PREDICTIVE FACTORS FOR OUTPATIENT PRESCRIBING ERRORS AT PHRACHOMKLAO PHETCHABURI HOSPITAL

Abstract

This case-control retrospective research investigated prescribing errors, the relationship between predictive factors and errors, and establishing a prescribing-error predicting equation in the out-patient department at Phrachomklao Phetchaburi Hospital. The population studied 91,699 prescriptions from 1 October 2021 – 31 March 2022. The data was stratified and randomly selected from the electronic hospital information system, which comprised 188 and 376 prescriptions, respectively, from the study and control groups. Statistical analyses on univariate and multivariate binary logistic regression at a significance level of 0.05 was conducted. The findings suggested that the factors related to prescribing errors in out-patients were patients’ estimated glomerular filtration rate (eGFR), international normalized ratio (INR), the number of drugs prescribed per prescription, prescriptions with high alert drugs, and prescriptions with cardiovascular drugs. Factors which can predict the occurrence of prescribing errors were eGFR of < 60 ml/min/1.73 m² (OR_adj = 2.625, 95% CI = 1.24 – 5.55), abnormal INR level (OR_adj = 7.978, 95% CI = 2.69 – 23.67), and prescriptions with more than 7 drug items (OR_adj = 1.960, 95% CI = 1.23 – 3.13). The probability of prescribing error occurrence (p) complies to the following equation: p = 1/[1+e^-log(odds)], where log (odds) = -1.013 + 0.965 (eGFR < 60) + 2.077 (abnormal INR level) + 0.673 (prescriptions with more than 7 items). Therefore, pharmacists responsible for analyzing prescriptions should be particularly careful when the predictive factors are present in order to minimize prescribing errors, prevent harm, and increase patients’ safety.