• News New Blog Banner

  • Study: Functional Limitation Reporting Isn't Delivering the Data Goods

    Since the beginnings of Medicare's Functional Limitation Reporting (FLR) program, APTA has maintained that the system was overly burdensome and unlikely to provide meaningful information. Now a new APTA-funded study fleshes out the case against FLR's usefulness, finding that the codes simply have not been submitted in ways that are consistent with regulations. The holes in data make it difficult to rely on FLR as a source for tracking improvement and outcomes—a deficiency pointed out by APTA to the US Centers for Medicare and Medicaid Services (CMS), and a factor in a CMS decision to propose dropping the FLR altogether.

    The study, e-published ahead of print in PTJ (Physical Therapy), used a 5% random sample of Medicare part B fee-for-service claims for outpatient physical therapy provided in 2014, the first year nonparticipation in the FLR system was tied to claim rejection. Authors tracked FLR and Severity Modifier (SM) coding used throughout episodes of care, analyzing not only the completeness of the reporting, but the projection and documentation of patient improvement from physical therapist (PT) initial examination (IE) to discharge. A total of 114,558 unique patients were included in the study.

    The good news for FLR, if there was any to be had, was that PTs had a high level of submitting complete FLR information at IE, with more than 90% of claims including both a current FLR status code and a projected goal status code. The bad news was that complete reporting fell off dramatically after that, with fewer than 17% of claims required during interim reporting periods—at least once every 10 treatment days—including current and projected status coding. Reporting rates were as low as 9.7% for interim reports of current status.

    Similarly, discharge claims also showed a significant drop in reporting, with an average completion rate of 36.8% for FLR discharge status.

    When it came to planned and documented improvement in functional status as reflected in changes to the SM code, most of reports did identify goals for positive change. For the FLR code sets related to specific functional limitations, at least 85% included estimates of planned improvement; the percentage was slightly lower—78.7%—for the code sets related to "other" PT/occupational therapist (OT) care categories, but that didn’t surprise researchers. The actual level of improvement ratings varied, but, overall, the most frequently used SM for projected goal status was "CI," indicating an improvement to 1%-20% impaired or restricted.

    The study arrives at a time when CMS has proposed the elimination of FLR as part of its move to include physical therapists (PTs) among the providers participating in the Quality Payment Program, a major shift toward value-based payment included in the proposed 2019 physician fee schedule. CMS is accepting comments on the proposed rule through midnight on September 10—APTA will submit comments and encourages members to do the same using a letter template the association created to simplify the process. APTA's analysis of FLR reporting was cited as a factor that led to CMS' decision to drop the program. [Editor's note: for more information on the proposed fee schedule, visit APTA's Fee Schedule webpage, and scroll down to the "APTA Summaries and Fact Sheets" header for a 3-part explanation of the proposal]

    Other highlights from the study:

    • Mobility was the most commonly used FLR code, present on 63.4% of all the PT evaluation claims; "carrying, moving, and handing objects" was next at 16%, followed by "changing body positions" comprising 11.7% of claims.
    • Overall, facility-based claims tended to have lower estimates of planned improvement compared with noninstitutional settings, the most significant being improvement ratings for mobility (80.3% for facilities versus 91.2% for noninstitutional) and self-care (83.7% for facilities, 91.9% for noninstitutional).
    • Most of the FLR codes reported at discharge showed improvement by way of changed SMs, with an average improvement of 1 to 2 steps in the graduated SM code set out in 20% improvement increments. Fewer than 2.5% of all FLR codes reflected worse function, and fewer than 15% reflected no improvement.

    Researchers theorize that the lower reporting rates at discharge may be related to the lack of specific Medicare discharge codes for physical therapy, and that patient drop-out also may come into play, "limiting the ability of the physical therapist to document the FLR codes at the time of discharge."

    The low rates of interim reporting, however, is another matter, according to authors, and a big problem for the FLR system.

    "The reason for the very low completion rate for the interim reporting periods is unknown, and to our knowledge this is the first study to report on this," they write, speculating that the "time burden" involved in completing reports at least once every 10 days of treatment could be a factor.

    "We know that the collection of functional data is core to physical therapist practice, so the question becomes why the FLR system has such significant data gaps," said Heather Smith, PT, MPH, APTA program director for quality who coauthored the study with APTA member Meghan Warren, PT, PhD. "Our study indicates that the real issue is the mechanism through which these data are collected—FLR adds burden and complexity without producing much in the way of useful data."

    While authors acknowledge their study's limits—which include the discharge issue, the potential for inaccurate coding, and an inability to generalize findings to apply to Medicare Advantage beneficiaries—they say the bottom line is clear: the FLR system is not producing the data it was intended to produce—at least not when it comes to physical therapy.

    "The Medicare FLR program, in policy, supports evidence-based practice…however, the current data collection process has significant issues that limit the use and application of the data," authors write. "The ultimate solution to these issues may be the collection of functional data through the use of standardized functional outcome measures that allow for benchmarking by patient condition, at the national and local level, and by setting of physical therapist care." They add that data could be strengthened through sources other than claims, including registries such as APTA's Physical Therapy Outcomes Registry.

    "Without feedback on data completeness and change in function over the episode of care, the collection of these data has limited use," authors write. "Therefore, feedback to providers on performance is a crucial component of making these data meaningful to physical therapists and their patients."