Talking Medicare: GAO urges CMS to continue prior authorization efforts

On May 21, 2018, the Government Accountability Office (GAO) issued a report to the U.S. Senate Finance Committee on the use of prior authorization models by the Centers for Medicare and Medicaid Services (CMS). The GAO was asked to examine: (1) the impact of prior authorization on total expenditures, and the potential savings for items or service subject to prior authorization, (2) the reported benefits and challenges of prior authorization, and (3) CMS’ monitoring of these programs, and its plans for future prior authorization. To conduct its study, the GAO looked at payment data and other information provided by CMS. The GAO also interviewed CMS, the Medicare Administrative Contractors (MACs), and selected provider, supplier, and beneficiary groups.

Prior authorization was first implemented by CMS in 2012 for certain power mobility devices (e.g., power wheelchairs) in seven states. Subsequent prior authorization models were implemented for non-emergency hyperbaric oxygen and home health services. Most relevant to our industry, CMS implemented a prior authorization model for repetitive, scheduled, non-emergency ambulance transportation in December of 2014. Originally, this model was implemented in only three states: New Jersey, Pennsylvania, and South Carolina. In January of 2016, the prior authorization model was expanded to include the states of Delaware, Maryland, North Carolina, Virginia, and West Virginia, as well as the District of Columbia.

The GAO’s key finding is that these prior authorization models have been effective in reducing Medicare’s expenditures for various items. The GAO’s analysis of actual expenditures found that the estimated savings from all demonstrations through March of 2017 could be as high as $1.1 to $1.9 billion. Given this fact, it should not be surprising that the GAO is calling on CMS to continue the use of prior authorization.

The majority of the data included in this report relates to non-ambulance services. However, I do want to highlight a few data points noted by the GAO.

From the model’s implementation in December 2014 through March 2017, MACs collectively handled more than 337,000 prior authorization requests, including a total of 3,231 requests for authorization of a repetitive, non-emergency, ambulance patient. This includes 2,620 initial requests, and 611 resubmissions (i.e., subsequent requests for prior authorization following the rejection of the initial request).

The GAO provisional affirmation rate for both initial and resubmitted authorization requests rose in each demonstration between the initial implementation date and March 2017. For example, the GAO noted that the affirmation rate (i.e., the rate at which patients are approved for repetitive ambulance transportation) during the first six months of the non-emergency ambulance model was 28 percent. This rose to 66 percent during the most recent six-month period (October 2016 through March 2017). The GAO noted that MAC officials attributed this increase, in part, to provider and supplier education, which they felt improved the documentation being submitted by providers and suppliers. While this is undoubtedly true, it is also likely the case that the MACs refined their approval process over time.

The GAO estimated the total potential savings from the prior authorization model for ambulance to be nearly $387.5 million from December 2014 through March 2017. Importantly, 90 percent of that savings was attributable to reductions in utilization in the original three states. Moreover, more than half the reduced expenditures took place within the first six months of the demonstration project.

In terms of fitting this report into the larger picture, I think it is best viewed as further confirmation of what we already suspected: namely, that the federal government perceives prior authorization to be an effective tool for combating the perceived overutilization of ambulance to transport patients to and from dialysis. CMS indicated as much when it adopted the program in 2014. Medicare payment data has borne out those expectations. Recently, CMS issued its first interim report on prior authorization’s effectiveness. The GAO’s report adds an independent imprimatur to that belief.

Big picture, all of the stars appear to be lining up for an expansion of prior authorization next year. Stay tuned!!

Have an issue you would like to see discussed in a future Talking Medicare blog? Please write to me at bwerfel@aol.com.