Patient characteristics including age, comorbidity and prognosis affect prescribing decisions and may be associated with under-use of many valuable drug therapies, particularly in frail older persons. The investigators propose epidemiologic studies that will: a) identify demographic and health-related factors associated with the decision to initiate specific drug regimens and to maintain adherence to previously prescribed therapies; and b) clarify the relationships between use of specific drugs and mortality in the elderly. Studies will focus on specific drug classes that are commonly under-utilized in spite of their demonstrated therapeutic value including: lipid lowering agents, beta blockers, angiotensin converting enzyme inhibitors, hypoglycemic agents, and glaucoma drugs. Additional methodologic aims are c) to construct a simple, drug-based comorbidity index that predicts mortality in older persons and d) to evaluate the extent to which treatment decisions are clarified by inclusion of information from HCFA's Minimum Data Set. The proposed studies will utilize data from two large populations of persons aged 65 years or older: 1) over 400,000 residents of New Jersey who had prescription drug coverage through either Medicaid or that state's Pharmacy Assistance for the Aged or Disabled program during the years 1990-2000; and 2) all residents of British Columbia, including over 500,000 persons during the years 1995-2000, who received prescription benefits through that province's universal program of drug benefits for the elderly. Data sources for these populations include: prescription drug claims from these benefit programs; data on hospitalizations including discharge diagnoses, information on outpatient visits and diagnoses, and on mortality available from parallel databases; and detailed information on patient characteristics available for New Jersey residents in nursing homes available from HCFA's Minimum Data Set. The study populations have high death rates and contain large numbers of subgroups at particular risk for under-treatment including those aged 85 and above, nursing home residents, women and blacks. Clarification of the epidemiology of drug use, based on current longitudinal data, can lead to improved prescribing practices and thereby decrease morbidity and mortality in high-risk populations.
Showing the most recent 10 out of 37 publications
