Participants and Recruitment
A randomised controlled trial (RCT) study was conducted among nurse leaders in a large general hospital in Beijing, China. The inclusion criteria were: 1) nurse leaders in clinical departments, 2) voluntary participation, and 3) no previous participation in Balint groups. We recruited 80 nurse leaders through advertising. Participants were randomly assigned to either a Balint group or non-Balint group, using opaque random envelopes by a researcher who implements random assignment. The participants were from different specialties, including internal medicine (n=36), surgery (n=30), gynaecology and obstetrics (n=4), paediatrics (n=4), emergency medicine (n=4), ophthalmology (n=1), radiology (n=2), and anaesthesiology (n=2). The nurse leaders were informed about the specific processes of the project and the potential risks and benefits. The sample size was established based on an overall difference in the outcome measure of The Maslach Burnout Inventory(MBI)scores (emotional exhaustion [EE], depersonalization [DP], and personal accomplishment [PA]) between participants in the control and intervention groups, where a sample size of 60 was sufficient to detect a difference in MBI scores [10]. Assuming a dropout rate of 15% throughout the study, the required sample size was determined as 71, with 35 participants in each group.
Before the study began, informed consent was obtained from all participants.
The study was approved by the Peking University People's Hospital Ethics Board (Registration Number 2020PHB151). The study was registered in ClinicalTrials.gov Protocol Registration(ClinicalTrials.gov Identifier: NCT05716828,16/01/2023).
Measures
Demographic variables
All participants completed a demographic questionnaire that included age, gender, clinical department, marital status, and length of medical service in mean years.
Burnout
Burnout was measured using the Chinese version of the Maslach Burnout Inventory-Human Services Survey (MBI-HSS) [11, 12]. This questionnaire includes 22 items across three domains: emotional exhaustion (EE; 9 items), depersonalization (DP; 5 items), and personal accomplishment (PA; 8 items). The scoring range for each item is 0 (never felt) to 6 (felt every day). The EE subscale evaluates the feeling of excessive emotional stress and exhaustion due to work, which is characterized by mental, emotional, and physical exhaustion. The DP subscale measures unsympathetic and impersonal response to patients, and is regarded as a form of alienation. The PA subscale assesses work-related ability and sense of accomplishment [11]. The score for each subscale is calculated separately and not combined into a single total score. In a previous study, the Cronbach alpha coefficients for the EE, DP, and PA subscales were 0.89, 0.79, and 0.87, respectively [10]. The Cronbach alpha coefficients for the EE, DP, and PA subscales in this study were 0.742, 0.803, and 0.862, respectively.
Self-efficacy
The General Self-Efficacy Scale (GSES) is a 10-item scale used to measure individuals’ overall self-confidence in dealing with different environmental challenges or unprecedented situations. The scale’s mean score is 2.86 [13]. Participants are asked to rate 10 questions by choosing one of four response options: 1 (completely incorrect), 2 (almost incorrect), 3 (relatively correct), or 4 (completely correct). A higher score indicates a higher level of general self-efficacy. The GSES is widely used in China, and the Chinese version has good reliability and validity. The internal consistency coefficient for the scale was 0.862 in previous research [6, 14, 15]. The internal consistency coefficient in this study was 0.756.
Group Climate Questionnaire
The Group Climate Questionnaire-Short Form (GCQ-S) assessed the group climate, using a three-dimensional construct comprising engagement, avoidance, and conflict [16]. The GCQ-S engagement measures the team's self-disclosure and work orientation. The GCQ-S avoidance examines the extent to which individuals depend on other team members or leaders to create and manage team interactions while avoiding taking responsibility for their change process. The GCQ-S conflict measures hostility, including anger, distrust, and rejection. Among these three factors, engagement is most closely related to the processes and results of other groups and individuals, followed by conflict [17].
Procedure
The intervention program began in July 2020 and ended in January 2022. Participants in the Balint group completed Balint training for a period of at least three months, which included two lectures and 10 small group discussion sessions held once a month for one hour at a time (on Thursdays from 12 to 1 PM). The 40 participants in the control group were placed on the waitlist for future Balint group sessions but did not receive any intervention in the meantime (We plan to carry out Balint team interventions for participants on waitlist in the next year; in the subsequent trial, we will also use the data from the waitlisted control group). Participants in both groups completed the MBI-HSS and GSES questionnaires, at the beginning and end of the intervention period. The Balint group also completed the GCQ-S. The GCQ-S was administered twice; the first measurement was after the first Balint group activity, and the second after the 10th Balint group activity.
The study’s purpose was explained to the participants, and they were informed that participation was voluntary. Figure 1 shows the consort flow diagram.
Intervention
Balint group intervention is a standardized training model, which is introduced in Germany [18]. Each group is a closed group including one to two group leaders and six to twelve participants, with sessions lasting 60 to 90 minutes. At the beginning of the first group session, leaders explain the basic norms and expectations for how members relate to each other during a Balint group activity; the members are asked to commit to respect these, to ensure there is a safe talking space for all group members and the leaders, allowing the group to work openly and effectively. The main expectations include: maintaining confidentiality, being respectful and non-judgmental, speaking on one’s own behalf, providing an opportunity for all members to speak, avoiding offering advice or solutions, and reaching an agreement about time, place, membership, and how to work together. This is called maintaining the frame [19].
Before each meeting, all participants were invited to prepare a challenging nurse-patient encounter case. At the beginning of the meeting, one presenter briefly described the case, and the others decided whether to choose it as the topic for that day. During the meeting, the presenter had to: (1) spend 5 to 10 minutes describing a nurse-patient relationship that is disturbing, frustrating, confusing, or uneasy; (2) answer short factual questions from other members; (3) step back from their chair, listen to, and reflect on the discussions of other group members; (4) focus on their own feelings when listening; and (5) return to the group and analyse their reflections. Other team members needed to: (1) explore the doctor-patient relationship in the given case; (2) share what would happen to them if they were that doctor or patient; (3) introspect themselves and use their imagination to explore the unacknowledged parts of the case; (4) pay attention to the differences among team members; and (5) generate new opinions and ideas on the case. The team leader needed to: (1) establish a clear agreement to create and maintain a safe group environment; (2) maintain the framework; (3) protect the care provider and other members of the group from falling into judgment and irrelevant discussion; (4) provide reasonable and timely intervention to encourage reflection, introspection, empathy, and open communication; and (5) participate in the development of the group to ensure that it follows the Balint task to explore the doctor-patient relationship [18].
The discussion emphasized the participants’ emotions and attitudes; medical technology was avoided, and they were not required to provide specific ways to solve problems. Participants were asked to consider their reactions, or the emotions and thoughts of a particular nurse. They were expected to consider the nurse-patient relationship from the nurse’s and patient’s perspectives [20]. Figure 2 shows the typical process of the Balint group session.
Data analysis
IBM SPSS Statistics version 26 (IBM, Armonk, NY, and the US) and SAS 9.4 (SAS Institute, Cary, NC) were used to conduct all analyses. Categorical data utilized a chi-square or Fisher's exact tests. After the end of the Balint group activity, we used an analysis of variance for repeated measures to compare the differences between the Balint and control groups, as we expected highly correlated values. We also calculated the effect sizes (Cohen's d). According to the criteria proposed by Cohen, a value of 0.2 denotes a small effect, 0.5a moderate effect, and 0.8 a large effect.