POCKET-COST-HF: Comprehensive Cost Disclosure improves cost-informed decision making

By Leah Kosyakovsky on

Key Points:

  • Out-of-pocket costs for GDMT in HFrEF have risen substantially with the advent of novel effective therapies. However, there is limited information regarding the utility of comprehensive cost disclosure in informing patient and clinician decision-making regarding prescribing GDMT.
  • The POCKET-COST-HF study was a stepped-wedge cluster randomized trial examining the utility of a tailored, comprehensive cost disclosure intervention on a primary endpoint of cost-informed decision-making, ascertained by transcription of audio recordings of a clinic visit for HF.
  • Comprehensive cost disclosure resulted in a higher proportion of encounters in which cost of medication was discussed, with further studies needed to inform the potential impact on medication prescribing and implementation strategies. 

While there have been great strides in medical therapy for HFrEF in recent years, each new effective therapy has led to increased expense. Average out-of-pocket (OOP) costs for those with Part D coverage is now estimated at >$2600 for four-drug GDMT, and both patients and clinicians report discomfort with integrating discussions about cost into clinical decision-making. In a breaking presentation at the 2023 AHA Scientific Sessions today, Dr. Neal Dickert and his team presented their study: “Integrating Cost Into Shared Decision-Making for Heart Failure With Reduced Ejection Fraction: A Trial Providing Out-of-Pocket Costs for Heart Failure Medications During Clinical Encounters,” or the POCKET-COST-HF clinical trial.

The POCKET-COST-HF trial (NCT04793880) was a stepped-wedge cluster randomized trial conducted across 6 clinic sites and 2 health systems. Cluster size was fixed at 40 patients, with a transition to intervention after 10, 20, and 30 patients. The study intervention was built on the pre-existing EPIC-HF checklist; the control group only received the checklist, and the intervention group also received the OOP cost for medications with higher cost (estimated by the TailorMed financial navigation company). Every visit was audio recorded; each patient was surveyed 2-3 weeks after the visit by telephone and also underwent an EHR chart review 3 months after the visit.

Participants were required to have HFrEF with an LVEF ≤40%.  Key exclusion criteria were advanced HF therapies, GFR <30, presence of amyloid cardiomyopathy, and life expectancy < 1 year. The primary outcome was “cost-informed decision-making,” defined as a clinician or patient mentioning costs of GDMT during the visit assessed by transcript review of the audio recording using a defined codebook. A generalized linear mixed model (GLMM) was used for the primary model, with random effects for site and adjustments for quarter, financial status, age, sex, and payor status. Key secondary outcomes included specific components of cost conversations (ie, specific cost of HF medication, qualitative cost of HF medication, discussion about value of medication, etc), as well as post-visit survey measures including medication changes and recall/perceptions of both the clinical encounter and the use of the checklist.

A total of 247 patients were randomized and had a successfully recorded encounter (121 in the intervention arm, 126 in the control arm). The median age was 63, 70% were men, and the two groups had similar income distributions.  Twenty five clinicians were involved in the visits in each arm, respectively.  In the primary analysis, there was a significant (but modest) increase in cost-informed decision-making in the intervention arm (68% vs 49%, p=0.021). There was a positive impact in 5 of the 6 included sites, with varying rates of cost discussions at baseline. In secondary analyses, the only component of cost conversations that differed significantly between the two groups was contingency planning considering the cost of medication (31.9% vs 16.5% in the control vs intervention, p=0.02). Fewer patients in the intervention arm started a new medication after the clinic visit than in the control group (21.2% vs 36.4%, p=0.01).

When discussing the clinical implications of the study at the Scientific Sessions, Dr. Dickert stated: “Tailored, comprehensive cost disclosure has a modest but significant impact on the proportion of encounters at which cost was discussed for patients with HFrEF…larger sample sizes and longer-term follow-up are needed to assess the impact of comprehensive OOP cost disclosure on medication choices, prescribing, and adherence.”