I would like to offer a response to the May 2018 Journal of Clinical Pathways article titled “Factors Influencing the Implementation of Value-Based Contracting Between Pharmaceutical Manufacturers and Payers.” This well-written article presents a cogent explanation of some fundamental technical reasons impeding the implementation of value-based (VB) design contracts.1
My perspective is based on my previous position as a predictive modeler/analyst for several years, working with teams of clinicians, actuaries, finance staff, IT, and analysts to prepare information necessary to create contracts in Oregon for coordinated care with shared savings back to provider groups.
Many payers continue to rely on claims-based diagnoses due to lack of more granular data. The development of applications that capture multifaceted information and the improvement and flow of electronic records will be increasingly valuable in the ability to show attenuation of risk over time. Additional information, such as patient engagement, access and transportation, or social support structures may allow the development of models that can more closely relate patient factors to risk reduction. For example, what is the role of improved transportation or social support in medication compliance? Feedback to clinical teams on the independent contributions of these factors may inform workflows generally or help develop patient engagement strategies.
The article mentions “Keeping It Simple,” which is an important mantra to have in decision support and contracting. The work underlying contracting is not simple, though, so results need to be translated from scientific language to financial language, eg, data scientists use terms like “beta estimates and likelihood.” But decision makers need information and communication in the form of questions like, “How much of a reduction needs to be shown?” And financial staff need to understand these likelihoods to apply monetization of treatments and effects (ie, “How do I price a reduction in negative outcomes?”).
Also, those involved in VB insurance design contracting may not always be asking for the information they need to create meaningful contracts, simply because their field of training has not prepared them in terms of the proper terminology. For example, how often are calculations such as sample size or statistical power requested of analysts? Does the analytic staff have the skill to prepare these? Yet, these calculations are necessary to foretell whether a meaningful outcome can be achieved within a specified period of time within a number of patients. Statisticians can provide information showing trade-offs in sample and effect size: How much does a metric need to be reduced to be both clinically meaningful and statistically significant? How much time would be necessary to show an effect? How many patients are needed to show a significant lowering of outcomes after a specific therapy?
Those who understand these concepts and deploy effective and informed models using expanded data frames and who can improve internal communications between modelers, data teams, and decision makers will likely create effective VB contracts that make sense to all parties.
Some may say that these perceptions are too idealistic, but I believe that that ideal world could be built with enlightened leadership (learn what you need and hire for that) and not unduly raise analytic budgets (academic analysts make half what business analysts make but are not hired because they are seen as “PhD researchers”). The insurers I worked for had to learn why one would use research-level analytic methods to obtain better results; the leadership then implemented enlightened analytics after they understood the reasoning behind it.
It is easy to say that the ideal cannot be attained—that we cannot get this done in a business setting. But I wholeheartedly applaud the efforts of those working in this field who are trying to attain the ideal, or at least get closer to it—to improve health and provide the best care at the lowest possible price.
Reference
1. Buyse M, Carter S, Sarnataro K. Factors influencing the implementation of value-based contracting between pharmaceutical manufacturers and payers. J Clin Pathways. 2018;4(4):27-30.
