OBJECTIVES

The objectives for this research are to: 1) understand how payers define unmet need and determine standards of care; 2) explore characteristics and acceptability of clinical and humanistic endpoints from RCTs and RWE; 3) understand methodological preferences for value evidence development in clinical trials and the real world.

METHODS

Senior-level decision-makers (n=10) from US payer organizations (n=9) representing 206.1 million member lives were recruited in May 2020 to participate in a web survey (10 topics) and qualitative telephone interview (approximately 25 minutes). Advisers were required to have a high level of knowledge surrounding formulary decision-making, clinical pathways and medical policy development, and manufacturer contracting. Descriptive statistics (e.g. Surveymonkey ranking scores, weighted means, % of mentions) and contextual analysis were used to analyze the results. Subanalyses were conducted by payer archetype.

RESULTS

When defining unmet need, payers most frequently look for a high level of HCRU (weighted mean 4.67), poor survival prognosis (4.4), and drug related issues (i.e. AE management, loss of response) (3.9). In determining appropriate standard of care and eligible patient population, payers most frequently reference clinical guidelines (100%) and KOL feedback (90%). Cost is the most important factor in determining standard of care (2.5). Direct endpoints (4.7) are the most acceptable; PROs (3.2) and novel/emerging endpoints (3.3) are the least acceptable. Most payers (70%) expect an active comparator in an RCT. Head to head RCTs (4.9) are the most acceptable form of treatment comparison, followed by RWE comparative effectiveness research (4) and network meta-analysis (3.2). Quality of PRO evidence (2.4) is the most important characteristic in evaluating humanistic outcomes. Electronic health record studies (4.3) are the most acceptable form of RWE.

CONCLUSIONS

Evaluation of clinical efficacy and safety evidence remain the gold standard for reimbursement decision making, while PROs continue to play a very minor role in the evaluation paradigm.