Study Design and participants
This randomized controlled trial (RCT) was conducted in four localities in three districts (Chaoyang, Haidian and Fangshan) in Beijing, China. Staff from primary care centers, community committees, Civil Affairs and the China Disabled Persons' Federation in the four localities were referred to the study if they: (a) were aged 18–60; (2) had more than one year work experience; and (3) were literate. Participants were told prior to participating that the intervention was aimed at reducing mental illness stigma. They were randomized to either the “education only” (control) or “education and contact” (experimental) group in order to compare the effect of lectures about mental illness or lecture plus contact with patients with a clinical diagnosis of schizophrenia or bipolar disorder. The intervention took place between December 2018 and Mar 2019. Six sessions were carried out with 18–25 participants in each session. The study was approved by the Peking University Institutional Review Board (Year 2017,No. IRB00001052-16077).
Each site had a project coordinator who issued a notice about the programme on mental illness and stigma, and invited colleagues who were interested to sign up, both in the facilities and by notifying potential participants by social media. All potential participants were informed about the project and all participants signed a written, informed consent sheet.
Randomization And Sample Size
Randomization
Participants had an equal probability of being assigned to the two study groups.. Using a computer-generated randomization a list was compiled through simple randomization, participants were randomly assigned (ratio 1:1) to the two groups by an independent third party. To ensure the concealment of allocation, this third party used a protected computer database for the randomization list. Those in the “education only” group received the lecture alone without contact, while those in the “education and contact” group were invited also to stay for the contact intervention session that followed immediately after this.
Sample size
In a previous trial in China, the mean score of MICA decreased from 47.92 ± 8.63 to 43.53 ± 9.61 after mental health training[26]. Based on this difference, it was calculated that at least 55 participants were required in each group to achieve a significance level of 5% (two sided) and a power of 0.70.
Interventions
Control condition
The lecture was divided into three parts: (1) a brief introduction of the classification and treatment of mental disorders, as well as community mental health services in China; (2) details of the prevalence rates, main symptoms, illness course, principles of treatment, effect of treatment and extent of recovery of schizophrenia and bipolar disorders, as well as introducing historical famous people who suffered with these two disorders. (3) an introduction to mental illness related stigmatization. MH held the first two parts of the lecture while part 3 was led by WZ. The lecture took approximately 2 hours. Participant were encouraged to ask questions during the lecture.
Experimental condition
The experimental condition also included the lecture described above and in addition there was a session on patient recovery stories (the social contact element). For the “education and contact” group, this contact part was composed of hearing both in person recovery stories told directly by patients with a clinical diagnosis of schizophrenia or bipolar disorder, and watching recovery stories recorded in the form of a film. Selected recovered patients, on the basis of the “recovery story checklist” provided by CH, wrote their personal recovery narratives, including reference to their symptoms, treatment and recent experiences. The stories of two recovered patients were adapted and converted into two short films, each film of about 4 minutes. In each contact session, two or three recovered patients participated and shared their own experience of recovery. After that, the film which included the two recorded recovery stories were played (one of a person with a diagnosis of schizophrenia and one with bipolar disorder). This contact session lasted about 1 hour. Participants were encouraged to communicate with the patients. Full details of the experimental and control condition interventions are available from the corresponding author on request.
Measures
Mental health related knowledge
Mental health related knowledge was measured using the Mental Health Knowledge Schedule (MAKS), which consists of 12 items and each item was rated on a 5-point Likert scale which ranged from strongly disagree to strongly agree [27]. The overall test-retest reliability of MASK score was 0.71 (Lin’s concordance statistic). Item retest reliability, based on a weighted kappa, ranged from 0.57 to 0.87, suggesting moderate to substantial agreement between the 2 time points. The overall internal consistency among items 1 to 6 was moderate (0.65). MAKS is a brief instrument which can be used in conjunction with other scales related with attitudes and behaviour.
Mental health related attitudes
The Mental illness: Clinicians’Attitudes (MICA) was used to assess participants attitudes towards people with mental illness [28]. The scale consisted of 16 items and a 6-point scale which ranged from strongly agree to strongly disagree. A higher score indicated a higher level of negative stigma-related mental health attitudes. The MICA scale showed good internal consistency (alpha = 0.79). The test-retest reliability (concordance) was 0.80 (95% CI: 0.68–0.91). The Chinese version of MICA, which has been tested also shows good validity and reliability (Cronbach alpha = 0.72)[29].
Intended behavior
Reported and Intended Behaviour Scale (RIBS)[30] was used to measure intended behavior, which was derived from the Star Social Distance Scale[31]. We use four items-living with, working with, living nearby and continuing a relationship with someone-to assess the desire of future contact with people with mental illness. A higher score indicates less desire for social distance. The total RIBS score was standardised. Studies showed that the overall test-retest reliability of total RIBS score was 0.75 (Lin’s concordance statistic). Item retest reliability ranged from 0.62 to 1.0, indicating moderate to substantial agreement between time points. The overall internal consistency (Cronbach’s alpha ) among the subscale items was 0.85[30]. The Chinese version of RIBS has been shown to have strong internal consistency (Cronbach alpha = 0.82) and test-retest reliability (r = 0.68)[32].
Statistical Analyses
All of the statistical analyses were performed using the Statistical Package for Social Sciences (SPSS), Version 26.0 (SPSS Inc., Chicago, Illinois). The continuous variables were described using summary statistics, such as the means and standard deviations. The categorical variables were described using frequencies and percentages. The baseline characteristics of each group were compared using a t-test for the continuous variables and the chi-square test for the categorical variables.
The outcome measures were the MAKS, MICA and RIBS scores. Firstly, the change in the MAKS, MICA and RIBS total scores (after intervention vs baseline) was calculated for each participant. Secondly, the main strategy involved repeated-measures analysis of variance (ANOVA). For the dependent variables, 4 time points were used as the within effect, and group was used as the between effect. If the time-by-group interaction effect was significant, then multivariate analysis of variance (MANOVA) was applied to examine the group differences at post intervention, 1 month and 3 months. Finally, one-way repeated-measures analysis of variance was used to see the differences in outcomes among the different assessment points after intervention. Significance was set at p < 0.05 using a 2-sided test.