Concerns with Patient Reported Outcome Measurement and Value Claims for Therapy Response: The Case of Mavacamten and Symptomatic Hypertrophic Cardiomyopathy (SHCM)

Abstract

Fundamental measurement is the basis for a rational assessment of patient reported outcome (PRO) value claims; both as response to therapy and the submission of credible and evaluable value claims to formulary committees and other health system decision makers. It is important to emphasize the importance of creating interval and ratio scales as opposed to nominal and ordinal scales to support value claims; a recognition that follows from acceptance of conjoint simultaneous measurement and the contribution of Rasch or modern measurement theory (RMT). Failure to appreciate the role of RMT has led thousands of researchers simply to apply numerals to events, inappropriately applying the techniques of classical statistical analysis, with the result that all that is produced are ordinal PRO scores. Instead, we should be aiming for interval and ratio scores based on a comprehensible latent trait and the application of the Rasch model. The purpose of this brief commentary is to review the measurement properties of PRO value claims for mavacamten (Camzyos; Bristol Myers Squibb) in symptomatic hypertrophic cardiomyopathy (SHCM) and to judge whether they have any validity when judged against the requirements of modern measurement theory. The assessment includes both the recent evidence report by the Institute for Clinical and Economic Review (ICER) for mavacamten as well as pivotal randomized trial (RCT) value claims that combine clinical endpoints with PROs that fail the standards of fundamental evidence. These include the Kansas City Cardiomyopathy Questionnaire (KCCQ), the New York Heart Association (NYHA) functional classification and the EuroQuol EQ-5D-5L multiattribute health related quality of life (HRQoL) preference instrument. The review concludes that apart from purely clinical claims based on the various pivotal trials, there are no PRO claims for mavacamten in SHCM that meet the required measurement standards.

Post-Menopausal Quality of Life Claims: Overlooking the Requirements of Normal Science and Fundamental Measurement in ICER’S Cost-Effectiveness Assessment of Fezolinetant for Moderate to Severe Vasomotor Symptoms

“…Absent Rasch measurement we are left with observations or counts which are always ordinal; we must recognize that meaningful measurement is based on the arithmetical properties of interval scales . Unfortunately, although advised on many occasions through public comments on draft evidence reports, ICER rejects the Rasch claims for single attribute interval scores, insistent (with no evidence or proof) that generic instruments such as the EQ-5D-3L/5L preference scores are not ordinal scores but actually ratio scores 13 14 . This is patently untrue 15 16 .…”

Formulary Submissions: Value Claims, Protocols and Outcomes Based Contracting in Rare Disease

“…As outcomes contracting is focused on response, and agreement on the validity of the proposed response measure, then we need agreement on the appropriate standards. The corollary here is that if we insist on these standards for measurable single attributes then we must reject multiattribute preference scores such as those for the EQ-5D-5L (and the preference base QALY) as well, as the majority of disease specific instruments that have only ordinal scores 4 . Claims that the EQ-5D-3L instrument, for example, has ratio properties to support QALYs is patently false when, with the UK value set, the preference equation results in 34.6% of the 243 health states having a negative score or categorized as states worse than death 5 .…”