- Study design
The positive psychology intervention was established as a classroom-based elective course, which was embedded in the regular school curriculum and open to medical students at the School of Medicine, South China University of Technology (SCUT), Guangzhou, China. The course was open to year-2 and year -3 students only. The attendees who voluntarily consented to participate in the research were asked to complete the questionnaires before and one week after the training course, respectively. All questionnaires were anonymous, with the exception of academic year, gender and age, to avoid stigmatizing participants and to obtain honest answers as much as possible. No compensation was provided for participation. To test the effects of the intervention, an identical set of interventions and surveys was performed on an independent cohort of students who attended the course voluntarily in the year following the preliminary study. None of the participants reported any previous experience with positive psychology intervention.
A total of 61 undergraduate medical students in year-2 or year-3 of academic study attended the training course voluntarily in the preliminary study. Only participants who completed the entire course study, assignments, and pre- and post-course questionnaires were included in the analysis. According to these inclusion criteria, of the initial 61 participants recruited to the study, 12 dropped out of the program, giving an attrition rate of 19.7% (12/61). For these 12 students who discontinued the study, their common reasons were that they were too busy to finish the assignments or return the post-course questionnaire. For the 49 participants, their ages ranged from 17 to 22 years (mean=19.5, SD=0.94). For the validation cohort, 52 students participated in the course initially, and 46 of the participants who fulfilled the above inclusion criteria were included in the current analysis. Six of them dropped out of the program due to time limitations. The detailed socio-demographic characteristics of both participant cohorts are described in Table 1.
The students who participated in the course were first invited to complete a packet of questionnaires (as detailed in the following Measures section and Appendix 1-5) prior to the commencement of the training course. They then received the training and finished the related assignments for 8 continuous weeks (as detailed in the following Interventions section). Finally, in the week following the completion of the course, they were given an identical packet of questionnaires to complete. The pre- and post-course questionnaires were then compared and analyzed. At the same time, an independent survey question would be given in the post-course questionnaire by asking ‘‘Overall, what do you think of this project?”. Participants were asked to rate their response with a four-item Likert-type scale including “No useful at all”, “A little useful”, “Useful” and “Very useful”.
The intervention was set as a 1.5-hour class once a week and lasted for 8 continuous weeks. The protocol of the intervention, which is detailed in Appendix 6, was derived from Dr. Martin Seligman’s theory of PERMA (Positive emotion, Engagement, Relationship, Meaning, Accomplishment)  with slight modifications. Briefly, in each weekly class, a different topic related to the cultivation of PERMA was discussed, such as the cultivation of positive emotional states (e.g., gratitude and appreciation), the cultivation of intrinsic motivation through ‘‘flow”, and learning to be in harmony with bad moods by highlighting the meaning of life. Additionally, multiple topics related to medical professionals (e.g., doctor-patient relationships) were introduced and discussed in class to guide the students to find the thinking traps using positive psychology theory. Multiple out-of-class exercises were also assigned, such as writing down good things and identifying key character strengths. One single teacher led and completed the entire intervention, and the teacher had been trained to be qualified before the class at the Center for Positive Psychology and Engineering Psychology, School of Social Sciences at Tsinghua University, Beijing, China.
The following five scales were used to measure the psychological status of the participants before and after the training: the trait hope scale, life satisfaction scale, subjective happiness scale and depression and anxiety scale (Appendix 1-5). All these questionnaires were first translated from English into Chinese by one of the authors who was fluent in both Chinese and English. The translations were then checked by another author of the study to ensure the consistency with the original meaning of the scale items. No other adaptations to these scales were made.
5.1 The trait hope scale. The hope trait was measured by using the 12-item Trait Hope Scale[38, 39] (Appendix 1). This questionnaire asks respondents to rate their agreement with 12 statements related to hope on an 8-point Likert-type scale ranging from 1 (definitely false) to 8 (definitely true) (Cronbach’s α=0.85). A sample item of the scale is ‘‘I can think of many ways to get the things that are important to me’’. Among the scale items, 4 items (item 2, 9, 10 and 12) measure the goal-directed energy (also called agency thoughts), and 4 items (item 1, 4, 6 and 8 in the questionnaire) measure the plans to meet goals (also called pathway thoughts). The total trait hope scale score is derived by summing the four agency and the four pathway items. The possible range is 8-64, with higher scores reflecting higher levels of hope.
5.2 Life satisfaction scale. The life satisfaction was measured using the 5-item satisfaction scale [40, 41] (Appendix 2). The scale utilizes a 7-point Likert-type response scale ranging from 1 (strongly disagree) to 7 (strongly agree) (Cronbach’s α=0.86). A sample item of this scale is ‘‘In most ways, my life is close to ideal’’. The total score is derived by summing all 5 items. The possible range is 5-35, with a score of 20 representing a neutral point, 5-9 indicating extreme dissatisfaction with life and 31-35 indicating extreme satisfaction.
5.3 Subjective happiness scale. The subjective happiness scale is a 4-item scale of global subjective happiness[42, 43] (Appendix 3). Each item is rated on a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree) (Cronbach’s α=0.89). Amongst, two of the items ask respondents to characterize themselves using both absolute ratings and ratings relative to peers, whereas the other two items offer brief descriptions of happy and unhappy individuals and ask respondents the extent to which each characterization describes them. The total score is derived by summing the four items. The possible range of the total score is 4-28, with a score of 18 to 22 representing an average range. A higher score reflects a greater happiness.
5.4 Depression and anxiety scale. The symptom levels of depression and anxiety were measured using the patient-reported outcome measurement information system (PROMIS)[44, 45]. PROMIS, which is a set of online measure systems developed by the National Institute of Health (NIH) of the United States (US), evaluates multiple physical and mental conditions, including anger, depression, fatigue, anxiety, depression and physical function measures[44, 45]. Both the depression and the anxiety questionnaires are five-item Likert-type response scales to measure the frequency with which respondents have experienced over the past week. Both of them are universal symptom screening tools rather than disease-specific diagnostic tools. A sample item from the anxiety scale is ‘‘I felt worried in the past seven days’’. Participants were asked to rate their agreement with 5 answer choices including “Never”, “Rarely”, “Sometimes”, “Often” and “Always” (Cronbach’s α=0.87). PROMIS has two different but highly comparable scoring options: a short form and a computer adaptive test (CAT). In the short form option, participants are asked to answer an entire set of questions, while the CAT is a response-based scoring system in which participant’s response to the first item will guide the system’s choice of subsequent items, and the computer will calculate the sum score automatically. We adopted a combination of both options: the participants were asked to finish the online CAT survey, and the score from the CAT report was used for the current analysis. They were also asked to answer a paper-version short form survey (12 items included, Appendix 4-5), to enable us know their response to each specific item. For both depression and anxiety in the CAT, a score of 50 is the average for the general population. A higher score represents a greater level of the symptom being measured.
- Data analyses
The data were analyzed by using SPSS (version 18). Independent t-tests were used to determine whether there were any differences between the pre- and post-test scores on each outcome measure. Estimated means were used to describe the average pre-test and post-test scores on the outcome measures. Regression analysis was also performed to examine the relationships between the intervention and effect measures (Appendix 7). An alpha level of 0.05 was used to determine the statistical significance of all results.