Transitional Care Challenges of Rehospitalized Veterans: Listening to Patients and Providers

Abstract

Readmissions to the hospital are common and costly, often resulting from poor care coordination. Despite increased attention given to improving the quality and safety of care transitions, little is known about patient and provider perspectives of the transitional care needs of rehospitalized Veterans. As part of a larger quality improvement initiative to reduce hospital readmissions, the authors conducted semi-structured interviews with 25 patients and 14 of their interdisciplinary health care providers to better understand their perspectives of the transitional care needs and challenges faced by rehospitalized Veterans. Patients identified 3 common themes that led to rehospitalization: (1) knowledge gaps and deferred power; (2) difficulties navigating the health care system; and (3) complex psychiatric and social needs. Providers identified different themes that led to rehospitalization: (1) substance abuse and mental illness; (2) lack of social or financial support and homelessness; (3) premature discharge and poor communication; and (4) nonadherence with follow-up. Results underscore that rehospitalized Veterans have a complex overlapping profile of real and perceived physical, mental, and social needs. A paradigm of disempowerment and deferred responsibility appears to exist between patients and providers that contributes to ineffective care transitions, resulting in readmissions. These results highlight the cultural constraints on systems of care and suggest that process improvements should focus on increasing the sense of partnership between patients and providers, while simultaneously creating a culture of empowerment, ownership, and engagement, to achieve success in reducing hospital readmissions. (Population Health Management 2013;16:326–331)

Background

Readmissions to the hospital are common, costly, and identify patients at high risk for adverse outcomes.^1–3In addition, hospital readmissions represent a marker of complex inefficiencies in care delivery systems that often extend beyond the hospital.^4,5 Reducing readmissions is a key policy initiative embedded in the Patient Protection and Affordable Care Act (PPACA),⁶ and now constitutes a major focus of health care systems nationwide.^2,5,7 Previous studies suggest that some readmissions can be avoided by redesigning the hospital discharge, particularly through incorporation of interdisciplinary teamwork and implementing evidence-based transitional care models.^2,8–14

Although the Veteran's Health Administration (VHA) is not subject to penalties for Medicare hospital readmissions now mandated under the PPACA, reducing readmissions and improving the quality of patient-centered transitional care are core strategic goals of the VHA. San Francisco Veterans Affairs Medical Center (SFVAMC) is part of the federally integrated VHA system for qualified US Veterans, and, with a 124-bed hospital and dozens of supporting ambulatory clinics and services, it delivers care to more than 310,000 Veterans living in an 8-county area of Northern California. Like other Veterans Affairs (VA) medical centers, hospital readmissions are a significant challenge for SFVAMC. Recent data suggest that Veterans older than 65 years of age who seek care at SFVAMC have 30-day readmission rates of 16.8% and 90-day readmission rates of 28.6%. Thirty-day readmission rates have risen by 3% over the past 5 years.

Despite increased national attention on reducing hospital readmissions and improving the quality and safety of care transitions, little is known empirically about patient and provider perspectives of the unique transitional care needs of rehospitalized Veterans. In 2010, SFVAMC joined a regional initiative—Avoid Readmissions through Collaboration (ARC) (www.avoidreadmissions.com)—supported by the California Quality Collaborative and Center for Quality System Improvement. ARC is a learning community forum for hospitals and their outpatient providers designed to increase awareness of characteristics of evidence-based transitional care models, engage in rigorous data collection and evaluation, enhance local networks supporting communication of implementation ideas, and provide expert mentorship within and across organizations. The ultimate goal for all hospitals involved with ARC is to reduce hospital readmissions by 30% by December 31, 2013.

Through ARC, the authors sought to understand more deeply the existing complexities, stakeholder roles, and common failure modes in systems that affect transitions of care at SFVAMC. One of the key objectives was to better understand patient and provider perspectives of the transitional care needs and challenges faced by rehospitalized Veterans. This article provides an overview of the authors' qualitative findings from patient and provider interviews and elucidates the unique transitional care challenges for this vulnerable population.

Methods

Patient perspectives

Semi-structured bedside interviews were conducted with 25 hospitalized Veterans on medical/surgical units between June 1 and July 15, 2011. Hospital administrative/census data were used to identify patients who were currently hospitalized after previously being admitted to SFVAMC within the prior 90 days. The patient's assigned nurse was approached to ensure that a bedside interview would not interfere with patient care. Patients were excluded who had any acute medical crisis, aphasia or inability to speak, or acute psychiatric symptoms.

Eligible patients and their accompanying family members or caregivers were approached during the day shift (8am–4pm). They were informed that the project was a quality improvement effort, in which the authors sought to understand the experiences and challenges each patient faced when recently readmitted to the hospital. After receiving verbal consent from the patient and/or caregivers, the interviewer (NS) reviewed the electronic medical record (EMR), spoke with the patient, and made brief handwritten notes. Interviews typically lasted 15–25 minutes.

Immediately following the interview, thorough field notes were completed and analyzed using basic Grounded Theory methods. Salient topics across interviews were highlighted and assigned a numericcode, which allowed for exploration of broader themes that appeared repeatedly across the sample. A second investigator (CS) then independently examined all field notes and the preliminary coding schema with excellent agreement on the themes that emerged. Interviews were conducted and analyzed until thematic saturation was reached.

Interdisciplinary provider perspectives

The EMR was used to identify the inpatient attending physician and the outpatient primary care provider (PCP), as well as various members of each Veteran's inpatient care team (ie, nurses, social workers, physical therapists, occupational therapists). A semi-structured interview was created using a secure online survey tool; an e-mail was sent to each patient's PCP within 10 days of the interview. The authors followed up with a phone call that included an explanation of the project and the survey tool. Members of the inpatient care teams assigned to patients in the cohort also were engaged in in-person or phone interviews utilizing the same semi-structured interview questions. Notes from interviews with all care providers were subjected to the same Grounded Theory methods already discussed to identify salient provider themes.

This project was reviewed by the University of California, San Francisco Committee on Human Research and the San Francisco VA Research and Development Committee and adjudicated as a quality improvement activity not requiring Institutional Review Board oversight.

Results

Descriptive characteristics of the patient cohort are displayed in Table 1. The primary readmission diagnosis-related groups for the sample were: kidney/urinary tract infections, heart failure, renal failure, intracranial hemorrhage/cerebrovascular accident, peripheral vascular disease, and nutritional/metabolic disorders.

Table 1.

Descriptive Characteristics of a Cohort of Veterans Readmitted in the Past 90 Days (n=25)

Age (mean years)	68
Race/Ethnicity (%)
Black	32
Hispanic/Latino	4
Native American	4
White	60
Marital status (%)
Married	24
Divorced	20
Never married/single	40
Widowed	16
Place of residence (%)
City of San Francisco	60
Santa Rosa area	12
Other California counties	20
Outside of California	8
Mental health/substance abuse history (%)	73
Average number of readmissions in prior 90-day period	2.6
Average days inpatient in prior 90 days	32
Average length of stay (days)	7

Table 2 describes the types of interdisciplinary providers interviewed regarding the cohort of readmitted Veterans.

Table 2.

Different Types of Interdisciplinary Providers (n=14) Involved with the Care of the Cohort of Veterans Readmitted within the Past 90 Days

Provider type	Inpatient	Outpatient
Physician	2	4
Registered nurse	3	1
Social worker	2	0
Physical therapist	1	0
Occupational therapist	1	0

Patient/caregiver perspectives

The dominant themes that emerged from patient interviews were: (1) knowledge gaps and deferred power, (2) difficulty navigating the health care system, and (3) complex psychiatric and social needs (Table 3).

Table 3.

Patient Perspectives of Their Readmission Challenges and Transitional Care Needs

Knowledge gaps and deferred power	“I never needed a doctor before so I don't know mine.” “I take so many pills, I have no idea. Go look in the computer.”
Difficulty navigating the health care system	“My doctor is really hard to get a hold of. I know when he works in the emergency room, so I just go to him when I need something-I just go down to the emergency room and he takes care of me there.”
Complex social and psychiatric needs	“I am on a fixed income. When my wife died of cancer and stopped eating, I stopped eating too.” “I use the VA vans. Without those vans, I am a deadman. Those vans are my lifeline.”

Knowledge gaps and deferred power

Knowledge gaps were exemplified by patient inability to recall assigned PCPs, active medications, or current care plan details. Many patients deferred to the EMR and often requested that the interviewer use the EMR to find details of care. In general, patients frequently deferred power to nurses and physicians and displayed a lack of perceived control or ownership over their care.

Difficulty navigating the health care system

The majority of patients expressed difficulty receiving appointments, obtaining medications, and paying for transportation to necessary outpatient follow-up. No patient in our sample contacted his or her PCP prior to seeking urgent or emergent care, even though the majority of patients were readmitted through the emergency department. Although they identified access to transportation, nursing, occupational therapy, and physical therapy at home as key factors helping them avoid rehospitalization, they often expressed that these services were difficult to obtain.

Complex psychiatric and social needs

The majority of patients interviewed had either a prior or active history of psychiatric illness, often with co-occurring substance abuse. Psychiatric comorbidities included depression, post-traumatic stress disorder, anxiety disorders, schizophrenia, and bipolar disorder. Many patients expressed severe financial concerns, and many articulated either current or recent homelessness.

Interdisciplinary provider perspectives

Interdisciplinary provider interviews revealed 4 major themes: (1) substance abuse and mental illness, (2) lack of social or financial support and homelessness, (3) premature discharge and poor communication, and (4) nonadherence with follow-up (Table 4).

Table 4.

Provider Perspectives Regarding Reasons for Readmission of Patient Cohort

Substance abuse and mental illness	Nurse:“Patient is not able to recognize limits in his abilities.” Inpatient MD:“We need an inpatient substance abuse program for detox.”
Lack of social/financial support and homelessness	PT/OT:“Patient is a continued fall risk and needs help at home.” SW:“Caregiver is unable to perform necessary duties to assist patient.” SW:“Patient is unhappy with home situation. No other options exist based on his level of income currently.” Inpatient MD:“Patient needs a subacute level of care.”
Premature discharge and poor communication	Nurse:“Patient was still not feeling well prior to discharge.” PCP:“Patient is not at his baseline mental status and shouldn't be discharged yet.”
Nonadherence with follow-up	PCP:“Patient was assigned to me but I've never met him. He only goes to the ED for care.” PCP:“I knew about my patient's last admission, but wasn't directly contacted about it.” Nurse:“Patient didn't answer 48-hour follow-up call but came to the ED instead.”

MD, medical doctor; PCP, primary care physician; PT/OT, physical therapist/occupational therapist; SW, social worker.

Substance abuse and mental illness

Nursing staff often expressed concern that patients with dual diagnoses (substance abuse and mental illness) frequently were not able to recognize limits in their abilities, which negatively impacted effective decision making regarding safety, adherence, and follow-up. One physician commented that Veterans and the medical center could potentially benefit from an inpatient substance abuse program for detox, given the complex medical and psychiatric care needs of some dual diagnosis patients.

Lack of social or financial support and homelessness

Similar to patients in the cohort, many members of the interdisciplinary team highlighted the lack of social and financial support, resulting at times in homelessness, as key challenges in reducing readmissions. Low levels of monthly income constituted an important barrier for patients seeking community-based services or improvements in their home environments. In addition, lack of an identified, available, or sufficiently skilled caregiver often presented safety concerns for Veterans who sought to remain at home alone, and frequently served as a primary cause of frequent admissions and ultimate placement in a long-term care facility. Many providers felt that addressing all of these interdependent and multidimensional challenges was very difficult during a single inpatient admission, in part because so many types of resources and segments of the care delivery system required coordination that extended beyond the hospitalization.

Premature discharge and poor communication

Premature discharge and poor communication also emerged as significant factors contributing to patient readmission risk. Providers expressed concern that some persisting physical symptoms were not fully addressed (or stabilized) during the initial hospital stay and therefore contributed to the patient's readmission. Even during patients' current readmissions, some providers perceived that certain information regarding patients' clinical status was partially ignored or disregarded because the discharge process had already been started. Poor communication also occurred between inpatient interdisciplinary providers, between providers and patients during hospitalization, and across care settings. In particular, there was a perceived lack of communication between inpatient and outpatient providers, especially regarding key measures of function such as patients' baseline cognition and prior capacity for independent living.

Nonadherence with follow-up

Across all interdisciplinary providers, but particularly among the inpatient and outpatient physicians, there was a strong perception that patient “nonadherence” or “noncompliance” with follow-up appointments and medical recommendations played a major role in hospital readmission. In one case, a PCP voiced frustration because he was responsible for a patient in our cohort, yet had never met the patient because of missed outpatient appointments. Although providers often commented on the impact of patients' behaviors on the severity of their illnesses, providers also recognized that larger system challenges and frequent breakdowns in communication often exacerbated patients' struggles with follow-up, transportation, and limited financial and social resources.

Discussion

Recently hospitalized Veterans at highest risk for return to the emergency department (a near miss) or rehospitalization typically face a bewildering combination of medical, psychiatric, and socioeconomic challenges, including homelessness.^15–20 By exploring patient and provider perspectives on the causes of failed transitions in care, this study identifies important barriers faced by this vulnerable Veteran population that must be addressed to achieve sustained improvements in health care quality and hospital readmissions.

Veterans in this cohort displayed significant knowledge gaps regarding their postdischarge needs and limited capacity to navigate a complex health care system. Such “gaps in understanding” also have been found in other studies.^21–25 Paradoxically, however, these knowledge gaps often did not spur Veterans to address deficits through self-advocacy or efforts to partner with providers. Rather, they often communicated a strong sense of deferred power to nurses and physicians and a general lack of perceived control, ownership over, or interest in proactively managing their care—a finding that has not been elucidated elsewhere. This attitude often led to an inability by these high-risk patients to engage with available services that potentially could have prevented rehospitalization.

Interdisciplinary providers who participated in this study often mirrored Veterans' sense of limited empowerment. Providers commonly felt that the sheer complexity, prevalence, and severity of co-occurring medical, psychiatric, social, and financial factors common in this patient population present a significant challenge to the current health care system. Although provider interviews demonstrated recognition that many programs are designed to assist the vulnerable Veteran population, providers also felt that these programs often were not adequately connected, transparent, or accessible to patients to avert readmissions.

Although patients and providers in this study tended to share similar views regarding the challenges and systems deficits leading to frequent readmissions, the authors did not find a strong sense of common purpose or shared responsibility for improvement. Rather, patients and providers often tended to defer joint efforts or cede responsibility for care to the opposite party. For example, rather than focusing on what they could do to ensure more appropriate follow-up, patients seemed to concentrate their energy on explaining how difficult the system was to navigate. Rather than focusing on how they could help simplify the process by which patients access care, providers spoke most frequently about patient nonadherence and substance abuse as barriers to collaboration.

This paradigm of disempowerment and deferred responsibility contributed to ineffective care transitions resulting in readmissions. Importantly, the themes from this qualitative data highlight cultural constraints on systems of care rather than isolated failure modes in established processes within the hospital, clinic, or elsewhere. The authors suspect that the sociocultural context and hierarchical nature of prior military service among Veterans may make them more likely to defer authority for decision making to providers. However, most of the providers involved in this study were not Veterans, and yet they often seemed equally influenced by these cultural norms. This finding raises larger questions about the relationships between patients, providers, and the systems that govern access to health care services when complexity is high and responsibility is diffused broadly. Because complexity and diffusion of responsibility are increasingly common characteristics of health care systems,^26,27 it seems likely that results would be similar were our study repeated in a non-VA setting.

Quality improvement efforts in health care settings must simultaneously focus on detailed care processes (eg, process mapping, failure modes, and effects analysis) and the pursuit of robust culture change when seeking to improve care transitions. Within the VHA, negotiating new cultural norms should focus on moving away from such a paradigm of deferred responsibility. Part of this work must include developing appropriate systems of care that support and empower shared decision making, greater patient engagement, and motivation for increased patient self-efficacy. An important first step for both patients and providers will be to acknowledge the mutual responsibility for and benefit of managing serious chronic conditions in the ambulatory setting, while also addressing factors such as substance abuse, mental illness, social isolation, and limited financial resources.

Two examples of this approach are already under way within SFVAMC. First, SFVAMC (and the VHA more broadly) is currently redesigning primary care delivery to incorporate a team-based model—the Patient Aligned Care Team (PACT)—that includes all elements of the patient-centered medical home. A fundamental aim of PACT is enhanced coordination among staff and use of evidence-based tools to encourage healthy lifestyle choices among patients.^28–30 Second, SFVAMC has used implementation of PACT as an opportunity to increase motivational interviewing, a style of patient-centered health care dialogue that helps patients identify and resolve discrepancies between their actual and desired behavior.^31,32 The authors believe motivational interviewing may be particularly powerful for Veterans, as prior research has demonstrated the effectiveness of motivational interviewing in reducing substance use, improving dietary habits, and engaging patients in chronic disease management.^32–35 Motivational interviewing may be a particularly powerful tool because the vast majority of Veterans in this study lacked someone who could serve as a caregiver or surrogate advocate.

In conclusion, these findings shed light on the unique patient and provider perspectives regarding the needs and challenges of rehospitalized Veterans. Understanding, incorporating, and addressing these important perspectives will be critical to implementing and tailoring evidence-based, patient-centered, transitional care interventions to decrease readmissions within the VHA. It is encouraging that care design initiatives already under way at SFVAMC are well suited to addressing the needed cultural change highlighted in this study. Increasing the sense of partnership between patients and providers, while simultaneously creating a culture of empowerment, ownership, and engagement, is critically important for success in ongoing efforts to reduce hospital readmissions and may prove beneficial for non-VA health systems as well.

Acknowledgments

We would like to thank the patients, providers, and medical center leadership at the SFVAMC for their time and commitment to this important quality improvement endeavor. The authors are also grateful to the California Quality Collaborative and Center for Quality Systems Improvement for sponsoring the Avoid Readmissions through Collaboration (ARC) learning collaborative. We further wish to sincerely thank Pat Teske, RN, MHA for her tremendous time, guidance, and expertise as our quality improvement ARC consultant. We are also grateful to the Gordon & Betty Moore Foundation and California Health Care Foundation, the VA Quality Scholars program, and the John A. Hartford Foundation/Atlantic Philanthropies Claire M. Fagin Fellows program.

The authors received the following financial support for the research, authorship, and/or publication of this article: The Gordon & Betty Moore Foundation and the California Health Care Foundation provided funding for this project; the VA Quality Scholars program provided postdoctoral training support for Drs. Stephens, Schopfer, Schmajuk, and Moy; the John A. Hartford Foundation/Atlantic Philanthropies Claire M. Fagin Fellows program provided postdoctoral training support for Dr Stephens.

Author Disclosure Statement

Drs. Stephens, Pierce, Schopfer, Schmajuk, Moy, Bachhuber, Wallahgen, and Lee, and Mr Sackett declared no conflicts of interest with respect to the research, authorship, and/or publication of this article.