Case Study – Power of Operational Analytics in Healthcare

Summer will be over by the time this article goes into publication. Overall, the U.S. economy is growing, albeit weakly in 2016, and the headwind of the November election might cause further slowdown in the healthcare sector. However, I do not expect any major shift in strategies adopted by the healthcare sector overall. The transformation that we are seeing in healthcare was long overdue. The Affordable Care Act (ACA) certainly catalyzed the change, but given the fact that despite spending 17 percent of GDP on healthcare, the fact that health outcomes in United States are worse than many other developed nations in the world was already driving change. Regardless of what happens in the presidential election, it is unlikely that the core focus will move away from the value-based care paradigm.

With the change in the payment models and move toward capitation, i.e., a fixed amount of payment per capita, many healthcare organizations are responding with a renewed focus on cost reduction through operational excellence. Providers and payers, both for-profit and not-for-profit organizations, have focused on this over the last several years. During this time, we have seen wide-scale implementation of the lean Six Sigma model in healthcare delivery organizations. Along with advanced process modeling and value stream mapping techniques, advanced analytics based on operational data played a significant role in such implementations. Organizations have improved their operational excellence and have built sustainable process changes with the help of operational analytics. This article describes a case study from my personal experience within a managed care organization where we focused on data and analytics to improve operational excellence and achieve substantial cost reduction.

Unmanaged Growth is a Problem!

Before ACA and the Medicaid expansion in California, my client organization successfully managed member lives of a small population of Medicaid enrollees for almost two decades. The State of California paid a capitated amount per member to manage their health. Before the expansion of Medicaid in California (also called Medi-Cal) 20 percent of Californians were covered under the program. After Medicaid expansion, anyone who earned below a certain percent of the federal poverty line along with the undocumented children became eligible for Medi-Cal coverage. This caused a steep increase in Medi-Cal enrollees. By 2015, nearly 30 percent of Californians came under Medi-Cal coverage, and 76 percent of the Medi-Cal enrollees were assigned to the Medi-Cal managed care program. My client received tens of thousands of members from that expansion population. Within six months of expansion, the number of members managed by my client under their managed care program grew by more than 60 percent. The organization wasn’t prepared for such a rapid growth. Healthcare utilization management, a function of the managed care operation, was severely affected by that explosive growth.

Healthcare utilization management is usually one of the highest cost functions within a payer organization since it employs registered nurses and physicians to review the medical necessity of requested medical services. To keep up with the rise in demand for services the department had to increase its workforce at a rapid pace. But at some point they couldn’t find enough qualified candidates to meet the demand. I was told that the department could not meet the state-mandated turnaround time for service requests despite a rapid increase in the overall personnel cost.

Within a week of time study, workflow shadowing and analysis of the production data, I was able to see some key issues. There was no process automation in place. The set-up time was very long as the input requests had a lot of variability. Data entry operators manually entered data from faxed forms. There were many variants of the form used by providers, and several wasteful sub-processes were in place. Operators were not capturing data consistently, causing difficulty in measuring real cycle time. In the absence of any defined staff productivity metrics, managers couldn’t accurately tell how many requests were getting processed by each person per day. As a result, when the backlogs increased, the department was asking for additional staff to manage turnaround time requirements.

Building a Solution, Powered by Analytics

The major area of improvement was found to be the data capture process and ensuring that expensive workers were not idle at any point in time. Wasteful sub processes were analyzed and eliminated as appropriate. We carefully standardized the input data form to eliminate variability in the process. To automate the data capture process we sought help from an optical character recognition service provider and developed an extract-transformation-load process to load the extracted data directly into the production system.

We started to reject all faxed requests that came via non-standard forms with automated fax backs. This helped in both reducing variability as well as the behavior change of the requesters. We set up an automated monitoring process of the input, powered by analytics, to ensure that repeat offenders were identified and contacted regularly by the departmental staff. We defined performance metrics for all departmental staff and enabled managers to analyze their performance with data visualization. A leader board showcased names of the high performers. Competitiveness increased the overall productivity of the department.

Within a few months of the implementation, we saw drastic improvement in the overall productivity of the department: backlogs disappeared, turnaround time was reduced by more than 50 percent, and the processing cost per request was reduced by 40 percent. During this time, membership peaked at almost 100 percent of what the organization had before Medicaid expansion but without any major impact on the departmental throughput or turnaround time. Moreover, overall profitability of the organization improved as the cost curve was bent while revenue increased steadily. Staff morale was improved, as they were able to focus on higher value-added services without feeling stressed by the constant fear of workload and regulatory compliance.

In conclusion, analytics plays a significant role in bringing operational excellence within healthcare. Healthcare organizations will have to elevate their competencies in operational analytics. The sector is going through many transformational changes, and this will continue for several more years. During such times healthcare organizations will have to stay efficient and nimble so that they can eliminate waste and change appropriately in response to the macro environment.