Hospital Discharge and the Transition Home for Poor Patients: “I Knew I Couldn’t Do What They Were Asking Me”

One of the most contentious aspects of the Centers for Medicare and Medicaid Services’ (CMS) Hospital Readmission Reduction Program, which penalizes hospitals for higher than expected readmission rates, is that patient socioeconomic status (SES) is not accounted for (i.e., not risk adjusted for) when calculating hospitals’ readmission rates. Hospitals argue that this disadvantages institutions that care for high proportions of low SES patients because poorer patients have inherently higher risk of readmission. They further contend that hospital interventions to prevent readmissions will not be successful.¹ Policy makers on the other hand point out that if hospitals caring for low SES populations have higher readmission rates that may reflect inadequate quality of care—exactly what the measures are intended to illuminate—and should spur improvement efforts to ensure high-quality transitions from the acute care setting.² At its essence, this is a debate about the extent of hospitals’ responsibility to patients of low socioeconomic status and the degree to which hospitals can influence, within the small window of the inpatient admission, a patient’s trajectory over 30 days following discharge. And all the while, millions of dollars are now at stake through CMS’s Hospital Readmission Reduction Program.

Many prior studies examining the relationship between SES and readmission rates have focused solely on patient-level risk, examining whether patients of low SES have higher readmission rates. These papers often, but not always, find patients of low SES are at higher risk of experiencing a readmission after hospital discharge. It is these studies that the hospitals point to when they argue that patient SES should be accounted for within hospital performance measures.^3,4 However, the more relevant question for the construction of both measures and policies to drive improvements in hospital performance and clinical outcomes is the relationship between SES and readmission rates at the hospital level. Policy makers point to studies (including ones done by our research group) that show wide variation in performance among safety net and non-safety net hospitals, with only small differences in average performance. These studies demonstrate that many hospitals caring for low SES populations achieve low readmission rates, supporting the argument that hospital incentive programs could bring about improved outcomes for low SES populations.^5,6 Few studies, however, have sought to examine the more fundamental question of how hospitals can improve clinical outcomes, such as readmissions, for low SES patients.

In this issue of the journal, a qualitative study provides some useful insights that should inform hospitals and policy makers debating the best approach to measuring performance and improving care.⁷ Kangovi et al. used in-depth interviews to explore the perspective of patients of low SES on their experience of discharge from the medical or cardiac services of the hospital and the transition to a non-acute setting. The study reveals many ways that we can better address the needs and concerns of our most vulnerable patients making the transition from the hospital to home. Led by a community member, the authors interviewed 65 patients from two hospitals in Philadelphia who were either uninsured or Medicaid eligible and came from particularly poor neighborhoods (>30 % living below poverty level). In brief, the low SES patients described powerlessness during hospitalization due to their illness and socioeconomic factors and the approach to discharge processes. In addition, they recounted socioeconomic constraints and competing priorities that complicated or prohibited their ability to perform recommended behaviors. Last, patients reported feeling abandoned after discharge and a loss of individual self-efficacy that resulted from failure to perform recommended health behaviors.

The methodological strengths of this study include the use of grounded theory, an iterative coding approach, and engagement with participants on the findings. However, one potential threat to the validity of the study, noted by the authors in the limitations, is its heavy reliance on the Integrative Behavioral Model for organizing the questions and findings, which implicitly makes assumptions about why a person does or does not perform health behaviors. Moreover, the authors somewhat cursorily explain why they limited their focus to only certain aspects of the model. It is not surprising, therefore, that their findings fit quite well within their pre-established framework. One cannot know whether, or how different, the results might look if the same study team and patients had approached the interview within a less restrictive structure.

Nevertheless, the most powerful and important findings from the study come from the direct statements of the participants. Their articulation of the experience of and their perceptions surrounding hospital discharge and the transition home brings nuance to the debate about responsibility for patient outcomes. One quote encapsulates the experience: “I knew I couldn’t do the things they were asking me to do. So, I just sort of gave up. I knew I would end up back in the hospital.”

This study demonstrates that the hospital discharge process is frequently inadequate for ensuring a safe transition from the acute care setting. Participants experienced misalignment of goals between the care team and patient—e.g., misunderstanding by the care team of what patients feels capable of handling and often a lack of understanding by patients of what is needed/expected—sometimes exacerbated by perceived discrimination and by hurried discharge processes that left inadequate time for questions. The study’s findings also point to challenges patients face upon returning home, including competing priorities and lack of social support. Not all the obstacles described by these patients can be addressed by hospitals, but many can.

The study illuminates important failings in our discharge processes that are not unique to patients with few resources. For example, two other recent quantitative studies demonstrated that at another prominent, high-quality, academic institution, the discharge process is frequently fraught with failures.^8,9 These studies similarly found that patients often had little warning prior to discharge and inadequate understanding of the reason for hospitalization and plan for transition home.

We, physicians and hospitals, must heed the concerns voiced by these vulnerable patients, as failures at discharge have important consequences. Many of the missed opportunities may be minor, or even seemingly meaningless, but as the authors of this study find for the patients they interviewed, the culmination of factors that hindered a successful transition often lead to a lack of self-efficacy, i.e., participants “gave up.” This shows the urgency of finding better approaches and/or interventions to support our most vulnerable patients at the time of hospital discharge, perhaps through collaborative goal setting, as suggested by the investigators. Finally, as the authors note in their discussion, these findings also may help to justify the decision by CMS not to “adjust away disparities” in the readmission measures and instead to “incentivize hospitals to provide care that is centered towards the needs and goals of the low-SES patient” and all patients. There are clear opportunities for hospitals to improve the transitions of patients from the hospital to home, ensuring that patients have adequate time to prepare for discharge, that clinical care barriers are addressed prior to discharge, and that patients are empowered in their care plans. By addressing these frequent failures at discharge, hospitals are likely to help patients avoid repeat hospitalizations and improve quality of care.