Prospective comparison of START-STOPP and MAI for the detection of inappropriate prescriptions in an acute geriatric ward

Introduction.- Inappropriate prescription (IP) and adverse drug events (ADEs) are crucial in older persons that account for 40% of all medication use. About 40% of geriatric patients suffer from ADEs, responsible for 10 to 30% of admissions in geriatric wards. Thirty to 60% of ADEs are preventable. To prevent ADEs inappropriate prescription must be detected: either by implicit evaluation based on physicians' knowledge as in the Medication Appropriateness Index (MAI), or by explicit lists of IP as the ''Screening Tool of older Persons' Prescriptions''-''Screening Tool to Alert doctors to the Right Treatment'' (STOPP-START). Our study aimed to compare STOPP-START to MAI for detection of IP and of drugs related to ADEs.
Method.- Prospective evaluation of consecutive admissions to a 24-bed acute geriatric ward in a general teaching hospital during 3 months. Blinded evaluation of admission and discharge treatments with STOPP-START versus 2 senior geriatricians based on MAI.
Results.- Eighty-seven patients were included. Mean age was 82.8 years and 63% were female. MAI and STOPP detected overpre- scription in 68 and 74% (n.s.) of patients at admission and in 47 and 76% (P <0.05) at discharge. MAI and START detected under-prescription in 47 and 63% (P <0.05) and in 44 and 62% (P <0.05), respectively. At admission the 2 scales were in agreement with a K of 0.36. The most important risk factor for overprescription was the number of drugs prescribed, for underprescription the number of former active pathologies. ADEs were documented in 22% of our patients, but the responsible drugs were not detected by either score in 36.8%. START-STOPP criteria detected 42.1% and MAI 36.9% of ADEs.
Conclusions.- START-STOPP criteria detect more IP than MAI. Screening at admission does not reduce IP at discharge and screening of discharge medications is advised. ADEs are not suffciently detected by either procedure.