Care Transitions: Using Narratives to Assess Continuity of Care Provided to Older Patients after Hospital Discharge.

BACKGROUND: A common scenario that may pose challenges to primary care providers is when an older patient has been discharged from hospital. The aim of this pilot project is to examine the experiences of patients' admission to hospital through to discharge back home, using analysis of patient narratives to inform the strengths and weaknesses of the process.
METHODS: For this qualitative study, we interviewed eight subjects from the Sheldon M. Chumir Central Teaching Clinic (CTC). Interviews were analyzed for recurring themes and phenomena. Two physicians and two resident learners employed at the CTC were recruited as a focus group to review the narrative transcripts.
RESULTS: Narratives generally demonstrated moderate satisfaction among interviewees with respect to their hospitalization and follow-up care in the community. However, the residual effects of their hospitalization surprised five patients, and five were uncertain about their post-discharge management plan.
CONCLUSION: Both secondary and primary care providers can improve on communicating the likely course of recovery and follow-up plans to patients at the time of hospital discharge. Our findings add to the growing body of research advocating for the implementation of quality improvement measures to standardize the discharge process.

INTRODUCTION

There is mounting concern within the medical community and general public regarding the oncoming “grey tsunami”,( the expected increase in the number of Canadians becoming seniors over the next few years. The influx of those aged 65 and older will place unprecedented demands on the Canadian health-care system. By acting as an entry point in to the health-care system and providing continuity of care, family physicians will play a critical role in caring for our aging population.

The management of a patient by a primary care physician, seemingly straightforward in a younger patient presenting with a single concern, can become daunting when treating someone of advanced age who has both medical and social complexity. Challenges include the heightened vulnerability of these patients to adverse outcomes, higher likelihood of cognitive impairment, limited availability of clinical resources and sufficient time to address their problems, as well as the added complication of involving caregivers in decision-making.( Concerns have been raised from the perspective of the older patient, as well. A qualitative study by Barnes et al.( of patients diagnosed with congestive heart failure found that communication issues with their primary care physician led to a poor understanding of the underlying disease process and prognosis.

The most common care transition encountered by older patients is the movement into and out of hospital. In 2013–2014, Canadian seniors aged 65 and older account for 40% of all hospitalizations.( These patients often arrive at their family physician’s office or clinic after discharge with new diagnoses, altered medications, results of laboratory investigations, and follow-up recommendations that need to be synthesized and acted upon by the family physician.

The discharge process from hospital back to the community is rarely seamless. Communication issues between the discharging and receiving physician in the community including delays, errors, and omission of important information can negatively impact the quality of post-hospital care.( One study found that up to 14.1% of older patients had at least one medication error on their discharge summary.( Canada’s aging population will lead to a rise in the prevalence of health conditions that will result in an increased consumption of health-care resources,( including more frequent hospitalizations of longer duration if we carry on with current practices. Ensuring safe transfers of care between hospital-and community-based providers will be a key requirement of a health-care system that serves older patients effectively.

There is increasing interest in the use of patient narratives to explore the consumer perspective of the health-care system. Traditionally, there has been a preference for standard question-and-answer interviews as a means of acquiring qualitative data. It has been argued that focusing the topic through questioning that is ordered and worded by the interviewer forces the interviewee to respond within an imposed structure. Emphasis is placed on what is deemed relevant by the interviewer. What the interviewee may think as most important may never be discussed.( This is partially rectified by use of patient narratives, which are spontaneously generated stories with very little prompting or direction from the interviewer. Narrative approaches within health-care settings have been used for both research and enhancing clinical care.( Patient narratives are particularly useful in identifying areas of quality improvement, as a story may discuss a problem in detail as well as provide clues to direct steps to better service and care.( To our knowledge there has been no previous use of patient narratives to explore the perceptions of older patients regarding the experience of being discharged home from hospital.

The primary aim of this pilot study was the examination of the hospital discharge experience from the patient perspective in the expectation that this would inform potential areas for process improvement. A secondary aim was to analyze the positive and negative aspects of the current primary care experience from the perspective of a patient recently discharged from hospital, as we felt this might uncover issues not evident when the interactions between patient and provider are more, for want of a better term, “routine”.

METHODS

This was a pilot study examining the feasibility of obtaining patient narratives about the experience of hospital to home transitions. Ethics approval was received from University of Calgary’s Conjoint Health Research Ethics Board. We obtained informed consent from all enrolled patients.

Sample

Subjects were recruited from the Sheldon M. Chumir Central Teaching Clinic (CTC) using the clinic electronic medical record (EMR) to identify potential subjects. The CTC is a family medicine teaching site affiliated with the University of Calgary that supports a team approach to patient care, with social work, a behavioural health consultant, licensed practice nurses, chronic disease nurses, pharmacist, and other allied health team members on site.

Potential subjects had to be 75 years of age or older, have at least three recorded visits to the clinic, and had been admitted to hospital overnight for any reason in the six months prior to being interviewed. Subjects had to be fluent in English and possess the capacity to both provide informed consent and a spontaneously self-generated verbal narrative. Those with dementia listed as a health concern on their EMR or flagged as having significant cognitive impairment by their primary care provider were excluded.

Potential subjects who met criteria were then approached by their primary care provider for permission to have one of the authors (CW) contact them regarding participation in the study. As this is a pilot study, a convenience sample of nine subjects was interviewed. This was determined by the capacity of the research team and the number of patients who met inclusion criteria.

Data Collection

From September to December 2014, interviews were conducted in subjects’ homes by a member of the research team (CW) according to the narrative interview technique described by Jovchelovitch and Bauer.( This four-phase approach to interviewing first involves introducing the subject to the topic of interest. Specifically, subjects were given a brief overview of the rationale behind the narrative study, and then asked to describe their experiences: “We’d like to know what things were like before, during, and immediately after your hospital stay, and how this transition affected you.” During the second phase and main portion of the interview, the subject narrated freely with no verbal interruption from the interviewer. This phase was audio-recorded. Once the subject completed their narrative, the interviewer proceeded to probe for more information regarding the events related. No opinions or attitudes were directly sought through questioning, per protocol. Finally, once this third phase was complete, the recording device was switched off. The interviewer then engaged in “small talk” during which the interviewee was asked about attitudes or opinions that might aid with the interpretation of the narrative.

Data Analysis

Immediately following completion of the interview, post-narrative impressions and notes were recorded by the interviewer. Narratives were transcribed verbatim with the removal of any identifying information. Members of the research team (CW and DH) read through the completed transcripts separately, identifying common themes and generating coding categories and phenomena,( using field notes and the original audio-recordings to help provide context (Box 1). The authors met to review their individual results and, through discussion, achieved consensus on the resulting list of categories and phenomena.

“So, then I ended up and said I’d go in for the operation… I had a lot of concern because they said how much pain, and how hard, and I’m on my own [fear regarding recovery]. And I thought, “How can I look after myself?” All of that bothered me. And then they said “take home care,” and that to me was my life-saver [home care, positive] [post-discharge assistance].

“And then my daughter … I came out of the hospital a couple days earlier than I was supposed to…or more. And I stayed with her out on the acreage [family support].

“The hospital was hot, and it was noisy, and the woman in the next bed to me had company all night long [negative hospital environment].”

Focus Group

Two staff (MD1 and MD2) and two resident learners (R1 and R2) working at the CTC were recruited to review the narrative transcripts themselves and then participate in a focus group. None of these physicians had any previous contact with the subjects interviewed. To guide the focus group’s discussion of each individual narrative, the members were prompted with four questions to consider (Box 2).( The focus group’s discussion was also recorded and transcribed.

What did you hear in the story?

Was there anything that surprised you?

Is there anything in the story that concerns you or makes you uncomfortable?

After reading this narrative, is there anything you would add/change with regards to management of this patient?

RESULTS

Initial search of the EMR generated 42 potential subjects based on identifying a hospital discharge summary being received over the previous six months. Manual review of these charts revealed 14 patients who met our inclusion criteria. Nine of these individuals agreed to participate in the study.

Subjects ranged in age from 76 to 94 years of age. There were six women and three men. Reasons for admission and length of stay varied (summarized in Table 1). One narrative was discarded as it became apparent during the interview that the subject was unable to recall and/or relate their experience without significant spousal assistance. The remaining eight interviews were transcribed for analysis. Spontaneous narratives lasted between 1:19 and 10:28 minutes. The complete interview durations ranged from 7:29 to 14:52 minutes. The spouses of subjects were present for two of the interviews, but their comments were eliminated from the final transcribed narratives.

TABLE 1.

Summary of hospitalizations of interviewed and analyzed subjects, as described in EMR discharge summaries

Subject	Reason for Admission	Length of Stay (days)
1	Total knee replacement	unavailable
2	Acute stroke	3
3	Pacemaker implantation	2
5	Heart failure	8
6	Cardiomyopathy	3
7	Acute exacerbation of COPD	9
8	Urosepsis	5
9	Heart Failure	11

Following analysis by CW and DH, six themes were identified in three or more of the narratives. These included an overall positive experience with their primary care provider, the unexpected nature of their hospitalization, uncertainty regarding discharge plans, feelings of shock or surprise with the residual effects of their illness and the slowness of their recovery, the necessary support of family members during these transitions, and displeasure with the hospital environment itself (Table 2).

TABLE 2.

Recurring themes found in narrative transcripts

Theme	# of Narratives	Example
Positive primary care experience	6	“We have nothing but praise for the clinic down there. It’s totally, totally wonderful.”
Admission unexpected	6	“…To us a stroke was right out of left field.”
Uncertainty of discharge plans	5	“I’m sure [the pharmacist] explained [the medications] to me, but… I remember nothing…”
Surprise at residual effects/rate of recovery	5	“Sometimes you just get impatient with what’s happening. Why hasn’t it fixed up, why isn’t it repaired?”
Family support	6	“[My daughter] takes me for groceries… she helps me lots.”
Uncomfortable hospital environment	3	“The hospital was hot, and it was noisy…. if I hadn’t come out of there, I’d likely run out!”

All narratives made some comment regarding their primary care experience following their release from hospital. Both positive and negative themes were found in the eight narratives (Table 3). Six of the eight subjects mentioned having an overall positive experience with their family physician.

TABLE 3.

Themes from narrative analyses specifically related to primary care experience

Positive	Negative
Accessibility of clinic (geography, close location)	Lack of communication between patient and family physician
Accessibility (availability of appointments, ease of scheduling)	Patients’ uncertainty regarding final diagnosis, prognosis
Attentiveness of resident learners	Perceived delay in care
Timeliness of primary MD involvement post-discharge	Unilateral decision-making, perceived lack of patient input
Perceived thoroughness, comprehensive care	Family physician unaware of hospitalization/diagnosis
Continuity of care	Appointment tardiness, long wait times in clinic

These themes were reiterated in the analysis of the narratives provided by the focus group. Of particular note, in five of the narratives focus group members commented on their perception that the communication to patients about their discharge plans was inadequate. As MD1 explains:

“I’m sure everyone in acute care is going to say, ‘We did the discharge planning, we wrote down stuff.’ I’m sure they’ll say that. But this isn’t the first time that [patients] can’t remember [discharge instructions].”

R2 stressed the importance of ensuring patients have written documentation related to their illness and future management:

“I think the take-home message is get [discharge instructions] written down. Because [the patient] didn’t know all this; I’m guessing he took out the information and read it. So anything that’s written is super useful.”

The focus group reiterated the importance of ensuring adequate communication between hospital and primary care physicians at the time of discharge. One resident stated after reading a particular narrative that “it [was] unclear whether the [discharge] instructions were given to the GPs [general practitioners].” This observation led to members reflecting on previous experiences where the discharge information they received seemed inadequate or inaccurate:

I think when we do get discharge information … it’s less comprehensive than it used to be, if we get it at all.

And sometimes they say they’ve done things and they haven’t.

Themes around patient fears and concerns emerged in various narratives. The focus group conceded that patients might be unwilling to share this without prompting in a follow-up family doctor appointment. As noted by MD2 regarding Subject 3’s overt concerns regarding being home alone after hospitalization:

“…You’ve got this vulnerable little old person…. She’s got all these fears but obviously these fears aren’t something she discusses with her doctor. Maybe we should ask about those fears? I think we ask about those in generalities, but not specifics.”

Finally, the focus group commented on the crucial role that family played in the successful transition back home for a number of the subjects, and the potential strain this may place on caregivers. As MD2 voiced:

“What would you do if you didn’t have a daughter? Every person needs a daughter! …They’re coming from all over the country…. So the family is trying to make these big flights across the country, and obviously it’s quite stressful for them.”

DISCUSSION

Despite a large body of evidence emphasizing the need for improvements in care transitions from hospital,( the results of this pilot study suggest that there still exists a need for improvement in the discharge process. Coleman( described the elements of effective transitional care—dealing with the logistics of discharging patients back home, educating them and their family with respect to follow-up care, and coordinating the discharge with community health-care providers. Two recurring themes emerged from our analysis of the narratives: uncertainty around discharge plans; and surprise regarding the residual effects of the illness that led to admission coupled with the slower-than-expected recovery time. Both offer opportunities for improved patient education. This could be as simple as providing patients with written plain-language discharge instructions, or being conscious of the patient’s mental and emotional state when giving information pertaining to appointments or new medications. For instance, one subject recalled that a pharmacist visited her while she was hospitalized, but due to newly prescribed analgesics altering her mentation, she was unable to remember the content of their conversation.

As noted by the focus group, it is essential not only for patients to be adequately informed, their primary care physicians need crucial transition information, as well. A previous study showed that post-discharge follow-up with the physician who regularly provided treatment to a patient prior to their hospitalization significantly reduced the risk of urgent readmission to hospital.( Despite the importance of ensuring timely and informed follow-up by community-based providers, a systematic literature review completed in 2007 found that direct communication between hospital and primary care physicians was a relative rarity, occurring in only 3–20% of hospital discharges. Distribution of written discharge summaries was unduly delayed, with only 12–34% of discharge summaries available at the time of the first post-discharge visit.( Our subjects expressed frustration and disappointment when information regarding their hospital stay was not known by the attending family physicians. Not only does this lead to dissatisfaction, but it may impact health outcomes negatively, as well.

Both qualitative and quantitative research demonstrates the desire of patients for physicians who are eager to comprehend their experience and provide education and counselling based on their concerns.( Recognizing that a discharge from hospital is a time with a high potential for confusion, worry, and fear should prompt family physicians to explore their patients’ experiences more thoroughly. Concerns that might otherwise go unstated may then be addressed. We found that the narrative interviews about care transitions from hospital to home could be done within 10 and 15 minutes, making them something that could be potentially implemented within a primary care practice.

As a pilot study there are a number limitations to our report. First, our sample size was small and heterogeneous in terms of length of hospital stay, reason for admission (e.g., planned procedure versus acute decompensation of chronic disease), and previous hospitalization experience. While qualitative research strives for an adequate sample size that achieves data saturation,( the vast differences in our subjects’ experience and limited sample pool made this aim difficult. Subjects were exclusively recruited from an academic teaching clinic associated with the University of Calgary, where providers are funded by an alternate funding plan as opposed to a fee-for-service remuneration model. This allows both preceptor physicians and resident learners to spend more time with complicated patients than might be feasible for a community family physician. Selection bias may exist among our subject pool for a variety of reasons. A number of the patients approached following their hospitalization declined participation due to the fact that they were still feeling unwell. Though we aimed to exclude patients with cognitive impairment, potential subjects did not necessarily have objective cognitive test results (such as a Mini Mental Status Examination) on record in their EMRs. To protect patient privacy, once potential subjects were identified, the research team had to request individual primary care physicians to contact patients on our behalf, possibly leading to the exclusion of subjects based on the discretion of their physician. The presence of the subjects’ respective spouses for two of the analyzed narratives was difficult to control, and may have influenced the responses received. Finally, the narrative analysis process is subject to the bias of the individual researchers. We attempted to minimize this by placing each narrative through two separate analyses, as well has a final reading by our focus group. To assert that there is one correct interpretation of a story is disputable.( Many proponents of narrative medicine argue that the method’s appeal lays in the malleable interpretation of the patient experience.(

Our study highlights some the challenges facing older patients who have been hospitalized and discharged back home. Some of these issues may be mitigated through quality improvement strategies to facilitate appropriate and timely communication of discharge information to primary care physicians and patients themselves. We also present evidence that it might well be feasible and valuable to prompt older patients to voice their concerns and raise questions about their hospital experience.

CONCLUSION

Older patients of the Central Teaching Clinic recently discharged from hospital reported overall satisfaction with their primary care physicians. Negative aspects of hospital-to-home-care transitions stemmed from communication issues between hospital physicians, community-based providers, and patients themselves. Family physicians should take the opportunity to ask their patients about their hospital and discharge experience.