Look behind the lecture: the value of value-based contracts
In this feature, Richard Gliklich (OM1, MA, USA) discusses his presentation from the 2019 Drug Information Association Annual Meeting (23–27 June, CA, USA) on the necessity and importance of value-based contracting (VBC).
Please could you introduce yourself and your institution?
I am the CEO of OM1 (MA, USA) — a real-world outcomes and technology company focused on measuring, comparing and predicting patient outcomes.
What is value-based contracting and why is it important?
Value-based contracting (VBC) is a performance-based approach to contracting for medical products or services where performance is evaluated by health outcomes or results. A VBC can be managed as a single bundled payment or a variable payment based on the degree of success of a product. VBC has emerged as an approach to flattening the ever-rising cost curve in healthcare; it is important because an increasing number of payers — including the Centers for Medicare and Medicaid Services (CMS; MD, USA) — are experimenting with value-based models to better manage healthcare costs.
How may we use provider clinical outcomes in value-based contracting for rare conditions?
If clinical outcomes are measurable, they can be used to determine the percentage payment that a manufacturer or provider might receive. For more common conditions, one might consider the impact of a particular product or procedure on a representative group of patients and evaluate the most relevant outcomes for that condition being managed on the population of interest. It is very important to utilize outcomes that are relevant to patients and providers in these analyses and not simply outcomes that are accessible from available data. This has been a problem over the past few years because outcomes are either not standardized, or hard to assess without more clinically deep information that is difficult to collect.
“It is very important to utilize outcomes that are relevant to patients and providers…and not simply outcomes that are accessible from available data.”
For rarer conditions, the problem becomes trickier because a single payer may not have a large enough sample size of treated patients to fairly assess the impact(s) of an intervention. For example, if a patient undergoes stem-cell therapy under a VBC, the results may be binary — the procedure succeeds or fails. However, if the sample size is small, the therapy provider or supplier may appear more or less successful than if the sample size were significantly larger. This creates challenges for developing reimbursement models that are fair, but still promote innovative therapies.
How may we supplement electronic health record and claims data endpoints with patient-reported outcomes, and why is this important?
The Outcome Measures Framework is a federally funded effort to develop a more standardized approach to measuring outcomes between and within different conditions. That framework, which has been vetted by hundreds of healthcare stakeholders to date, categorizes patient and clinician relevant outcomes into five domains across most conditions. One of those core domains is patient-reported outcomes (PROs); understanding the patient experience is critical to understanding the ultimate outcomes of many conditions and procedures. Supplementing electronic health records and claims data with PROs requires direct collection of information from patients. This may create administrative and patient burden but is incredibly valuable and increasingly being looked to as a means to better understand outcomes.
How may we standardize RWE to formulate endpoints for prospective and retrospective studies?
As mentioned, the Outcome Measures Framework — and other efforts at endpoint standardization by the FDA, National Institutes of Health (MD, USA) and other groups — are critical to our successfully using RWE to compare results between studies and/or aggregate smaller studies to answer key questions. These guidelines will also allow us to monitor the same endpoints before and after approval of drugs and devices.
What are some of the challenges associated with employing RWE and electronic/modern data sources (including wearables and social media-sourced data)?
Data sources for RWE need to be representative of the target populations that studies are seeking to draw conclusions for. Beyond technical issues with including certain electronic/modern data sources, one needs to consider if focusing on wearable or social media-sourced data introduces any biases into the data. For example, from an economic perspective, how representative are patients who have wearables?
“Quality and validity in data generation is a planned, not accidental, process.”
Similarly, how demographically representative are patients who frequent certain social media platforms? How can information entered on certain social media sites be validated? These questions need to be rigorously answered as we begin to include and employ these important, but sometimes inherently biased or flawed, data sources.
How may we work towards generating data that increasingly high quality and valid?
Quality and validity in data generation is a planned, not accidental, process. If we aggregate information without intent or purpose, we will end up with data exhaust and poor-quality data. Data needs to be gathered intentionally — leveraging existing sources wherever possible — with clear plans and efforts to ensure quality at every stage.
How do you see the use of RWE evolving in the future?
I believe that the use, and perceived value, of RWE will dramatically increase in the future as we increase the clinical relevance, representativeness, scale and quality of the real-world data being utilized, and stakeholders become more familiar with the value and limitations of such data. I think we are still only scratching the surface of the use of high-quality real-world data in drug discovery, clinical development, peri- and post-approval research, market access, value-based care, personalized medicine and decision support. The opportunities for RWE use are enormous, but the technical and scientific challenges are very real.
Disclosures:
Richard Gliklich is an employee and shareholder in OM1, Inc.
The opinions expressed in this feature are those of the interviewee/author and do not necessarily reflect the views of The Evidence Base® or Future Science Group.