Application Evaluation of Prescription Comment Software in Our Hospital / 中国药房

ABSTRACT

OBJECTIVE:To evaluate the application of prescription comment software in our hospital. METHODS:552 pre-scriptions were evaluated by pharmacist A,B and software,and the Kappa values were calculated to evaluate the reliability of the software. The results of comment by pharmacist B and by software were combined (serial test and parallel test),and then com-pared with the results of comment by pharmacist A. The sensitivity,specificity and Youden's index were calculated to evaluate the reliability of the software. The comment results of 552 prescriptions by pharmacist A for twice were compared to calculate Kappa value,and it was used to evaluate the comment consistency of pharmacist A. RESULTS:The pharmacist A,B comment results and software comment results showed poor consistency,and Kappa value were 0.020 2 and 0.002 0,with significant difference (P<0.001);pharmacist A comment results had strongest consistency with pharmacist B comment results,and Kappa value was 0.843 0,with no significant difference(0.1software sensitivity was 69.09%,and the specificity was 32.58%. The sensitivity of pharmacist B was 84.55%,and the specificity was 97.74%. The parallel test had high sensitivity (93.64%),no change was found in specificity (32.35%). The sensitivity of serial test decreased a little (54.55%) and the specificity improved significantly (97.96%);but there was significant difference,compared with pharmacist A comment (P<0.005). The preliminary evaluation and re-evaluation of pharmacist A were most consistent,and Kappa value was 0.971 4,with no significant difference (0.5CONCLUSIONS:Compared with the pharmacist artificial prescription comment in our hospital,software com-ment has low sensitivity and specificity,and poor authenticity overall. Pharmacist combined with software for prescription comment can not improve the sensitivity and specificity of results.