmacy and Poisons Board. Conclusion: Health workers’ knowledge on adverse drug reaction reporting was a major determinant of reporting of adverse drug reactions. Health workers who had been trained and had more knowledge on adverse drug reaction reporting were more likely to report adverse drug reactions than those who had not been trained nor had less knowledge about adverse drug reaction reporting. Recommendations: In order to increase reporting of adverse drug reactions by health workers, there is need to train all health workers in order to increase their knowledge, to include all cadres of health workers in reporting and to avail the adverse drug reaction reporting forms in all the wards and clinics of the hospital.