Premium
Physician characteristics associated with prescription of inappropriate medications using Beers criteria
Author(s) -
Imai Hirohisa,
Fick Donna M.,
Waller Jennifer L.,
Maclean J. Ross
Publication year - 2007
Publication title -
geriatrics and gerontology international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.823
H-Index - 57
eISSN - 1447-0594
pISSN - 1444-1586
DOI - 10.1111/j.1447-0594.2007.00428.x
Subject(s) - medicine , medical prescription , beers criteria , family medicine , logistic regression , pharmacy , confidence interval , managed care , incidence (geometry) , medline , emergency medicine , health care , nursing , physics , political science , optics , economics , economic growth , law
Background: The prescription of potentially inappropriate medications (PIM) for elderly patients represents a major problem. In the published work, various practice characteristics associated with physicians prescribing habits have been reported. However, existing data has shed little light on the characteristics of physicians who tend to prescribe PIM. We examined whether personal, professional or practice characteristics differ between physicians who prescribe PIM and those who do not. Methods: The subjects comprised primary care and general practice physicians. Physicians were identified from the pharmacy database of a managed care organization as having prescribed medications for Medicare patients over 65 years enrolled in a managed care plan. We adopted Beers criteria to describe the prevalence of PIM use. The physicians were divided into three groups according to number of PIM prescribed. To examine the extent of associations between all the physician‐related characteristics studied, polychotomous logistic regression was conducted. Results: Physicians who prescribed one to five PIMs were 0.63 (95% confidence interval [CI], 0.41–0.98) times more likely to have publications than physicians who prescribed no PIM. Physicians who prescribed more than six PIM were 3.18 (95% CI, 2.05–4.95) times more likely to be certified by an internal medicine board, 0.48 (95% CI, 0.30–0.78) times more likely to have publications, and 1.84 (95% CI, 1.01–3.35) times more likely to be in solo practice than physicians who prescribed no PIM. Conclusion: In this study, we found three predictors of PIM prescribing incidence. Since the current study could only describe associations and not causality, further research is necessary.