BACKGROUND

Year-end resident clinic handoffs occur when patients transfer resident primary care providers (PCP) at the time of resident graduation. Annually, year-end resident clinic handoffs impact approximately one million patients.1 Recently, it has been demonstrated that resident clinic handoffs put patients at risk and lead to discontinuity of care.2–4 Despite Accreditation Council of Graduate Medical Education (ACGME) requirements for resident competency in handoffs, clinic handoffs remain an unaddressed patient safety issue in resident education, and there is little evidence for effective solutions.1,5

The risks of clinic handoffs for patients are numerous. After a clinic handoff, patients often do not have a follow-up appointment, are lost to follow-up, and test results are missed.3,4 These handoffs also create many missed opportunities to provide preventative health and routine management for chronic conditions.3 Even when patients receive appointments, they often miss visits and are ultimately lost to follow-up.4 They may also experience more acute visits in the emergency room or hospital as a consequence of delayed care.4 Although poor outcomes attributable to clinic handoffs have been demonstrated, there are few interventions demonstrating improvement in clinic handoff outcomes. One multifaceted handoff intervention in a psychiatry residency clinic demonstrated improved timeliness of follow-up and patient outcomes.6,7 Two interventions in internal medicine clinics increased the number of handoffs completed and improved the number of clinical tasks that were followed up after the transition.8,9

To date, there has been little examination of the patient perspective of year-end clinic handoffs. One earlier study identified that less than half of patients were fully satisfied with their transfer in an Internal Medicine (IM) resident clinic.10 The main predictor for increased patient satisfaction in that study was personal notification of the transition by the departing resident. Other predictors included whether or not the patient felt the departing resident had done everything possible to facilitate the transfer, whether there was opportunity to discuss the transfer with the departing resident, and the patient’s overall perception of the medical center. A subsequent intervention consisting of educating departing residents on how to approach the transfer with patients and sending handoff notification letters to patients increased patient satisfaction.11 These studies were based upon patient questionnaires and did not elicit patient experiences and perceptions of the handoff process.

Several agencies including the Institute of Medicine (IOM) and the Agency for Healthcare Research and Quality (AHRQ) advocate for redesigning processes of care to focus on delivering patient-centered care to improve quality.12,13 Thus, understanding the patients’ perspectives is necessary to improve clinic handoffs and design patient-centered care transitions. Eliciting the patient perspective is especially critical to improving clinic handoffs, since certain patient factors may be associated with poor outcomes, such as understanding why patients miss visits during the transition. Additionally, the risks of these handoffs may be underestimated, since prior studies show that patients are often able to report adverse events that would be missed by chart review.14,15 Furthermore, patients may be aware of the quality of inter-physician communication during care transitions and may give additional insights into methods to improve communication during clinic handoffs.16

Therefore, to improve clinic handoffs, more insight into patient experiences and patient needs during this transition are crucial. Our aim was to examine patient perspectives and satisfaction with resident clinic handoffs using semi-structured interviews. In particular, we strove to identify patients’ perceptions of positive and negative experiences during clinic handoffs. Ultimately, these findings can be used to design patient-centered handoff processes.

METHODS

Setting and Study Design

This study occurred at a single, academic IM resident continuity clinic. Approximately 30 IM residents per class have clinic at this site, spending half-days in clinic for the duration of their residency supervised by faculty preceptors consistent with ACGME regulations.5 Patients were recruited for this study from October 2011 to January 2012 after being identified on a sign-out by graduating residents during a year-end clinic handoff in June 2011. This study was approved by the University of Chicago Institutional Review Board.

Clinic Handoff Process

In June 2011, as part of a multifaceted clinic handoff protocol, departing residents were asked to list patients they believed were “high-risk” during the handoff on a sign-out worksheet. Residents had previously received education on the risks of clinic handoffs and guidance on how to select high-risk patients. Suggestions for high-risk patients included complex patients with multiple comorbidities, nonadherent patients, patient who frequently miss visits, patients with frequent hospitalizations, patients with psychiatric diseases or challenging social situations, and patients undergoing active work-up.4,17 During a scheduled conference, departing residents discussed patients on their sign-out with the post-graduate year (PGY) 2 resident assuming care for their patients after the handoff. Patients are handed off to PGY2 residents instead of interns, since interns would be unable to receive an in-person sign-out. Patients received letters notifying them of the transition in May, and many patients were also notified in person by the departing physician. When possible, patients received appointments with their new physician at the time of their last visit with their departing physician or were placed on a wait list. High-risk patients were scheduled to be seen with priority as soon as possible after the handoff.

Data Collection

Sign-outs were collected to record the names and medical record numbers of high-risk patients identified by residents. Departing residents and residents assuming care gave email or written consent for their patients to be contacted to participate in interviews.

A patient interview script was developed to elicit positive and negative experiences with the clinic handoff process (Appendix 1, available online). The script was designed for easy readability and reviewed with a patient champion to ensure adequate understanding. The interview questions were semi-structured. (For example, “Did anything bad or inconvenient happen following your transition to the new doctor?”). Probing questions were added to the script and improvised by the interviewer (KB) to prompt patient comments. Patient survey questions were also added to assess satisfaction with various aspects of the clinic handoff and quantify patients’ perceptions of the quality of the clinic handoff.18,19 Patients were asked to rate their level of agreement on a 5-point Likert-type scale from 1 = strongly disagree to 5 = strongly agree for specific statements about the clinic handoff. Patients were also asked to name their old and new PCP and whether they were notified about the transition.

High-risk patients were contacted by a trained research assistant (KB) by telephone and invited to participate in interviews during the post-handoff period (October 2011–January 2012). Consent was obtained to record the conversation prior to beginning the interview. Interviews were recorded and transcribed using digital telephone audio recording.

Data Analysis

Two physician investigators (AP& WL) and one research assistant (RA) (KB) each independently coded 21 (20 %) of the interview transcripts to establish a coding scheme, and subsequently met to discuss discrepancies and to refine the coding scheme. Discrepancies were resolved by consensus. After the coding scheme was finalized, an additional ten (10 %) transcripts were coded independently each by two physician investigators (AP& WL) and one trained RA (KB) to further test and refine the categories. Another author (AP) reviewed all coded quotations to ensure correct assignment. Qualitative analysis was completed using Atlas.ti 5.2 (Berlin) software program.20

Patient responses to Likert-type items were examined and data were dichotomized for analysis (“agree” was defined as the combined sum of agree- and strongly agree). Whether patients could correctly name their new PCP, whether or not they had recalled receiving notification of the handoff, and their satisfaction ratings were analyzed using STATA.21 Chi square, Fisher’s Exact, and t-tests were utilized, as appropriate, to test associations between these variables.

RESULTS

There were 26 departing residents who listed 323 high-risk patients on their sign-outs. After contacting all (323) patients, 95 % (307/323) were deemed eligible. Patients who could not consent, were deceased, changed insurance providers or were no longer patients of the clinic were excluded (Fig. 1). After attempting to contact all eligible patients twice, 103 patient interviews were completed. The mean age of participants was 67 (range 31–91), 71 % were women and 88 % were African American. On average, non-respondents were younger; their mean age was 63 (range 24–90, p = 0.01), but had similar proportions of women (66 %, p = 0.41) and African American patients (85 %, p = 0.46). Inter-rater reliability for coding was assessed using a three-way kappa statistic, which was 0.6.

Figure 1.
figure 1

Patient enrollment for the Engineering Patient Oriented Clinic Handoffs (EPOCH) project.

Full size image

Patients’ Negative and Positive Experiences

Patients identified 28 negative experiences with clinic handoffs that were categorized into four overarching themes: (1) doctor-patient relationships (i.e. difficulty building rapport); (2) clinic logistics (i.e. difficulty rescheduling appointments); (3) process of the care transition (i.e. patient unaware that transition had occurred); and (4) patient safety-related issues (i.e. missed tests) (Table 1). The most common experiences by patients were resignation with the handoffs (52), difficulty building relationship or rapport with the new physician (50), inability to recall their new physician’s name (36), and not being made aware of the handoff prior to its occurrence (35) (Appendix 2, available online).

Table 1 Categories of Patient Reported Experiences (Events) in Clinic Handoff Experience with Illustrative Quotations
Full size table

Patients most frequently reported interruption of the doctor–patient relationship as their experience of clinic handoffs (172). Patients reported difficulty with frequent physician turnover, doctor–patient communication and often felt anxious and uncomfortable with the new physician PCP (Table 1). Patients most frequently expressed tolerance and resignation with the system of clinic handoffs and experienced frustration with frequent turnover (Table 1).

Clinic logistics and the process of the care transition were the next most frequently occurring themes. Overall patient safety-related issues occurred as a theme in 88 interviews. Specifically, patients reported seeking acute care visits (ER or urgent care) due to delayed care, missed test results or running out of medications during the transition period (Table 1).

Patients identified several positive experiences during the clinic handoff process. There were 15 distinct positive categories identified by patients, which fit into the same four themes as the negative experiences (Table 2). The most common identified positive experiences were being made aware of the handoff before its occurrence (53), being able to build a relationship or rapport with the new doctor (45), being called by their new doctor before the first visit (44), and good doctor–patient communication (42). Patients valued being prepared by their doctors for the transition (37) and some found it helpful to prepare themselves for their first visit with their new doctor. They also reported that ‘personal sharing’ from the new PCP helped build rapport. Personal sharing occurred when patients learned personal information about their doctor, such as being made aware of an important personal event in their doctor’s life, i.e. marriage or a birth of a child (Appendix 2, available online). Also, patients who were aware of their role in the educational mission as educators of residents were more understanding of the process (Table 2).

Table 2 Categories of Patient Reported Positive Clinic Handoff Experiences and Solutions, with Representative Quotations
Full size table

Patient Satisfaction and Patient Reported Outcomes

During interviews, although 73 % (75/103) of patients could correctly name their old PCP, only 59 % (61/103) of patients could correctly name their new PCP (Table 3), even though the majority of patients (83 %, 86/103) reported they had seen their new PCP at the time of the interview. A few patients (15 %, 15/103) reported difficulty getting a visit after the handoff and 19 % (20/103) reported having missed a visit with their new PCP. About half of patients had tried to reach their PCP after the handoff (40 % 41/103), and of these, 29 % (12/41) reported having difficulty communicating with them. While satisfaction with most aspects of the clinic handoff was high (Table 4), only 63 % (62/99) of patients were satisfied with the process of changing doctors overall. Patients who had seen their new PCP at the time of their interview were more satisfied with the clinic handoff than those who had not yet seen their PCP (57/86 [66 %] vs. 6/17 [35 %], p = 0.027).

Table 3 Patient-reported Clinic Handoff Outcomes
Full size table
Table 4 The Proportion of Patients Who Answered Agree or Strongly Agree, or Satisfied or Very Satisfied, to Likert-Type Interview Statements
Full size table

Overall, 72 % (74/103) of patients recalled receiving a letter from the clinic notifying them of the handoff, and 35 % (36/103) reported having a telephone conversation with the new PCP prior to their first clinic visit. In total, 82 % (84/103) of patients were notified of the transition by either a letter or telephone call. Patients who recalled receiving a letter or being contacted by telephone were more likely to report they were aware that their doctor was changing when compared to those who did not (95 % [80/84] vs. 79 % [15/19], p = 0.02). Notification of the clinic transition by letter or telephone call with the new PCP was associated with a higher rate of patients correctly naming their new PCP (65 % [55/84] vs. 32 % [6/19], p = 0.007). This notification was also associated with more patients reporting their new doctor assumed care for them immediately (81 % [68/84] vs. 53 %[10/19], p = 0.009) and reporting satisfaction with communication between their old and new doctors (80 % [67/84] vs. 58 % [11/19], p = 0.04). There was no association between receiving a letter or a telephone visit (61 % [51/84] vs. 58 % [11/19], p = 0.82) and correctly naming their new PCP (64 % [39/61] vs. 55 % [23/42], p = 0.35) and overall satisfaction with the clinic handoff.

DISCUSSION

In this study, high-risk patients identified positive and negative experiences of clinic handoffs after switching to a new resident PCP. Lack of patient notification and preparation for the handoff are common problems, as is inability to recall the new PCP’s name. Patients expected good communication to occur between the old and new doctor, as well as for the new physician to be aware of their medical history at the time of the first visit. They also expressed a lack of trust with the new PCP and being unhappy with major changes at the first visit. Patients were also aware of patient safety-related issues during their handoff and scheduling constraints. Additionally, many patients could not name their new resident physician when asked. Furthermore, many patients felt resignation with frequent turnover, and yet were tolerant of the process. Patients also expressed problems with systems issues outside of resident education, such as trouble communicating with their physician, difficulty with transportation to clinic and difficulty scheduling appointments.

Despite the numerous negative experiences of clinic handoffs identified, for the majority of patients, satisfaction was high. Interestingly, patient satisfaction with individual components of the handoff process was higher than overall satisfaction with the clinic handoff. One possible reason is that even with a good handoff process, patients do not like switching PCPs. It is also possible that we did not ask about components of the handoff process with which patients are most dissatisfied that would have correlated with their overall satisfaction. Our patient satisfaction with the clinic handoff was higher than satisfaction ratings reported in previous studies. Other investigators reported findings similar to ours that satisfaction correlated with notification of the handoff by letter or telephone.10,11

The finding of patients’ resignation is also interesting. Although many patients identified problems and aspects of the clinic handoff that they disliked, they were tolerant of it. It is possible that this resignation is because patients like having a resident PCP because they receive more attention, have longer visits, and like being seen by young doctors. It is also possible that patients did not think they had a choice because they wanted to receive care at our tertiary clinic rather than at other clinics.

This study has implications for a patient-centered clinic handoff process. It is important to ensure that patients receive clear notification of the handoff well in advance either in person, by telephone or by letter. It is equally important to help patients identify and pronounce the name of their new physician. Patients should be made aware of how physicians prepare for the handoff including communication, since patients value knowing that physicians review their chart prior to the visit and communicate about their care. Training residents in patient-centered communication during the handoff and working on improving their doctor–patient communication overall will also be helpful. Acknowledging patients for their role in educating resident physicians may be helpful. In addition, telephone visits with their new physician prior to the first visit may help improve the transition. Lastly, ensuring patients are notified about test results during handoffs and have methods of getting medications refilled seamlessly throughout the transition time period will improve patient safety. Scheduling high-risk patients to be seen in clinic early after the handoff is crucial.

There were several limitations of our study. First, it was a single institution study possibly limiting the generalizability our findings. Second, non-response bias may have influenced results, as it was difficult to reach patients by telephone and we were thus unable to interview all patients in our sample. In addition, we solicited the views of “high-risk” patients who likely had more chronic conditions, nonadherence, missed visits and hospitalizations. This also highlights the difficulty of contacting this high-risk population by telephone. Lastly, we had already implemented a clinic handoff protocol and an educational intervention for resident physicians prior to this study, so we are likely underestimating the negative experiences of patients and overestimating patient satisfaction compared to clinics without handoff protocols in place.

In summary, patients face negative experiences during clinic handoffs. Patients frequently are not aware of the transition, cannot name their new physician, and are at risk of experiencing patient safety-related issues during the handoff time frame. Patients are anxious about seeing a new physician and have difficulty establishing rapport after experiencing frequent physician turnover. Good physician handoff communication, doctor–patient communication and resident preparation prior to the first visit may mitigate these effects. Redesign of clinic handoff processes to be more patient-centered is needed and should incorporate these findings.