Value-Based Penalties Target Hospitals With High Risk Patients

April 28, 2017 - Hospitals that serve greater volumes of African-American patients and those with more severe conditions are more likely to receive a value-based penalty under the Medicare Hospital Readmissions Reduction Program (HRRP), a new JAMA Cardiology study showed.

UT Southwestern Medical Center researchers also found that the 30-day hospital readmission metric for myocardial infarctions (MI) used by CMS to determine financial penalties did not indicate long-term clinical outcomes.

“Our findings raise concern about the fair and equitable allocation of CMS penalties for readmissions,” James de Lemos, MD, UT Southwestern Internal Medicine professor and senior author, stated in a press release. “Hospitals that take care of larger numbers of patients with socioeconomic disadvantage, including a higher proportion of race and ethnic minorities, are more likely to be penalized, even though quality of care measures and long-term outcomes were not worse for these hospitals.”

Researchers found that almost 43 percent of the 377 studied hospitals faced a potential 1 percent financial penalty in 2013 under the HRRP. CMS penalized the hospitals for higher than expected 30-day readmission rates for heart failure, myocardial infarctions, and pneumonia.

However, the patient population treated at hospitals in the HRRP penalty cohort was 7.6 percent African-American individuals.

The patient population was only 4.5 percent African-American individuals at hospitals with 30-day hospital readmission rates lower than the national average.

Hospitals with higher than average readmission rates for MI also treated more patients with social risk factors. The socioeconomic status scale at hospitals with above average hospital readmissions for MI was negative 0.07 compared to just negative 0.3 at HRRP facilities with below average rates.

Patients with decompensated heart failure at admission, severely depressed ejection fractions, and major bleeding events also tended to receive care at hospitals with higher readmission rates according to the HRRP.

Additionally, the study uncovered that the 30-day readmission rate for MI used in the value-based purchasing program was not associated with care quality during the index hospitalization or clinical outcomes post-discharge.

Researchers found no significant correlation between MI performance measure performance and 30-day hospital readmission rates. Hospitals with below and above average rates in the HRRP similarly performed in-hospital MI guidelines, such as prescribing aspirin upon admission, assessing ejection fractions, and providing reperfusion therapy within 90 minutes for ST-Elevation Myocardial Infarction (STEMI) patients.

Both hospital groups also similarly performed discharge guidelines, including prescribing therapies, offering smoking cessation advice, and referring patients to cardiac rehabilitation.

After adjusting for potential confounders, researchers found that hospital readmission rates above a national average did not result in higher mortality risks within 30 days of discharge and after one year of discharge.

Although the hospital’s performance on the HRRP measure was associated with a greater risk for readmission within a year of discharged. The risk of all-cause readmission within one year was 6 percent higher per 0.1-unit boost in hospital readmission rates above the HRRP standard up to one.

The risk was 4 percent greater per 0.1-unit increase in the rate after the threshold.

In the analysis for potential readmission 31 days to one-year post-discharge, the HRRP measure had no impact on the risk for readmission.

Researchers suggested that CMS should revise the 30-day readmission metric used in the HRRP to ensure hospitals that treat more patients with social risk factors and greater disease severity are not disproportionately penalized.

“It is fundamentally unfair to penalize hospitals for factors that are beyond their control,” stated de Lemos. “We support proposed changes to pay for performance that would consider socioeconomic status in the risk-adjustment methods to calculate rewards and penalties.”

The value-based penalty methodology should account for race and ethnicity as well as measures of symptoms or ejection fraction, researchers added. The program currently only adjusts for age, comorbidity burden, and some disease severity measures.

Researchers pointed out that the value-based penalty methodology laid out in the 21st Century Cures Act may help the HRRP better identify hospitals that deserve the financial setback. The act mandated that CMS compare hospital performance by peer groups divided by similar patient populations, rather than a national average comparison approach.

The HRRP should also include incentive policies that use more diverse care quality and patient-centered outcome measures to ensure the penalties reflect low-quality care. The measures should be directly attributable to care furnished in the hospital setting.