Dean had previously attempted other ways to automate the referral process so members wouldn't be caught in limbo waiting for approval for medical care, and we knew that while automating referrals was a critical first step, we wanted to go much further than that. Ultimately we wanted to begin to capture in-depth utilization data to better understand how our members were using services and to optimize our performance. We also hoped to free up resources for important medical management efforts such as disease and care management.
Drowning in manual inefficiency
Dean's Achilles' heel became apparent when we analyzed the impact of our inefficiencies: The volume of calls to our call center regarding the status of referrals and claims was high. The manual process of handling referrals often exceeded five days and that of communicating referral status back to the provider averaged one week.
In addition, we saw a significant number of claim denials because referral authorizations were not being entered in a timely manner, leading to inefficiencies as well as dissatisfaction among the provider and member communities. Dean also lacked an effective way of tracking claims and referrals. Plus, all of the data about these processes was trapped in paper files, so any analysis we wanted to do required labor-intensive data entry from paper.
As we embarked on our search for an electronic provider connectivity solution, we were looking for a quick-to-deploy and highly customizable proven solution that would map to our unique business processes. In addition to a host of functionality requirements and ease of integration with Dean's and third party vendors' systems, we sought predictable, low-cost pricing, a partnership relationship and technical expertise.
Our research led us to Cambridge, Mass.-based NaviMedix Inc. and its NaviNet product; a Web-based solution that connects health plans with affiliated providers in real time, enabling us to initially automate provider transactions such as eligibility and benefits verification, and referrals and claims processing. Later, we would have the ability to add more features and other types of providers (hospitals, behavioral health providers, etc.). In addition, we were able to use NaviMedix's provider services, including deployment, adoption, usage management, training and support.
Together, we began by stepping back and evaluating our standard operations, existing workflow and data-capture requirements. We evaluated each administrative process or transaction to see how it could be transformed from manual to automated. We worked with our providers to understand their information needs and the processes they thought were most valuable.
Following a deep discovery process, product design and project planning, deployment of our Web portal across our network of more than 1,200 providers was completed within 90 days, followed by ongoing provider support and training. Dean offered the Web portal to providers representing 80 percent of our claims volume and has achieved an adoption rate of 98 percent of our eligible providers and hospitals, and 90 percent of all possible transactions are conducted via the Web portal. Dean's commitment to on-site training and NaviMedix's online support were key to this achievement.
Fifty-two percent of Dean's market share is in rural physician offices. Technology is not always top-of-mind for these smaller providers, and we worried about their acceptance of the Web portal. However, after deployment, we found the adoption rate for these rural providers was nearly 95 percent. These offices were satisfied with the outcome, especially the turnaround time for inquiries and availability of a solution to facilitate real-time communication with Dean.
Improved medical management
The greatest benefit of this Web initiative is that it has laid the foundation for Dean's longer-term medical management initiatives. By automating several time-consuming processes, we are now able to capture detailed information on referral patterns, authorization activity and other common provider communications. This facilitates utilization analysis and enables us to implement process improvements based on our findings.
We have been able to reassign medical management staff for disease management efforts, and we immediately found several ways to achieve cost savings through decreased mailing and human resource utilization.
Once we had the data to analyze the value of the process, one of the first things we did was reduce the number of in-network referrals. The total number of monthly referrals dropped from a typical 8,000 to between 4,000 and 5,000 (see sidebar). We reduced referral and authorization review time from an average of five days to just one day or same-day review. Additionally, Dean decreased notification of authorization determinations from an average of seven days to three days. The most telling sign of success? Since implementing our portal, calls to our utilization management and referral departments have deceased by 27 percent each.
Dean also automated drug prior authorization and can now track provider activity in that area. The hard data on drug prior-auth use has allowed the health plan to eliminate a number of drugs from its drug prior-auth list, resulting in more streamlined patient care and improved provider satisfaction. As new drugs come out, Dean's medical directors can evaluate provider behavior to determine which drugs should be on the prior-auth list.
We can very quickly develop the drug-related questions providers must answer and implement drug prior-auth for new or expensive drugs that should have limited usage. Dean has made every effort to make our prior auth processes transparent to our providers so they can see, before submission, whether or not it will be approved, which allows them to consider alternative therapies or treatment.
Dean has become a more efficient organization, enabling the utilization management department to reallocate three fulltime nurses to focus on case and disease management activities, thus containing medical costs and improving member and provider satisfaction.
The benefits are not limited to improved medical and utilization management, however. Dean has seen a 37 percent reduction in phone calls to our customer service call center, postal costs have been reduced by more than 30 percent and paper costs by 20 percent.
The average number of hours needed for the mail process has been reduced by 33 percent and filing time has been reduced by 60 percent. By eliminating administrative tasks and most of our paper usage, we have been able to reduce the utilization department by five full-time employees. Plus, now with everything entered electronically, no handwritten notes means that errors due to illegibility are greatly reduced.
Dean Health Plan is now a data-driven organization and our automated, Web-based provider communications solution has given the health plan the ability to capture and analyze our medical management data, and predict and prepare for future growth opportunities. By looking retrospectively at quantitative data to see volumes and trends, we can better determine the direction we need to move toward as a business.
Dean Health Plan Reduces Referrals
Prior to automation, since the entire process was paper driven, we had no way of determining what types of referrals we always approved or which reviews of referrals brought value to the process. Once NaviNet was implemented, we were able to review the statistics on the number of referrals approved by category and by specialty to determine the percent we approved. This allowed us to eliminate some referral requirements to improve our efficiency and customer and provider satisfaction with our processes.
A prime example of this was our decision to eliminate the in-network referral requirement. Previously we had required a referral when someone needed to go to a provider outside of their PCP site (mental health, specialty care, etc.). Using statistics from NaviNet, we were able to show we approved virtually all of those referrals and, in some instances, only used it to remind members of limited benefits or potential non-covered care. We determined our members could obtain that information in other ways and eliminated the in-network requirement except for those procedures which required prior authorization (a limited number compared to the total number of in-network referrals).
J.P. Author Information:Julie Pofahl is director of utilization management, Dean Health Plan, Madison, Wis.
