Few would argue that the US health care systems’ complex reform measures and payment methodologies sometimes create incentives or disincentives that are not thoroughly understood prior to their implementation. Certainly that may be the case for payment reforms mandated by the “Affordable Care Act of 2010” involving the withhold of part of the provider or plan's negotiated payments based upon performance measures and paid at some point in time long after the services have been rendered.
The STAR ratings applied to Medicare Advantage plans and hospitals require the collection and dissemination of many metrics in order to establish the scores, adding considerable burden to the plans and providers alike. These scores, do, however, attempt to “level the playing field” by providing the basis for comparison, assuming there are enough common data points between providers or plans that allow a fair comparison.
In my experience studying value-based contracting and performance metrics, both plans and hospitals have struggled, in particular, with the measure relating to hospital readmissions. The penalties are onerous for exceeding the threshold, but the means to impact them is often elusive.
How, for example, can a hospital or a payer predict and prevent the worsening of a CHF patient who defervesces over several days after being discharged in reasonably stable condition? What if the plan or the hospital has a higher percentage of such fragile members?
Health executives have argued for at least four years now that the way the Centers for Medicare & Medicaid Services predicts readmissions rates is not the most accurate nor the fairest, and this assertion has been backed up now by researchers at Harvard Medical School who have authored a study published in the May issue of the Journal of General Internal Medicine. They concluded that a patient’s “functional status, rather than comorbidities, was a better predictor of whether someone would be readmitted to the hospital.”
(See the link: http://www.healthleadersmedia.com/content/QUA-317519/CMS-Predictive-Readmission-Models-Not-Very-Good )
Jeffrey Schneider, MD, the primary investigator, comments that, "This raises a question of whether Medicare is really using the best predictors to really understand readmission," as well as questions about how fairly hospitals are being financially penalized.
Schneider references CMS fines to more than 2,200 hospitals that totaled $280 million in 2013 for the excess 30-day hospital readmissions measure alone. Clearly, this indicates that using an accurate readmission model is critical to the provider community.
The researchers concluded that the models used by CMS are not very good predictively, relying on simple demographic information like age, gender or comorbidities. They also claim that there is growing evidence that a patient’s functional status is a good predictor of all sorts of hospital outcomes when they reviewed retrospective data from over 120,000 patients in a medical rehabilitation database who had been admitted over a nine year period to inpatient facilities with a medically complex impairment group code. The patients themselves are heterogeneous but when measured using the Functional Independence Measure (FIM), which looks at 18 tasks such as eating, dressing, bathing, toileting, grooming, and climbing stairs, a strong correlation exists between the patient’s ability to take care of himself independently and a lower likelihood of readmission.
Schneider and his colleagues recommend that providers and payers begin piloting functional interventions and determining functional measures at discharge to help with risk-stratifying for readmissions.
That, of course, and lobbying CMS to take the sting out of an ineffective readmission measure that has harmful financial consequences.
