Residents' Exposure to Educational Experiences in Facilitating Hospital Discharges

Nearly 1 of 5 patients discharged from the hospital will experience an adverse event. Many will result in avoidable morbidity and added health care costs.^1,2  Significant public attention is focused on improving transitional care.³  In hospital-based settings, resident physicians play an important role within the complex interdisciplinary discharge process.⁴  This has led to a national consensus that successfully orchestrating care transitions should be reflected in the core competencies of specialties involved in inpatient care.^5–7 

A diverse collection of curricula has been published focusing on transitions of care,⁸  yet the degree to which they are implemented is not well understood. In a 2009 survey of IM residency program directors,⁹  only 16% reported having formal discharge planning curricula. This may understate resident experiences, as many educational modalities happen on an individual level (eg, discharge summary review, on-rounds teaching) and may not be catalogued by programs. Thus, it is important to understand learners' perceptions of educational experiences to create a full picture of how this competency is being addressed.

The education community also has an incomplete understanding of the efficacy of different educational approaches to care transitions. Curricula in this area are difficult to compare because they are content specific and their impact on behaviors is complex and difficult to measure. One theory argues that a person's willingness to adopt a specific behavior is dependent on his or her attitude toward the behavior.¹⁰  Thus, experiences in transitioning care may teach different skills, but ultimately these experiences need to influence participants' fundamental perceptions of responsibility for this aspect of care to affect practice.

We sought to determine the prevalence of specific educational interventions being used to teach safe hospital discharge practices within a sample of US internal medicine (IM) residency programs as well as assess correlations with perceived responsibility for ensuring safe hospital discharges.

We conducted a cross-sectional, 24-question survey (provided as online supplemental material) of IM residents at 9 US university- and community-based residency programs. We used purposive sampling and professional relationships to identify potential sites. We explicitly sought to identify programs with diverse program sizes and geographic locations. Program sizes ranged from 38 to 170 resident physicians (mean = 91), and programs were balanced across geographic regions within the United States (table 1).

There is no existing instrument with validity evidence for measuring responsibility for patients transitioning from inpatient to ambulatory care. Thus, we developed a new instrument that approached this construct through 2 methods. We first utilized a single question derived from prior work in this area to assess perceived duration of postdischarge responsibility (“How many days are inpatient providers responsible for their patients after they are discharged?”).¹¹  Existing consensus guidelines do not establish a standard for inpatient providers' duration of responsibility after hospital discharge.³  We then developed specific questions to assess residents' perceived responsibility within 4 key domains in posthospital transitions of care identified from existing literature: (1) medication reconciliation¹; (2) lab follow-up¹²; (3) follow-up appointments¹³ ; and (4) communication with ambulatory providers.¹⁴  Within each domain, we used 1 question to assess resident perceptions of responsibility for discharge practices that are considered the standard of care based on policy statements from professional societies.^3,15  A second question in each domain assessed resident perceptions of behaviors considered above and beyond the standard of care. Responses to the 8 questions were ranked on a 4-point Likert scale (4, strongly agree, to 1, strongly disagree), and items were summed to create a cumulative score for perceived responsibility for transitional care tasks.

Residents also were asked to identify the educational experiences they had completed: lectures, small group discussions, postdischarge home visits, postdischarge telephone calls, postdischarge clinic visits, on-rounds teaching, formalized discharge summary review and feedback, and online modules. These interventions were selected based on expert opinion and review of existing literature.⁸  Residents were asked to rank the impact of these approaches on their motivation to ensure safe transitions of care using a 1 to 10 scale (1, no effect, to 10, significantly impacted my motivation). Questions were refined in 2 sequential, heterogeneous focus groups of 6 to 8 attending and resident physicians, with “think aloud” cognitive interviewing to ensure consistency in question interpretation.

The survey was administered between September 2014 and February 2015 at participating programs using SurveyMonkey (SurveyMonkey Inc, San Mateo, CA) or in paper form. This was followed by 3 reminder e-mails at all sites except 1, where 2 reminders were sent based on Institutional Review Board restrictions.

The study was approved by the Institutional Review Board at each participating site.

Data were analyzed using SAS version 9.3 (SAS Institute Inc, Cary, NC). Survey responses were analyzed using descriptive statistics. Between-group comparisons were made using chi-square tests, and correlations between variables were analyzed using Kendall tau correlation coefficient.

Of 817 IM residents invited to participate, 469 (57%) completed the survey, with response rates for the various sites ranging from 45% to 76%. Respondents were balanced across postgraduate years, and the majority of respondents (61%, 254 of 420) indicated an intention to pursue a subspecialty career (table 1).

On-rounds teaching (74%, 327 of 439) and lectures (52%, 226 of 439) were the most common educational methods experienced by residents pertaining to posthospital care transitions (figure 1). Over a quarter of residents (32%, 142 of 439) had taken part in a postdischarge clinic visit, while other forms of postdischarge patient contact, such as telephone calls or home visits, were less common (12% [53 of 439] and 4% [18 of 439], respectively). Less than a quarter of respondents (23%, 99 of 439) had received formal review of and feedback on a discharge summary (PGY-1: 18%, 22 of 124; PGY-2: 25%, 31 of 126; and PGY-3: 24%, 31 of 131).

Residents rated the impact of the educational experiences on their motivation to ensure safe hospital discharges between 3.62 and 7.61 on a 1 to 10 scale (mean = 6.5; figure 2). Online modules and lectures were ranked lower than all other experiences (3.62 and 5.95, respectively). Residents rated postdischarge home visits, discharge summary feedback, and postdischarge clinic visits the highest (7.13, 7.38, and 7.61, respectively). None of the educational experiences were correlated with perceived duration of responsibility (table 2). However, the 3 educational experiences involving contact with patients after discharge (clinic visits, telephone calls, and home visits) had small but statistically significant correlations with cumulative perceived responsibility for transitional care tasks as measured by the sum of responses to the 8 domain-specific questions (correlation coefficient 0.12 [P = .004], 0.1 [P = .012], and 0.13 [P = .001], respectively).

In this multisite survey of IM residents, direct patient care was the most commonly reported means through which residents learned to successfully facilitate hospital discharges. Despite being a cornerstone of patient care, less than a quarter of residents indicated that they had received formal feedback on their discharge summaries. Furthermore, opportunities for residents to engage with patients across the postdischarge care continuum were uncommon, despite being rated highly by residents and having the strongest correlation with their perceived responsibility for ensuring safe transitions of care.

These results have several implications for medical educators. This work expands on prior qualitative data¹⁶  suggesting that residents learn transitions of care skills through clinical practice more commonly than didactic experiences. While experiential grounding is a vital component of adult learning theory,¹⁷  ad hoc work-based education allows for role modeling of heterogeneous practice patterns. This is of concern, given variable attitudes toward transitions of care on the part of attending physicians,¹¹  who also often fall short of the ideal behaviors outlined in consensus guidelines.^18–20  Research also suggests that higher patient census compromises the quality of attending physician care²¹  and that time constraints preclude high-quality discharge practices.^18,22,23  Collectively, this suggests that ad hoc on-rounds teaching may deteriorate as workload rises.

Overcoming the inherent heterogeneity in clinical role modeling requires that attending physicians assess residents' transitional care competence using standardized metrics. Discharge summaries are an ideal starting place, as the literature provides several examples of how to teach and evaluate residents' discharge summaries.^24–26  Similar to other studies,^16,27  our results suggest that formal discharge summary review is underutilized, despite evidence of its efficacy. Conducting discharge summary reviews is an excellent starting point for programs seeking to build transitions of care curricula rooted in competency-based education.

While it is an important start, teaching specific skills alone (eg, discharge summaries) may not transform the practice of busy residents. The theory of planned behavior argues that a person's willingness to adopt a behavior depends on his or her attitude toward that behavior, along with his or her perceived self-efficacy and external pressures.¹⁰  In other words, residents require more than skills and external pressure to improve transitional care practices. To truly achieve this competency, they must adopt an attitude of responsibility for ensuring safe transitions of care. Prior work²²  indicates that in the absence of downstream follow-up it is easy for residents to abdicate responsibility for patients in the postdischarge period.

Our study found that only interventions that connected residents with patients after discharge had a significant correlation with their perceived responsibility. The literature contains examples of curricula that connect learners with patients after discharge through telephone calls, clinic visits, or home visits,^28–30  yet our results indicate that these experiences are not widely used in IM programs. As residency programs seek to build their transitional care curricula, our results suggest that some aspect of patient follow-up may maximize the impact on residents' perceived responsibility for patients across the care continuum.

Our study has several limitations, including the relatively small number of sites that were selected using purposive sampling. It is possible that using professional relationships in selecting sites might have led to selection of programs with faculty particularly interested in transitions of care, biasing the results toward overrepresenting the actual prevalence of transitions of care education experiences nationally. An added limitation includes the response rate of 57% and the risk of nonresponse bias. This study's focus on physician responsibility, as care transitions often rely on interprofessional collaboration, is also a limitation. Finally, the survey lacks established validity evidence, and dependence on resident recall of prior educational experiences introduces the possibility of recall bias.

Further work is needed to develop tools to teach and assess residents' transitional care competency in ways that are efficient and resource sensitive to facilitate broader uptake in graduate medical education.

Internal medicine residents learn to facilitate hospital discharges most commonly through direct patient care. Opportunities to interact with patients across the postdischarge continuum are uncommon, despite being rated highly by residents and correlating with increased perceived responsibility for ensuring safe transitions of care.