Are Hospital Readmission Reduction Programs Truly Evidence Based?

When politicians attempt to make changes to health policy based on sentiments rather than on evidence-based conclusions, physicians often find themselves trying to meet unrealistic expectations.

Although there’s no denying that trying to keep patients out of the hospital is a good thing, hospital readmissions reduction programs have, over the last several years, proven to be a thorn in the sides of both hospital administrators and physicians alike. Even worse, recent data published in 2018 suggest that the assumption that lower readmission rates means better quality of care may have been overstated. In some cases, the effort to reduce hospital readmissions has resulted in worse outcomes for patients.

The hospital readmissions reduction program (HRRP) is a provision of the Affordable Care Act (ACA) that seeks to link hospital payments for inpatient admissions with quality of hospital care. In short, the HRRP requires the Secretary of the Department of Health and Human Services to reduce payments to hospitals when they are found to have excess readmissions.¹

The idea is that the program would provide incentives for hospitals to improve care coordination and postdischarge planning aimed at reducing hospital readmissions, particularly for acute myocardial infarction, chronic obstructive pulmonary disease, heart failure, pneumonia, coronary artery bypass grafting, elective primary total hip arthroplasty, or total knee arthroplasty. Any unplanned readmission that happens within 30 days of discharge from the index admission counts as a readmission, regardless of the reason for the readmission.¹

Unfortunately, the policy ignores patient-related factors that may affect readmissions but are beyond the physician’s control. For example, despite my efforts to reduce hospital readmissions for congestive heart failure, I have found that readmission rates plateau because of a small number of patients who are readmitted repeatedly because they are either noncompliant with medical therapy and/or continue to abuse alcohol and illicit drugs.

Despite best efforts to set these patients up with home monitoring, frequent cardiology follow-ups, and pharmacy telephone and face-to-face visits for medication counseling, these patients continue to be readmitted 20 to 30 times per year. In resource-poor hospitals, the decreased reimbursement for caring for these patients may inadvertently deplete their already limited funds.

In fact, a recent randomized controlled trial published in JAMA that included 2494 patients hospitalized with heart failure across 10 hospitals in Canada suggests that even in patients who are presumably compliant with therapy (who, for example, are not taking illicit drugs), the implementation of a patient-centered transitional care model did not improve a composite end point of all-cause readmissions, emergency department visits, or death within 3 months.²

The intervention included a nurse led self-care education program, structured hospital discharge summaries, and a family physician follow-up within 1 week after discharge and/or structured nurse home visits followed by more frequent heart failure clinic follow-ups for people with low-complexity heart failure.²

Study investigators speculated a few possible reasons for the lack of benefit. First, they felt that services shown to be effective in previous randomized controlled trials may not have been effective when applied to a healthcare system. Further, the trial included patients with “suboptimal health literacy or self-care” and these patients had an overall higher prevalence of comorbid conditions when compared with patients included in prior randomized controlled trials. Last, unlike prior randomized controlled trials looking at patient-centered transitional care, this trial titrated efforts to risk such that only those patients with the highest risk got home visits and heart failure clinic follow up.²

Although the national readmission rates for ambulatory care-sensitive conditions have dropped since the HRRP was implemented, there is growing concern that the program may have provided incentives for physicians and hospitals in the wrong direction.

An article published in 2018 recognized that financial penalties may have led some physicians to inadvertently avoid readmissions that were indicated.³ In a study aimed at determining whether the implementation of HRRPs was associated with an increase in patient-level mortality, investigators identified a significant increase in postdischarge mortality for patients hospitalized with heart failure and/or pneumonia.³ The study looked at 8.3 million hospitalizations among Medicare beneficiaries for heart failure, acute myocardial infarction, and pneumonia. Although the HRRP was associated with a 0.9% reduction in 30-day and 1-year readmissions, researchers found that there was an overall increase in 30-day and 1-year mortality from 31.3% to 36.3% or hazard ratio of 1.1.³

Thus, the situation is complicated. On one hand, patient-centered transitional care services applied to specific populations with ambulatory care-sensitive conditions have been shown in randomized controlled trials to be beneficial at reducing readmissions and overall cost of healthcare. However, when applied to all patients with heart failure within a health system, the same interventions did not yield a measurable improvement in readmission rates.² In some cases, providing incentives for physicians and hospitals to avoid readmissions for sick patients with chronic diseases may be inadvertently increasing mortality for those patients, thereby blunting the expected benefit of offering patients improved transitional care services.³

I suspect that Bayes’ theorem may apply here. Patients with chronic conditions who are at low risk are likely already doing very well. Very few resources are needed to keep them stable and out of the hospital. Likewise, patients with chronic conditions at high risk may be so sick that it does not matter how much support is thrown their way. They are not going to get better and it may be reasonable to consider palliative care services or more advanced therapies if they qualify (eg, transplant in people with heart failure).

Therefore, the highest benefit for HRRPs may be in targeting those at intermediate risk who are more likely to benefit from standardized care and guideline-directed medical therapy so that they can return to the low risk categories or so that their disease does not progress to a higher-risk state. Unfortunately, identifying these patients early can be tricky. Current risk scores for assessing risk in heart failure have been either too cumbersome or mediocre at best.

In the meantime, as we all work diligently to try to figure out how to best care for these chronic conditions, it may help for Centers for Medicare and Medicaid to reduce pressure. By allowing exemptions that take into account not only the patient’s overall prognosis but also their willingness to comply with guideline-directed medical therapy, hospitals and physicians would be able to focus more of their resources toward a population of patients most likely to benefit from transitional care services.

References

1. Centers for Medicare and Medicaid Services. Hospital Readmission Reduction Program (HRRP). https://www.cms.gov/medicare/medicare-fee-for-service-payment/acuteinpatientpps/readmissions-reduction-program.html. Updated January 16, 2019. Accessed April 1, 2019.
2. Van Spall HGC, Lee SF, Xie F, et al. Effect of patient-centered transitional care services on clinical outcomes in patients hospitalized for heart failure. JAMA. 2019;321(8):753-761.
3. Wadhera RK, Joynt Maddox KE, Wasfy JH, Haneuse S, Shen C, Yeh RW. Association of the hospital readmissions reduction program with morality among Medicare beneficiaries hospitalized for heart failure, acute myocardial infarction, and pneumonia. JAMA. 2018;320(24):2542-2552.

This article originally appeared on Medical Bag