Factors associated with nonprescription of evolving modes of therapy--a model for drug use evaluation in hospitalized medical patients.

OBJECTIVE: To measure the rate of inappropriate underuse of drug therapy among hospitalized patients and to identify factors associated with nonprescription of two effective modes of therapy. DESIGN: A survey was conducted by retrospective chart extraction to measure the rates of nonprescription of two therapeutic modalities--aspirin and angiotensin converting enzyme inhibitors. Stepwise logistic regression was used to determine which factors were significantly associated with nonprescription of either drug. SETTING: Two internal medicine departments in a major community and referral hospital in Israel. PATIENTS: Those who were hospitalized through 1990 with a diagnosis on their discharge sheet of angina pectoris or congestive heart failure. A random sample selection was made to define the study population. RESULTS: The rates of nonprescription among patients in whom there are no drug contraindications were 54% (95% CI 47-62) for aspirin and 34% (95% CI 27-41) for angiotensin converting enzyme inhibitors. Three variables correlated independently with nonprescription of both drugs: hospitalization in ward B, nonuse of the drug prior to hospitalization, and secondary priority of the diagnosis angina pectoris or congestive heart failure on discharge sheet. CONCLUSIONS: Nonprescription of essential drugs is a common phenomenon that varies significantly among physicians. Inconsistency in application of knowledge to clinical practice, as well as a diversity of approaches toward interpretations of clinical study results, significantly influences the prescription rates of evolving modes of therapy.