Both within and outside the VA, inappropriate use of medications is common, affecting approximately 20-25% of Veterans aged 65 years or older. As such, VHA has shown great interest in the American Board of Internal Medicine's Choosing Wisely campaign, which asks medical societies to identify five commonly used tests or treatments in their specialty that provide little benefit to patients, may cause harm, and are costly. One of the Choosing Wisely recommendations advised against inappropriate use of vasodilators such as phosphodiesterase-5-inhibitors (PDE5I) for certain types of pulmonary hypertension (PH). These medications are costly and can be harmful when used inappropriately. In our pilot work we found that over 4000 Veterans were treated with PDE5I for PH from FY 2005-2012, with 100-fold variation in frequency of inappropriate use across sites. In order to reduce inappropriate prescribing, we must first understand the factors driving decision-making surrounding prescribing of PDE5I for PH, as well as the outcomes associated with its use in VA. This work will identify lessons on how to reduce inappropriate prescribing that can be applied to de-implement use of other high cost medications with little benefit and potential for harm. We propose a mixed methods study with the following 3 specific aims: 1: Identify patient-, clinician-, and site-level determinants of inappropriate use of PDE5I for PH 2: Evaluate processes of prescribing and approval of PDE5I for PH in local contexts and identify best practices 3: Assess the outcomes of inappropriate use of PDE5I for PH. The study aims include:
Aim 1 : Using national VA data and linked Medicare claims, we will build mixed effects regression models to assess factors associated with inappropriate prescribing of PDE5I for PH, comparing patients prescribed PDE5I with similar patients who did not receive PDE5I.
Aim 2 : Through protocol review and in- depth qualitative interviews with key informants (patients, clinicians, pharmacists, VA leadership), we will conduct qualitative evaluations of patient-provider and inter-professional processes that result in decisions about use of PDE5I for PH at 6 VA sites (3 with high utilization, 3 with low utilization).
Aim 3 : We will form cohorts of PH patients who did or did not receive PDE5I inappropriately and identify outcomes using national abstracted data. Risks for important outcomes, including mortality, hospitalizations, and adverse effects of PDE5I, will be compared in the two cohorts. We will perform propensity score matching to adjust for potential confounding. Promoting safe and effective prescribing is a primary goal of the VHA, and of our operational partners in the Pharmacy Benefits Management office and the National Program Office for Pulmonary Medicine. This proposal will lay the groundwork to inform a future proposal to develop and implement targeted strategies to reduce inappropriate prescribing of PDE5I for PH. Moreover, this work will serve as a generalizable test case to inform how several factors influence prescribing: 1) tensions between guidelines, VA policies, and individual patient care, 2) how non-VA prescriptions for dual care patients are handled, 3) role of pharmacists in gate-keeping and point-of-care decision-making, 4) local implementation of national policies. This work is firmly in line with the call to de-implement ineffective care outlined in the Blueprint for Excellence (Strategy 2).
Inappropriate care, such as prescribing ineffective medications, costs the VA billions each year and exposes Veterans to iatrogenic harm. The VA is interested in ways to de-implement ineffective care. However, interventions to selectively de-implement care are more successful when built on a sound understanding of the complex factors that drive inappropriate use, the barriers encountered and best practices used to minimize inappropriate use, and the harms of inappropriate use. We propose a mixed methods study to examine factors and outcomes associated with inappropriate prescribing of phosphodiesterase-5-inhibitors for pulmonary hypertension. These medicines are expensive, offer little benefit, and are potentially harmful, yet are used inappropriately across VA sites with use quadrupling from FY05-12. This test case will produce generalizable knowledge relevant to de-implementing prescribing of other high-cost, risky medications with little benefit.
