Basel Discharge Communication Project

RATIONALE However essential the transferred information at discharge might be, it can only have a positive effect on the patient if he or she recalls its content after discharge. Little is known about patients' ability to subsequently recall instructions received during hospital discharge (Sanderson, Thompson, Brown, Tucker, \& Bittner, 2009). Using telephone interviews to gauge the ability to recall discharge instructions, Sanderson et al. (2009) found that many patients were unable to even name their diagnosis or list risk factors as contributing causes. Examining elderly patients' comprehension of discharge instructions, a further study found that 21% did not understand their diagnosis, and 56% failed to comprehend their return instructions (Hastings et al., 2011). A study of Chau et al. (2011) showed that even immune-compromised patient's knowledge of oral drugs at discharge was merely moderate. Finally, Isaacman et al. (1992) observed that less than half of the important discharge information, including medication details and indicators of worsening of the patient's clinical status, was recalled during an exit interview. These few available findings suggest that there is ample room for discharge communication to be optimized and for finding and testing techniques to do so.

As psychological theory and associated empirical findings suggest that information structuring can be a powerful tool in improving memory recall and understanding, the question has been raised whether structuring the information conveyed during the discharge communication could also improve patients' memory of the respective content. Few researchers (Doak, Doak, Friedell, \& Meade, 1998; Ley, 1979) have suggested that structuring communications in a clinical setting could improve patients' recall. They argue that structured information would be easier to recall than nonstructured information; however, these authors did not provide strong evidence to support this hypothesis. Furthermore, structured approaches and tools may provide support for physicians in increasing communication competences and potential solutions to improve the quality of communication and prevent subsequent patient harm. To date, only a few attempts have been made to investigate whether conveying information in a structured way results in better outcomes, measured in terms of learning and recall. The power of information structuring and associated chunking mechanisms has primarily been studied in the laboratory; no previous studies have investigated its role in improving discharge information delivery. Could information structuring also improve patients' recall and understanding of discharge information? If so, how should physicians best structure information at discharge to achieve these goals? How do possible effects on patients' recall translate to better adherence to recommendations? Studying these questions experimentally in the Emergency Department (ED) would be demanding and potentially stressful for ED patients. In a first step, we therefore decided to take advantage of students as proxy patients. Various previous studies have used proxies (mostly health care professionals or family caregivers) to evaluate certain patient outcomes (such as health-related quality of life (Pickard \& Knight, 2005), functional ability (Loewenstein et al., 2001), or symptoms(Nekolaichuk et al., 1999)). To our knowledge no previous study has used students as proxies to gauge the recall performance of patients.

Prior knowledge facilitates the processing of new incoming information by providing a structure into which new information can be integrated (Brod, Werkle-Bergner, \& Shing, 2013); therefore, structured incoming information should not be recalled better than nonstructured information by individuals who can build upon prior knowledge. If there is, by contrast, no internal structure because of absent prior knowledge, externally imposed structure could yield similar benefits. Thus, the extent to which the possible superiority of a structured presentation of ED discharge information over a nonstructured presentation relates to relevant prior medical knowledge was the secondary goal of this study. Or, in other words, could the availability of relevant prior knowledge enable the receiver of information to store it efficiently, even when its presentation lacks structure? To answer this question, we recruited, besides the first-year psychology students, two additional independent populations, namely first-year medical students and third-year medical students. Specifically, the three participant groups differed notably in their knowledge of cardiac pain (first-year psychology students \< first-year medical students \< third-year medical students).

TRIAL OBJECTIBVES Primary objective First, assuming that externally imposed structure yields benefits in terms of memory performance if internal structure is unavailable, the goal of this study is to investigate whether first-year psychology students, i.e. students with little to no prior medical knowledge, who serve as surrogate patients, recall more information when it is presented in a structured way as compared to a nonstructured presentation. Second, assuming that structure should benefit mostly those individuals who cannot make use of previous knowledge to build memory chunks and to better control the experimental setup between the two parameters "structure" and "prior medical knowledge", we set out to oppose various degrees of relevant prior medical knowledge to structured and nonstructured content presentation, respectively.

Secondary objectives

1. To compare differences between the structured and nonstructured conditions in terms of participants' ratings of the comprehensibility of the physician, the structure of the dialogue, and their willingness to recommend the physician to friends and relatives.
2. To establish whether the effect of information structuring on the number of recalled items is independent from the influence from the students' current mood and level of attention.

TRIAL ENDPOINTS Primary endpoint Number of items recalled, separately for the structured and the nonstructured condition as well as for the three participating groups Secondary endpoints Visual analog scale (VAS) measures of participants' current mood, level of attention (to test for moderating effects of the students' status), and ratings of the comprehensibility of the physician, the structure of the dialogue, willingness to recommend the physician to friends and relatives.

TRIAL DESIGN AND METHODS Trial design Prospective cross-sectional multicenter trial Participating sites University of Mannheim, Germany University of Basel, Switzerland Study schedule Participants We will recruit students during regular lectures at the University of Basel, Switzerland, and Mannheim, Germany. Psychology students will receive course credit for participation. Participants being younger than 18 years will not be included because of limited ability to give informed consent.

Study Procedure The study will be conducted in each two quiet auditoriums of the Universities of Basel (for first-year psychology students and third year medical students), and Mannheim (for first year medical students). We will randomly divide participants into the structured and the nonstructured condition, respectively. At the outset, after giving informed consent, participants of both conditions will be instructed that they are about to watch a video of a physician-patient interaction, requesting them to take the perspective of the patient. After that, a video displaying an experienced emergency physician who orally presents structured content to a patient will be shown to one group. A second video displaying the same emergency physician, presenting the same content, but in a nonstructured manner, will be shown to the other group. We will then confront the students with a blank piece of paper, and we will ask them to mark everything occurring to them concerning the communication (immediate recall). Informed students will observe their peers in order to inhibit taking notes. After that, students of both conditions will be confronted with a multiple choice test that was constructed in order to measure students' medical knowledge, and five VASs in order to assess the secondary endpoints. Information about demographic variables will be collected in conclusion.

Data Analysis Primary endpoint is the recall performance of the participants expressed as the number of items remembered from the discharge communication. In a primary, un-adjusted analysis, the difference in recall performance between the two groups receiving structured and nonstructured discharge information, and the three groups with different levels of medical knowledge, as well as their interaction, will be analyzed using an analysis of variance (ANOVA). In addition, an analysis of covariance (ANCOVA), which adjusts for the two VAS measures concerning students' status, will be performed. Non-parametric Mann-Whitney tests will be performed to probe for inter- and intragroup differences in medical knowledge of first-year psychology students, first-year medical students, and third-year medical students. A t-test analysis will be used to compare differences between the structured and non-structured conditions in terms of participants' VAS ratings of the comprehensibility of the physician, the structure of the dialogue, and willingness to recommend the physician to friends and relatives. All tests will be performed using a significance level α = 0.05.