Picture of Abraham Dada

Abraham Dada

UCSF Medical Student

Socioeconomic Journal Club

Article: Teaching and Safety-Net Hospital Penalization in the Hospital-Acquired Condition Reduction Program

The principles of beneficence, nonmaleficence, justice, and respect for autonomy are cornerstones of clinical care. The Hospital-Acquired Condition Reduction Program (HACRP) stands as a testament to these values, adjusting Medicare payments according to the quality of care provided. The program tracks avoidable in-hospital complications, penalizing the lowest performing hospitals with a 1% Medicare payment reduction. While well-intentioned, the HACRP has historically disadvantaged large teaching and safety-net hospitals. Although Medicare has made methodological overhauls to mitigate this bias, recent findings by Serpa et al suggest that these changes may not be enough.

The study found that safety net hospitals (odds ratio [OR], 1.41), major teaching hospitals (OR, 1.94), public hospitals (OR, 1.62), and level 1 trauma centers (OR, 2.05) still bear the brunt of financial penalties. Notably, safety-net hospitals with major teaching responsibilities were more than twice as likely to face penalties compared to non-safety-net, non-teaching hospitals (OR, 2.15).  This pattern is concerning as it threatens the stability of hospitals that are critical to underserved populations and to the cultivation of the next generation of clinicians. Moreover, the 2020 penalty status (issued under the old HACRP) proved challenging to reverse in 2021 (under the revised HACRP)–especially for those with high teaching responsibility or public hospitals–with only a meager 16.7 percent of major teaching hospitals improving their status compared to 41.7 percent of non-teaching hospitals.

It stands that much is to be desired from the current renditions of the HACRP. The authors propose a solution inspired by the Hospital Readmission Reduction Program (HRRP), which successfully reduced over penalization by considering the socioeconomic complexities of patient populations. By stratifying hospitals based on the proportion of patients dually enrolled in Medicare and Medicaid, the HRRP was able to decrease their over-penalization of teaching and safety-net hospitals, suggesting a viable path forward for the HACRP.

The findings from Serpa et al. serve as a call to immediate action to address the systemic inequities perpetuated by the current HACRP methodology. As we strive for excellence in clinical outcomes, it is crucial that our pursuit does not inadvertently penalize those committed to educating the next generation of physicians and serving our most vulnerable population. As policymakers and healthcare leaders contemplate the future of hospital quality improvement programs, it is imperative to prioritize adjustments that ensure fairness and equity.

Picture of Anthony DiGiorgio, DO, MHA

Anthony DiGiorgio, DO, MHA

Value based purchasing, where hospitals & physicians are reimbursed based on quality metrics, was a good idea, in theory.

It has completely failed in practice.

Facilities that treat poorer patients are disproportionately penalized.  We showed this happens to neurosurgeons in the MIPS program.  Many other studies have shown disproportional penalties to safety-net facilities and hospitals which treat minority patients.  Quality metrics also impose astronomical costs to the healthcare system while lacking any evidence they actually improve outcomes.  This alone should be enough evidence for CMS to go on a “quality diet.” 

Why the disproportional penalties to safety-net facilities?

The first reason is that quality metrics are expensive.  Johns Hopkins recently reported they spend $5.6 million on 162 quality metrics annually.  The average 161 bed community hospital employs 4.6 full time staff just for quality metrics.  Hospitals which are getting by with mostly Medicaid patients simply don’t have the resources to spend on quality metrics.  Collecting metrics and gaming the numbers takes money.  Institutions pay personnel to comb through notes and find those wonderful “coding queries” to help improve their quality metrics.  Safety-net hospitals can’t afford those luxuries when they can barely afford essential staff.

Second, it is impossible to adequately adjust for socioeconomic status (SES).  As pointed out above, CMS is attempting to account for SES in metric risk adjustment.  This is a welcome step, as SES was previously unaddressed.  A homeless individual with a lumbar stenosis and a wealthy CEO with a lumbar stenosis affected your expected complication rate equally.  Now, there is some accounting for SES.  This will never be enough, because not everything that counts can be counted.  There is no way to measure, and adjust for, things like trust relationships and social support networks.  Human beings are not reducible to numbers and risk scores.  To think they are, in the name of value based purchasing, is overly simplistic. 

It’s far passed time CMS (and Congress) realize this.  We need to end the quality metric regime.