Reliability of a modified medication appropriateness index in ambulatory older persons

OBJECTIVE: To evaluate the reliability of a medication appropriateness index (MAI) modified for elderly outpatients in a non-Veterans Affairs setting. DESIGN: Reliability study. SETTING: General community, PARTICIPANTS: Ten community-dwelling elderly (> 65 y) taking live or more regularly scheduled medications and participating in a university-based health service intervention study. MAIN OUTCOME MEASURES: Interrater reliability of MAI ratings of 65 medications made by two clinical pharmacists for individual items and for an overall summed score was calculated by use of kappa statistics and intraclass correlation coefficient. RESULTS: The interrater agreement for each of the individual MAI items was high for both appropriate and inappropriate ratings and ranged from 80% to 100% (overall kappa = 0.64). Overall agreement for the summed score was good (intraclass correlation = 0.80). CONCLUSIONS: The modified MAI is a reliable instrument for evaluation of medication appropriateness in a non-Veterans Affairs, ambulatory, elderly population and may provide pharmacists with a practical and standard method to evaluate patients' drug regimens and identify some potential drug-related problems.