Lalan Wilfong, MD
There is an old saying, “What you don’t know, won’t hurt you.” Unfortunately, when it comes to implementing the CMS Physician Quality Reporting System (PQRS) in a large multidisciplinary oncology practice, nothing could be further from the truth, at least from my personal experience as a medical oncologist at Texas Oncology.
We are an independent practice of more than 400 providers working across 165 sites, and we are affiliated with The US Oncology Network, one of the nation’s largest networks of integrated, community-based oncology practices. I became medical director of quality programs for the practice in January 2015, and one of the first things that landed on my desk was the PQRS initiative, a quality improvement incentive program that provides financial rewards or penalties for reporting healthcare quality data to CMS.
Whereas PQRS was originally an optional program, that is not the case today. Now physicians and advanced practice providers (APPs) who do not provide quality data during the 2016 reporting year may realize penalties as great as -6% of their Medicare reimbursement, which will be assessed in 2018 on all Medicare Physician Fee Schedule charges. By 2021, the penalties or rewards through the proposed CMS quality reporting programs will widen to plus or minus 9%.
There are several ways to report quality data to CMS. In 2014, we chose group-based reporting, which is fairly easy compared with other reporting models. Providers in the group are listed with their tax ID number, and specific metrics are reported. The problem with this model, however, is that nearly 90% of the metrics relate to preventive care rather than things we do in daily oncology practice. Primary care physicians ended up being responsible for most of the results we reported in 2014.
For 2015, we wanted to report on metrics that were more meaningful to oncologists and would accurately reflect the services we provide, so the data could be used to help advance the quality of cancer care. After discussions with The US Oncology Network, we switched to individual-based reporting using an individual reporting model. Each provider would report on 20 patients over age 55, 11 of whom were traditional Medicare patients with an oncology diagnosis tied to office administration of chemotherapy or radiation.
It sounded ideal, since we would be reporting on things we actually do in daily practice, which in turn, should reflect our quality of care.
Good Intentions Lead To Difficult Reporting Challenges
For most of our oncologists, the reporting went smoothly, but we did encounter some serious complications. Under the oncology measures group, providers report on 20 qualified patients. This sounds simple enough, but it turned out to be a huge problem. Physicians coming out of training joined us late in the year and did not have time to fulfill the requirement. Additionally, a large group of providers merged with us in November, but they could not report on their prior patients because those patients were not seen under the provider’s current tax ID. Also, fee-for-service Medicare was required, and some of our sites had many patients with managed Medicare plans, making those patients ineligible.
In the end, about 50 physicians did not qualify under the oncology measures group. Our only option was to report them under the preventive care group, but it was still a struggle to get enough patients and quickly meet the requirements. For example, a specific depression screening questionnaire was required, and many physicians were scrambling at the end of the year trying to find patients to screen.
We also encountered problems reporting our more than 100 APPs, since they did not have chemotherapy charges under their National Provider Identifier (NPI). Consequently, they had to report under the preventive care group. Most APPs bill under the physician’s NPI for patient care services, so we had to quickly find qualified patients for each APP to bill under their own NPI.
Surgeons also had problems because the surgery reporting group required utilizing a specific risk-based calculator with patients. Most surgeons used other risk calculators, so we immediately had to transition them to the required process.