Pilot project compares EHR data against traditional life insurance requirementsOctober/November 2018
Many life insurance carriers have implemented some type of accelerated underwriting program for life insurance. As promising as some of these programs are, however, there is still a balance between efficiency and risk management that needs to be maintained to facilitate continued growth. Carriers could implement an automated process with a fully underwritten price by obtaining medical records at the time of application rather than waiting for labs and physician statements (as is current practice).
Evaluation of EHRs Versus Attending Physician Statements
Nationwide (the company where I work) recently conducted a pilot project with a third-party provider to capture data from EHRs and then compared the EHR-sourced data against data gathered from traditional underwriting requirements. The pilot ran both retrospectively—where we reached out to customers with inforce policies and asked them to provide access to medical records—and prospectively—where we gathered EHR data in parallel to an application going through the traditional underwriting process.
Choice of EHR Source
Life insurance companies are exploring two methodologies for gathering EHR data. Our initial pilot used a portal to allow the life insurance applicant to sign into his or her account with his or her medical provider, and then port the medical records to the insurance company for review. The second method involves the applicant providing authorization, and then a third-party aggregator or HIE collects the medical records from the various data sources. This is similar to how insurance companies work with providers to gather prescription drug data. The third-party provider in this case has created a special process or algorithm to aggregate medical records from multiple parties in the health care industry, and the third-party company contracts with several health care data sources (including health plans or provider networks) to facilitate the data collection for the insurance carrier.
There are pros on cons to each approach. The biggest challenge we observed in our pilot for the user-directed sign-on approach is that applicants and advisers are hesitant to use it. It represents a change in the usual process, and even with incentives, it is difficult to change behavior. When the EHR data was shared, we found the EHR either provided the same quality of data as the attending physician statement (APS) or it would have led to the same underwriting decision as the traditional process in most cases.
For the data aggregator or HIE, the hit rates are still very low, as those working on these solutions are still growing their network of data contributors. It is a concern that using this type of process could provide several cases where either no data is returned or the data is very sparse and captures only a fraction of the applicant’s medical history. There is a fear that implementing this type of approach with a low hit rate could actually make the underwriting process worse because the traditional underwriting requirements are delayed while waiting for the EHR to be retrieved and assessed.
Potential Beyond Replacing the APS
When reviewing the data gathered from EHRs, we looked at it based on two dimensions. First, does the EHR match the data provided by the APS for the given insured? Some people may want to stop here; however, EHRs provide the potential to go beyond the traditional point-in-time requirements and look at the health of the applicant over a period of time. Longitudinal information can give considerably more insight into the health of an applicant over time. We are excited by the possibility of revisiting underwriting guidelines and looking closely at disease progression.
Based on our initial pilot project comparing the data available from EHRs against traditional life insurance requirements, we recognize EHRs are not yet ready to be used for underwriting. For a process to be adopted, we need to make sure it delivers quality data in a consistent manner. User portals are still seen by the applicant as an extra step in the process and are not being embraced despite their potential. Health records obtained by data aggregators are returning a low hit rate that won’t satisfy the requirements of the insurer to reduce the amount of time and effort spent to underwrite a case.
Despite the issues observed in our pilot project, the potential is high in the near future for insurers to use EHRs for underwriting. First, it would improve the quality of underwriting by using verified medical data at the time of the application. Second, it would improve the user experience by reducing the burden placed on the users to answer questions.
The effort required to make this a reality is not trivial, and insurers are looking for innovative partners to help them achieve this automation rather than driving this research and development work internally.
Copyright © 2018 by the Society of Actuaries, Schaumburg, Illinois.