Scale Development and an Educational Program to Reduce the Stigma of Schizophrenia Among Community Pharmacists: Study Protocol for a Randomized Controlled Trial

Background Stigma associated with mental disorders is rooted among many pharmacists, and represents a major barrier to patient support in community-based psychiatry. In the current study, we rstly developed an assessment scale that is specically designed to assess the level of stigma that pharmacists may have toward patients with schizophrenia, and then examined the effects of reducing stigma with an educational program that focuses on communication with schizophrenic patients using a newly developed Stigma Scale towards Schizophrenia for Community Pharmacists (SSCP). Methods was developed by exploratory factor analysis with rotation based on responses from 822 randomly selected community pharmacists. Furthermore, a randomized controlled was conducted to clarify the of were individually to two a lecture group only schizophrenia) or communication group communicating and the The stigma assessment using SSCP was immediately before and after the educational

schizophrenia has not yet been demonstrated.
We previously employed stigma scales to measure general attitudes towards mental disorders and social distance manifested as avoidance responses to the mentally ill in society [13,14]. To con rm the effectiveness of educational programs for pharmacists, stigma scales that accurately simulate actual settings, in which pharmacists manage patients with schizophrenia, are essential. Therefore, a scale to assess the stigma of schizophrenia among pharmacists was developed in the present study and its reliability was veri ed. A randomized controlled trial was then conducted to con rm the effectiveness of the pharmacist educational program with a focus on communication with patients to reduce the stigma of schizophrenia.

Methods
The present study was approved by the Meijo University Research Ethics Board to ensure privacy, con dentiality, and anonymity (H30-3). It was conducted according to the principles expressed in the Declaration of Helsinki.
Development of a scale to assess the stigma of schizophrenia among community pharmacists To develop a scale to assess the stigma of schizophrenia among community pharmacists, items related to the constructs of stigma as a barrier for pharmacists to provide professional support for patients with schizophrenia were extracted from existing scales [5,[15][16][17][18]. A pool of items was created with these items and those newly developed upon deliberations among 3 pharmacists (1 university professor conducting research on psychiatry, 1 communication specialist, and 1 researcher). There were 67 items, representing 3 constructs; 1) recognition of patients with schizophrenia, 2) social distance, and 3) self-disclosure/help-seeking behavior. The appropriateness of the 67 items in terms of content and expression was then examined with 12 pharmacists, including 2 of the above-mentioned pharmacists and 3 pharmacists specializing in psychiatry, and a 33-item scale was created.
The assessment tool was distributed with a survey on personal background information to 1,500 pharmacies in Aichi Prefecture between December 1, 2018, and January 31, 2019. These pharmacies were randomly selected. A 5-point scale was used in the assessment tool as follows: 1: strongly agree, 2: agree, 3: neither agree nor disagree, 2: disagree, 1: strongly disagree. Negative responses were given 5 points, and positive responses 1 point. Correlations between each item and the total score (item total correlation) were con rmed from the scale proposals collected. Based on the answers provided, an exploratory factor analysis was performed using the principal axis factoring method with a promax rotation to further examine the factors constituting the assessment tool. The 33-item scale created by 12 pharmacists described above was extracted into a 27-item scale by the factor analysis.
Cronbach's α con dence coe cient was calculated for each item and factor of the assessment tool to assess the level of internal consistency. The correlation between the total score and social desirability scale [19,20] was con rmed from the 27-item scale. The test-retest reliability of the 27-item scale was veri ed by conducting a retest for those who consented among the sub-sample of community pharmacists eight to twelve weeks after the completion of the survey, and the intraclass correlation comparing the total scores at the 2 points they completed was calculated.
Furthermore, to con rm criterion-related validity, the correlation between the total 27-item scale score and total score from the following 2 conventional scales was used to assess the stigma of mental disorders: the Whatley Social Distance Scale (WSDS) [17] and Index of Attitudes toward the Mentally Ill (IATM) [15]. WSDS examines attitudes that prevent social participation by patients with mental disorders (social distance), and IATM measures the levels of negative recognition of these disorders. We used the 2 scales in our previous studies, involving pharmacy students [13], and an internet survey [10], conducted by physicians, nurses, pharmacists, and general citizens. The scale developed through these processes was named the Stigma Scale towards Schizophrenia for Community Pharmacists (SSCP), and was used in subsequent studies.

Randomized controlled trial
A randomized controlled trial was conducted to clarify the effects of reducing the stigma of schizophrenia using an educational program for community pharmacists with a focus on communication with patients with schizophrenia.

Participants and study design
An outline of the study design is shown in Figure 1. Community pharmacies in Aichi Prefecture in Japan were randomly selected to receive a document to recruit participants by e-mail or post, and consent was obtained from 120 pharmacists belonging to these pharmacies. They participated in the educational program twice, on July 28 and November 10, 2019, and were divided into 2 groups of 60, adopting the strati ed block randomization method using a computer-generated randomization list with a block size of four: a lecture group (only attending a lecture on schizophrenia) and communication group (communicating with patients with schizophrenia and attending the lecture). Randomization was strati ed by (1) sex (female versus male), (2) age (<30 years versus >30 years), (3) experience of communicating with patients with mental disorders (whether participants have or do not have this experience). Following the exclusion of 4 participants in the lecture group and 1 in the communication group, who withdrew, the nal numbers in each group were 56 and 59, respectively (a total of 115 participants). Approximately 50% of each group participated in each session. After the lecture, the group to which each participant belonged was disclosed.
Outline of the educational program All 115 participants attended a 60-minute lecture on schizophrenia given by a psychiatrist. The contents of the lecture were the epidemiology, symptoms, diagnosis, treatment methods based on the latest evidence, treatment effects, main side effects, and prognosis of schizophrenia.
After the lecture, 59 communication members formed groups of 4 or 5 to perform the following activities in a single room: a lecture staff introduction and breaking the ice (self-introduction in each group); a lecture on mental disorders and the associated stigma (prejudice and discrimination); group work 1: holding a group discussion on the management of patients with schizophrenia, and making a presentation with 1 member of each group (a total of 5 groups) as the presenter; group work 2: holding a group discussion and offering opinions on the points of an interview with patients with schizophrenia to clarify their experience; an interview with a patient with schizophrenia, who was allocated to the table of each group (a total of 6 patients), and introduced him/herself for 20 minutes using a self-introduction sheet previously lled out; interviews with 2 other patients (a total of 3 rotations); group work 3: holding a group discussion on points of learning by pharmacists from patient experiences, making a presentation with 1 member of each group (a total of 5 groups) as the presenter, followed by a lecture to summarize the opinions offered at the group presentations.
The 6 (4 males and 2 females) patients with schizophrenia who shared their experiences belonged to a patient group in Nagoya city. They had taken antipsychotics for 5 years or longer and were visiting psychiatric hospitals as outpatients. Their signed consent was previously obtained using a written document specifying the study objective.
After the consent process, they entered: 1) medical history, 2) di culties associated with pharmacotherapy, 3) issues they may or may not consult about with community pharmacists, 4) cases in which they had perceived stigmatizing behaviors/attitudes (prejudice and discrimination), and 5) demands to be ful lled by community pharmacists, in a selfintroduction sheet, and rehearsed self-introductions using this sheet.

Stigma assessment
The stigma of schizophrenia among community pharmacists was assessed using SSCP at 3 points: before the lecture (both groups: T1), immediately after the lecture (lecture group: T2), and immediately after communicating with patients (communication group: T3). SSCP consists of 27 statements to be evaluated on a 5-point scale: <Strongly agree>, <Agree>, <Neutral>, <Disagree>, and <Strongly disagree>, which were scored as 1-5, respectively. Five items among 27 statements were reverse scored (i.e., #11, 20, 24, 25, and 26). The total score ranged between 27 and 135. Scores that were higher than or equal to a median of 81 represented more favorable attitudes.

Statistical analysis
Statistical analyses were performed using IBM SPSS statistics ver. 22. The attributes of the 2 groups were compared using the Mann-Whitney U test and chi-squared test. The Wilcoxon signed-rank test was used to compare between before and after the educational program. The effect size r was calculated using the standardized test statistic (Z) and sample size (N) (r=Z/√N). An effect size of 0.1 was considered to be small, 0.3 medium, and 0.5 or more a large effect. The signi cance of differences was set at two-tailed p<0.05 unless otherwise speci ed.
To identify fundamental factors from the 33-item scale, an exploratory factor analysis was performed using the principal axis factoring method with a promax rotation. The choice of the number of factors was based on the scree plot. Items with factor loadings lower than 0.4 were deemed meaningful and assigned to the given factor, with only the highest factor loading for each item being considered. Even if an item had a factor load of lower than 0.4, the item was adopted if researchers found it necessary to explain the construct to which the item belongs. We labeled each factor based on what best characterized the group of items that loaded on a particular factor. The internal consistency of the SSCP and subscales was evaluated using Cronbach's α coe cient and construct validity for SSCP was examined using Spearman's correlation (rs).

Results
Testing of the 33-item scale to assess reducing stigma Response rate and respondent characteristics There were 822 responses (response rate: 54.8%), and 806 were valid for analysis (Valid response rate: 53.7%). The mean age of respondents was 42.3 ± 12.1 years. There were 490 (60.8%) males and 316 (39.2%) females. Their mean length of pharmacy experience was 15.6 ± 10.4 years.

Item-total correlations
An item-total correlation of 0.2 to 0.5 is considered to be the most appropriate [16]. None of the 33 statements showed an item-total correlation of lower than 0.2.
Exploratory factor analysis Table 1 shows factorial patterns after the promax rotation, inter-factor correlations, Cronbach's α coe cient, and the median (inter-quartile range) for each item. Prior to the factor analysis, the mean score and standard deviation for each item were calculated to con rm the absence of a ceiling or oor effect. In the Kaiser-Meyer-Olkin measure of sampling adequacy, it is recommended that a range of 0-1 be allowed and values need to be higher than 0.6. The factor model showed a coe cient of 0.91, con rming its su cient sampling adequacy.
Signi cant results (χ2 6578.2, df=351, p<0.001), con rming the appropriateness of this model, were also obtained from Bartlett's test for Sphericity. Following the examination of all 33 items through an exploratory factor analysis, 4 factors and 27 items were selected. The cumulative proportion of variance explained was 45.8%. Six items with a factor loading lower than 0.4 were removed. The 6 items removed were as follows: "When patients with schizophrenia present with physical symptoms (e.g. nausea, back pain, headache), I might think that they are manifested because of mental issues", "I feel that there is nothing I can do for patients with schizophrenia in terms of their recovery", " I cannot understand the behavior of patients with schizophrenia that is caused by hallucinations and delusions", "I think I can actively identify speci c problems that patients with schizophrenia may have", "I cannot be friends with a patient with schizophrenia", and "If I had schizophrenia and was not able to control the symptoms myself, I would consider myself a weak person".
However, items judged to be necessary for explaining the components were not excluded, even if factor loading was lower than 0.4. The 4 factors were named as follows: Factor I; social distance at work, Factor II; recognition of patients with schizophrenia, Factor III; self-disclosure, and Factor IV; social distance in daily life. The median (inter-quartile range) of the total score was 70.0 (63.0-78.0) for the entire scale, 29.0 (25.0-34.0) for factor I, 19.0 (17.0-22.0) for factor II, 11.0 (10.0-12.0) for factor III, and 10.0 (9.0-12.0) for factor IV. These scores indicated a relatively positive response because they were lower than 50% of the total score.
Testing of the 27-item scale to assess reducing stigma Internal consistency Cronbach's α was 0.89 for the entire scale, 0.88 for <Factor I: social distance at work>, 0.76 for <Factor II: recognition of patients with schizophrenia>, and 0.62 for <Factor III: self-disclosure> and <Factor IV: social distance in daily life>.

Test-retest reliability
Among the initial samples, a subset of 81 consented to the retest. The intra-class correlation between the total scores at the 2 points was 0.90 (95% CI 0.84-0.93, p <0.001), exceeding 0.7, and, thus, the value con rmed su cient reliability.

Social desirability bias
No correlation was observed between the total SSCP score of 806 pharmacists and the total social desirability scale (rs= −0.117, p 0.213).

Criterion-related validity
The correlation between the total SSCP score and total WSDS/IATM scores was calculated. Total score correlations between SSCP and WSDS or IATM were 0.58 (p<0.001) and −0.62 (p<0.001), respectively.
Randomized controlled trial to assess effects of contact-based educational programs on lectures alone.
No signi cant differences were observed in participant backgrounds between the 2 groups ( Table 2).

Effects of the contact-based educational program on reducing stigma
In the communication and lecture groups, the total SSCP score and Factor I, II, and III scores after the educational program were signi cantly better than the baseline score (Table 3). Educational program-related changes in the median (interquartile range) total SSCP score were −9.0 (−16.0 -−5.0) in the communication group and −3.0 (−7.0 -1.0) in the lecture group (improvement rate, 15.5% and 5.2%, respectively; p < 0.001) ( Table 3).
The improvement rates for each factor in the communication and lecture groups were as follows: Factor I: 18.3 and 7.1%, II: 23.0 and 4.5%, III: 5.0 and 6.7%, and IV: 1.0 and 0%, respectively, revealing marked improvements in scores for Factors I and II (p = 0.001 and p < 0.001, respectively). The effect sizes associated with the contact-based educational session were as follows: entire SSCP: 0.41, Factor I: 0.32 and II: 0.49, respectively, revealing a moderate difference among both groups.
Effects of demographic characteristics on differences in stigma between two groups Table 4 shows the effects of demographic characteristics on differences in stigma between the 2 groups. Among participants in their 40s and older or who had experience of schizophrenia via family members or close friends or provide medication counseling for a few patients with schizophrenia each week, no additional effect of the contact-based educational intervention on the lecture was observed. However, in all subgroups other than that described above, participants in the communication group showed a signi cantly greater change than those in the lecture group.

Discussion
The role of community pharmacists in Japan is becoming an important challenge because the number of outpatients with schizophrenia is expected to steadily increase in the future. However, the stigma of community pharmacists towards schizophrenia is a major barrier to medication support for these patients, and also keeps pharmacists themselves from making the most of their professional skills. To the best of our knowledge, the present study is the rst to develop a scale to speci cally assess the stigma of schizophrenia among community pharmacists in Japan. By achieving Cronbach's α of 0.7 or higher for the entire scale, SSCP had su cient internal consistency and appears to be a reliable and valid scale for assessing the stigma of schizophrenia among community pharmacists. Furthermore, a randomized controlled trial design was adopted in the present study and con rmed the effects of contact-based educational interventions using SSCP to reduce the stigma of schizophrenia among community pharmacists. A previous study reported that the recurrence rate was approximately 5-fold higher among patients with schizophrenia who discontinued their medication than among adherent patients [21]. The stigma of schizophrenia among pharmacists negatively affects medication-related behaviors by patients, which may lead to worse symptoms due to medication withdrawal [22,23].
Many community pharmacists at work in Japan keep their social distance from patients with mental disorders [10], and concerns have been expressed over stigmatization by pharmacists, which increases the di culties associated with appropriately managing these patients.
In pharmacotherapy for schizophrenia, for which poor medication adherence is regarded as problematic [23,24], community pharmacists need to continuously provide medication support, including con rmation of the therapeutic and adverse effects of drugs, which helps patients to continue their medication and improve their quality of life in a responsible manner. The total SSCP and 4 factor scores of 115 pharmacists who participated in the educational program as part of the present study were lower than the midpoints. Thus, participants were individuals with relatively positive attitudes towards patients with schizophrenia. Moreover, their values were similar to those in our previous study to develop SSCP, involving 822 community pharmacists.
In comparisons of the 2 groups (lecture and communication) based on the results of the randomized controlled trial, SSCP scores improved more in the communication group, supporting the usefulness of the educational program combining educational interventions based on knowledge of schizophrenia and communication with patients to reduce the stigma of this disorder among community pharmacists. The lecture on schizophrenia itself also effectively reduced this stigma; however, the effect was enhanced by adding communication with patients to the lecture. The effectiveness of an educational intervention to reduce the stigma of mental disorders has already been con rmed in meta-analyses performed in other countries [11]. Social contact or contact-based interventions have been identi ed as the most effective strategy [12]. Gri ths et al. previously reported that the effects of social contact to reduce stigma is enhanced by combining it with knowledge-based education [11], and these ndings support the present results. Pettigrew and Tropp reported curricula complementing knowledge-based elements of education with social contact to effectively reduce stigma by decreasing anxiety and promoting empathy [25].
The present study adopted an unconventional program to provide interventions for pharmacists, with a focus on communication with patients. In this program, pharmacists initially held group discussions to clarify the di culties associated with managing patients with schizophrenia in daily services, and then classi ed their questions about these patients to directly ask them to actual patients for con rmation during a communication session for each group. Some of the questions from pharmacists, such as "How would you describe the stigmatic attitudes, statements, and behaviors any pharmacists have ever shown towards you?", "How did you feel when you actually perceived any stigmatization by pharmacists?", and "What do you expect from pharmacist?", highlighted the issue of stigma, and patients freely answered these questions. At the end of the program, the communication group discussed approaches to be adopted by pharmacists in the future based on their experience of communicating with patients. This educational intervention content was originally provided in this program, and its usefulness was con rmed by the stronger stigma-reducing effect achieved in the communication group than in the lecture group. In any case, there is currently no other educational program that speci cally addresses the stigma of schizophrenia among community pharmacists in Japan or other countries.
Among the SSCP subscale scores, <Factor II: recognition of patients with schizophrenia> improved the most, possibly contributing to the improvements observed in scores for <Factor I: Social distance at work>, which prevents pharmacists from making the most of their professional skills.
In our previous study, the feasibility of reducing the social distance of community pharmacists from patients with schizophrenia by resolving their misunderstanding of these patients as dangerous was also suggested [26]. Furthermore, the establishment of an equal relationship between healthcare professionals and patients through contact-based

Conclusion
The present results suggested the usefulness of our original scale of SSCP and educational program for community pharmacists with a focus on communication with patients to assess and reduce, respectively, the stigma attached to schizophrenia by these pharmacists, among whom stigmatization is more serious than in other areas of mental health services. The present study was approved by the Meijo University Research Ethics Board to ensure privacy, con dentiality, and anonymity (H30-3). It was conducted according to the principles expressed in the Declaration of Helsinki. All participants were informed about the study, and all provided written informed consent.     The values in the table represent changes in the median (inter-quartile range) total SSCP score