Comparison of tools for the assessment of inappropriate prescribing in hospitalized older people


Professor James C. McElnay, Clinical and Practice Research Group, School of Pharmacy, Queen's University, Belfast, 97 Lisburn Road, Belfast BT9 7BL, UK, E-mail:


Rational, aims and objective  The aim of the present study was to compare the ease of use and the capability of four approaches [Medication Appropriateness Index (MAI), the Beers' criteria 2003, the Improved Prescribing in the Elderly Tool (IPET) and Health Plan Employer Data and Information Set (HEDIS)] in assessing changes in medication appropriateness in elderly patients over a period of hospitalization.

Methods  A retrospective observational study in two hospitals in Northern Ireland using the four measures was undertaken, involving a cohort of 192 patients (aged > 65 years). Medication appropriateness assessments were made at three stages during the patients' hospital ‘journey’, that is, at admission, during their inpatient stay and at discharge. The identifying rates of inappropriate prescribing in elderly patients in hospital used validated screening tools: MAI, the Beers' criteria 2003, the IPET and HEDIS.

Results  The MAI was the most comprehensive approach but was also the most time consuming to apply. Data derived using the MAI indicated clearly that there was improved medication appropriateness over the three hospital stay stages (P < 0.001). Although this trend was also significant for the Beers' criteria 2003 (P < 0.05) and the IPET (P < 0.05) approaches, the HEDIS was unable to differentiate changes in appropriateness over time. There was a good correlation between data derived using the MAI and the Beers' criteria 2003 and the IPET approaches; this correlation was not evident for the HEDIS.

Conclusions  The MAI is the most convincing tool in evaluating medication appropriateness, but is very time consuming to apply. Beers' criteria 2003 and the IPET perform to an acceptable standard within the clinical setting and are more practical in their application. The HEDIS, although simplest to apply, does not have the sensitivity to measure change in appropriateness over time.