Little is known about the intended effects of differential drug cost sharing (DCS), an increasingly common strategy to discourage """"""""unnecessary"""""""" use of newer, expensive medications in most HMOs and government insurance programs. The proposed cohort study in British Columbia, using time series and comparison series, will assess the impact of DCS for blood pressure lower drugs, specifically angiotensin converting enzyme inhibitors (ACEI) and calcium channel blockers (CCB), on drug utilization and drug switching in elderly recipients of these drugs (N approximately 160,000). To examine discontinuities in utilization rates, segmented linear regression will be used to estimate changes in level or slope of these rates. Specific research questions are: 1) To what extent do monthly/quarterly rates of utilization and expenditures for blood pressure lowering drugs change over a period of 23 months before the introduction of DCS, a 2 month transitional period during implementation, and 11 months following the policy change standardized by age and gender? 2) Among patients receiving ACEI or CCB, is there a net reduction in overall use of anti-hypertensive agents? (This may be an indicator of underuse of need anti-hypertensive treatment due to restricted access.) 3) To what extent do patients receiving ACE inhibitors or calcium channel blockers switch from cost-sharing medications to non-cost drugs within these classes after the policy change? 4) To what extent do patients receiving ACE inhibitors or calcium channel blockers replace thee medications with other drugs unaffected by the DCS policy? 5) What patient and/or physician characteristics (e.g. age, gender,, income, co-morbidity, year of graduation from medical school) predict whether individual patients switch from cost-sharing to no-cost medications? It is hypothesized that low-income patients will be significantly more likely to switch to no cost-sharing drugs, and might therefore be limited in their choices. 6) What patient and/or physician characteristics (e.g. age, gender, income, comorbidity, year of graduation from medical school) predict whether individual receive prior authorization exemptions? It is hypothesized that low income patients and patients with a higher degree of comorbidity will be significantly more likely to receive prior authorization exemption. This study will fill important gas in the understanding of the immediate effects of DCS, especially in the elderly, and will be essential for the development of evidence/based policies in pharmaceutical benefits programs of public and private health plans.
| Schneeweiss, Sebastian; Maclure, Malcolm; Soumerai, Stephen B et al. (2002) Quasi-experimental longitudinal designs to evaluate drug benefit policy changes with low policy compliance. J Clin Epidemiol 55:833-41 |
