Effectiveness of a structured short intervention against stigmatisation in chronic visible skin diseases: Results of a controlled trial in future educators

Abstract Background Chronic visible skin diseases are highly prevalent, and patients affected frequently report feeling stigmatised. Interventions to reduce stigmatisation are rare. Objectives This study aimed to evaluate the effectiveness of a structured short intervention in reducing stigmatising attitudes towards psoriasis in future educators. Methods The intervention consisted of four components: (1) self‐reflection, (2) education on skin diseases, (3) contact between participants and a person with psoriasis and (4) practising of knowledge via case studies. A quasi‐experimental, pre–post study design was chosen with a nonrandomized contemporaneous control group that attended regular lessons. The main outcomes were participants' desire for social distance, stereotype endorsement, illness‐related misconceptions and intended behaviour. Intervention effects were analysed using mixed repeated‐measures analysis of variance, with Bonferroni post‐hoc tests for pairwise comparisons. Results The sample consisted of 221 students attending vocational training as educators (n = 118 intervention group, n = 103 control group). While no effect of the intervention was found in social distance, small to large effect sizes were observed for intended behaviour (r = .14), illness‐related misconceptions (r = .28) and stereotype endorsement (r = .42). The intervention group reported significantly higher satisfaction with the seminar compared to the control group. Conclusions Overall, the short intervention was effective at reducing stigmatising attitudes in future educators. In perspective, revised versions could help in reducing stigmatisation in various demographics and promote patient empowerment by acknowledging and including them as experts on their own behalf. Patient or Public Contribution Patient advocate groups were consulted and involved in the superordinate destigmatization research programme and intervention.

Patient or Public Contribution: Patient advocate groups were consulted and involved in the superordinate destigmatization research programme and intervention.

K E Y W O R D S
adults, dermatology, intervention, psoriasis, stigmatisation 1 | BACKGROUND AND OBJECTIVES Chronic visible skin diseases are a highly prevalent health problem. In Germany, around 10 million people are affected, most commonly by psoriasis, acne, rosacea and atopic dermatitis. 1,2 In addition to medical consequences, patients have a substantial psychosocial burden. A study among dermatological outpatients in 13 European countries 3 reported the presence of depression in 10.1%, anxiety in 17.2% and suicidal ideation in 12.7% of all patients. Patients with skin conditions also frequently report an impaired health-related quality of life 4 and feelings of stigmatisation. 5,6 Goffmann 7 (p. 3) defined stigma as a 'deeply discrediting attribute' that reduces the stigmatised individual 'from a whole and usual person to a tainted, discounted one'. Over time, the definition was extended and adapted, but essential components of stigma are generally agreed to be the recognition of a difference and devaluation and that stigma is a social construct. 8 Link and Phelan 9 broadened previous stigma concepts by including the perspective of the stigmatised and highlighting the influence of sociocultural processes in stigmatisation. According to their conceptualisation, stigma exists when four interrelated components converge: Individuals identify a difference and 'put a label' on it. Subsequently, (negative) stereotypes are linked to labelled individuals and they are placed in distinct categories to establish a sense of separation between 'us' and 'them'. Finally, labelled individuals experience status loss and discrimination that leads to unequal circumstances. The occurrence and associated negative consequences of stigmatisation have been reported in previous studies with regard to numerous illnesses like human immunodeficiency virus/acquired immunodeficiency syndrome, mental illness and obesity. [10][11][12] A substantial number of studies on the topic of stigmatisation in skin diseases so far have focused on psoriasis 5 and the World Health Assembly declared it to be a noncommunicable disease with particular relevance for health care delivery in 2014. 13 The visibility of skin changes in psoriasis facilitates the identification and labelling of a difference. Stereotypes about issues such as inadequate body hygiene lead to attribution of responsibility for the disease, while fear of infection or disgust results in a desire for social distance. Misconceptions about psoriasis and stigmatising attitudes are frequent in the general public. Halioua et al. 14 report that 16.5% of respondents believe that psoriasis is contagious and 6.8% believe that skin disease is related to personal hygiene. About half of the sample showed discriminatory behaviour towards patients with psoriasis, for example, reluctance to maintain friendship ties, to have a meal with a person with visible skin changes or to shake hands.
Moreover, 8% of respondents reported negative feelings about psoriasis, such as disgust, fear and rejection. Similarly, a representative survey investigating the perception of psoriasis among the German population found that 9% of respondents considered the disease to be communicable, 13% of the respondents stated that they would not shake hands and 7% of respondents would not want to eat at the same table with a person with psoriasis. A total of 10% reported that they would not want to live in the same household as an affected individual, 23% did not want to share the same swimming pool and 27% did not want to be in a personal relationship with someone affected by the disease. 15 For individuals affected by psoriasis, internalised stigma is frequent. 16 In addition to (self-) stigmatisation, previous studies also report a high perceived burden of social isolation based on the fear of social rejection, and impaired body image in affected individuals, all of which were associated with depressive symptoms. 17 In line with these findings, stigmatisation in psoriasis has been linked to impaired quality of life and mental health. 18 In psoriasis patients, stigmatisation was even found to be the most powerful predictor of depressive symptoms. 19 Consequently, affected individuals face a double burden of disease: the dermatological diagnosis itself and its potential stigmatisation.
It is therefore necessary to increase awareness and take steps in increasing acceptance and reducing stigmatisation of skin diseases as urged by recent World Health Organization recommendations. 20 There is clear consensus among health care professionals, researchers and patient organisations that systematic action is needed to, for example, raise awareness among the general public that psoriasis is a chronic condition that is not contagious, to identify and adequately treat patients who might be particularly vulnerable to stigmatisation and its negative consequences and to develop interventions that target the societal stigmatisation and discrimination of psoriasis. [21][22][23] Link and Phelan 9 (p. 367) specified that the term stigma can be applied 'when elements of labelling, stereotyping, separation, status loss, and discrimination co-occur in a power situation that allows the components of stigma to unfold'. It is therefore important for intervention efforts to address (potential) community facilitators in positions of influence, to reduce their stigmatising attitudes as early as possible and subsequently turn their societal or professional position to account. Individuals or groups that are WEINBERGER ET AL.
| 1791 frequently considered to be authority figures could for example contribute towards spreading awareness, correcting common misconceptions and preventing discrimination of individuals with skin diseases in future interactions. Moreover, patients are confronted with stigmatisation not just in adulthood but also already from childhood and adolescence. 24 The current short intervention was therefore focused on preschool teachers in training at a vocational college in Germany to take advantage of their unique position in not only educating children from a very young age but also being in frequent contact with parents of different societal and professional backgrounds.
The development of the current intervention was based on both practical considerations regarding implementation, future adaptability to different contexts and target groups and past evidence.
The primary objective of this study was to evaluate how effective the developed short intervention would be in reducing stigmatising attitudes in the participants of the intervention group in comparison to the control group. Ascertaining the predictive value of group status as well as confounding variables with regard to intervention effects were key secondary objectives. Another secondary objective was to investigate participants' satisfaction with the intervention to identify aspects that were viewed (un-)favourably, so future versions of the seminar can be modified accordingly.

| Trial design
The methodological approach of the superordinate destigmatization research programme and intervention has been detailed before. 25 In addition to different professions working in research and treatment, patient advocate groups were consulted and involved in the development and long-term implementation of the overarching research programme as well as the communication of its activities and results. Moreover, patients with psoriasis were directly involved in the short intervention as experts on their own behalf in the trialogue with participants and health professional.
Evaluation of the short intervention followed a nonrandomized quasi-experimental, pre-post study design with a contemporaneous control group. Due to the structure of the vocational education, assignment to study and control groups had to be done on the class level to avoid multiple or repeated participation by students of the same academic year. Participants in the control and intervention group participated in the study in parallel and were surveyed three times: immediately before (t 0 ) and after the intervention/regular class (t 1 ), and at 3 months of follow-up (t 2 ) to ascertain medium-term effects.

| Participants
The final sample consisted of N = 221 students (n = 118 intervention group, n = 103 control group) attending vocational training as educators at a college of social pedagogy in the city of Kiel, Germany.
Students eligible for participation had to be older than 18 years of age. Participants provided written informed consent and received €100 for their class fund at the end of the study. The study was approved by the local institutional review board (AZ: D521-18) and the Ministry of Education of the local government of Schleswig-Holstein. Sociodemographic data of the sample as well as statistical differences between the control and intervention groups are presented in Table 1.

| Outcomes
The primary endpoint with respect to the effectiveness of the intervention was the reduction of stigmatising attitudes in participants in the intervention group in comparison to the control group from baseline to t 1 , with scores remaining consistently low at follow-up as measured by the stereotype endorsement scale, the social distance scale, the psoriasis myth endorsement scale and the reported and intended behaviour scale (see below for details). Secondary analyses were done to ascertain the role of group status as well as confounding variables (see below for details), and to analyse satisfaction of participants with the seminar.

| Primary outcome measures
For their study of stigmatising attitudes towards persons with psoriasis in laypersons and medical students, Pearl et al. 26 adapted and developed instruments to assess stigmatisation in cooperation with dermatologists with expertise in psoriasis. The current study adapted the scales for use in German samples with their permission and undertook an initial validation during a pilot study. Two of the original scales ('Emotions', 'Attitudes toward treating patients') were not included due to time constraints and irrelevance to the current sample, respectively.

Stereotype endorsement
A semantic differential scale consisting of 11 adjective pairs (e.g., unattractive-attractive) was used to assess stereotype endorsement of individuals with psoriasis. Participants were asked to mark the circle closest to the adjective that they considered to describe a person with psoriasis (ranging from 1 to 5, with 1 being closest to the positive adjective). Scores were averaged, with higher scores indicating greater endorsement of negative stereotypes.

Social distance
The social distance scale measures a participant's desire to eschew persons with psoriasis in nine different social situations (e.g., 'share a meal'). Items were rated on a 5-point Likert scale ranging from 'definitely' to 'definitely not'. Item scores were averaged, with lower scores indicating lower desire for social distance.

Illness-related misconceptions ('myth endorsement')
Agreement with common misconceptions about the disease was analysed with 15 statements about psoriasis (e.g., 'psoriasis is caused by poor hygiene'). Participants rated the statements on a 5-point Likert scale ranging from 'strongly disagree' to 'strongly agree'. Scores were averaged, with lower scores indicating lower endorsement of the misconceptions.

Reported and Intended Behaviour Scale
The Reported and Intended Behaviour Scale (RIBS) 27 was adapted for use in skin diseases to explore stigmatising behaviour in the current study. Four items of the RIBS assess the prevalence of T A B L E 1 Descriptive statistics (mean and standard deviation) of sociodemographic data, depressive symptoms, personal affliction with a skin disease, contact with people affected by a skin disease and stigmatising attitudes at t 0  Table 1). In addition, participants were screened for depressive symptoms to explore their potential influence on their stigmatising attitudes and their evaluation of the attended seminar.
Patient Health Questionnaire-9 The depression module of the Patient Health Questionnaire (PHQ) is a valid and reliable self-report measure to assess depressive symptoms. 28 Participants were asked to report for nine depressive symptoms, respectively, in terms of whether and how often these had bothered them in the previous 2 weeks. The total score-a measure of depression severity-ranges between 0 and 27, with scores of 5, 10, 15 and more indicating mild, moderate and moderate to severe symptoms of depression.

Satisfaction with the seminar
A short self-developed questionnaire was used at data collection point t 1 and t 2 to measure participants' satisfaction with the (intervention) class. The 12 items were selected to ascertain for example, students' satisfaction with the length and content of the seminar, personal estimation of its relevancy in their (working) life and whether they would recommend the seminar to other students. Participants rated the statements on a 5-point Likert scale ranging from 'strongly disagree' to 'strongly agree'. Scores were averaged, with higher scores indicating higher satisfaction.
In addition, a free-text feedback box that allowed general comments and suggestions was displayed at the very end of the questionnaire.

| Sample size calculation
Due to the lack of intervention studies addressing stigma in this field, the results of a previously conducted pilot study were used for an estimation of effects. An a priori power analysis was conducted using G*Power. With an alpha level of .05, minimum power established at .90 and a moderate expected effect size of .45, 29 210 participants would be necessary to find a statistically significant effect in the model.

| Group allocation
As was indicated previously, the allocation of students to the control and intervention groups took place on the class level. To decrease the risk of bias, group assignment was decided by the school timetable of September 2019 in combination with the availability of the psoriasis patient and the medical expert, respectively. Class hours were compared to dates and time given by the patient and medical expert, and classes matching these dates were allocated to the intervention group, while classes not matching the dates were allocated to the control group.

| Statistical methods
Data analyses were carried out using SPSS 25.0.0. 30 T tests, Mann-Whitney U tests and χ 2 tests were used to analyse differences between groups in sociodemographic as well as the main outcome variables (see Table 1).
Intervention effects were analysed using mixed repeated- Multiple regression analyses were conducted to examine whether the four main outcome scores at t 1 and t 2, respectively, can be predicted based on study group membership, gender, age, depressive symptoms, personal affliction with a skin disease, frequency of contact with individuals with skin diseases and previous vocational training in a social field (see Tables 2-5).
p-Values were obtained from two-tailed tests, with a p < .05 indicating statistical significance.

| Participant flow and recruitment
Overall, 253 students filled in questionnaires at the three data collection points within their classes. Three students did not fill out the t 0 questionnaire, n = 11 did not report data for the

| Baseline data
The control and intervention groups did not differ significantly regarding their sociodemographic data or the main outcome measures at t 0 . Descriptive statistics for the total sample as well as the control and intervention groups separately are illustrated in Table 1.

| Numbers analysed
Analyses included all participants who filled in both the t 0 and t 1 questionnaires (n = 221). A total of 21 participants of the control group and 17 participants of the intervention group were lost to follow-up t 2 ; thus, data from 183 students were available for primary analysis.

| Stigmatising attitudes towards individuals with psoriasis
Over the survey period, no interaction effect between study group and time was determined for stereotype endorsement and t 0 -t 2 (0.29, p < .001), while no significant difference was found between t 1 and t 2 (−0.04, p = .925). In the control group, scores differed significantly between t 0 -t 1 (0.21, p < .001) and t 0 -t 2 (0.20, p = .014), but not between t 1 and t 2 (−0.01, p = 1.00).

| Role of group status
All variables and their statistical contribution to the prediction of the main outcome scores at t 1 and t 2 are illustrated in Tables 2-5.
Participants' study group status, gender, age, depressive symptoms, frequency of contact with individuals affected by a skin disease, personal affliction with a skin disease and previous vocational training in a social field explained a significant proportion of variance in social distance at t 1 (R 2 = .05, F(7, 199) = 2.42, p = .021), but not at t 2 (R 2 = .03, F(7, 166) = 1.78, p = .094). At both time points, group status did not contribute significantly to the prediction of social distance.
Further, the predictor variables explained a significant proportion of variance in stereotype endorsement at t 1 (R 2 = .13, F(7, 199) = 5.21, p < .001) and at t 2 (R 2 = .11, F(7, 164) = 3.98, p < .001).  32,33 Positive results were also found for information-based approaches as well as approaches based on contact with affected groups to foster patient participation. 34 However, previous stigma interventions mainly targeted patients with leprosy and were conducted in low-and middleincome countries. Moreover, substantial heterogeneity in study design as well as quality was observed, making direct comparison of results-especially with regard to the interventions' effectiveness-difficult. 31 The current results bear greatest similarity in both the methodology and the findings to interventions addressing the stigma of mental illness. Corrigan et al. 35

| Limitations
The following limitations need to be taken into consideration when interpreting the current results: Randomised assignment to control and intervention groups was not feasible considering the structure of the vocational training. To minimise selection bias, group assignment did not depend on the researchers involved as much as it did on matches in schedules. The two groups did not differ significantly in any of the control or outcome variables at the beginning of the study.
Since the study was designed and implemented under as Relating thereto is the sample itself: As future educators at a college of social pedagogy, students had working experience in a social profession either due to a 1-year internship and/or their previous jobs. Taken together with their ongoing vocational education, there might have been more awareness of and familiarity with the topic of exclusion/bullying among them even though the term stigmatisation/discrimination may not have been explicitly stated in their regular classes. As indicated above, on average, stigmatising attitude scores in the sample were lower compared to the layperson sample and more similar to the sample of medical students in the study that originally used the current assessment tools, 26 further corroborating this assumption. Future studies should therefore also include different vocational colleges to ascertain the potential influence of their curricula on stigmatising attitudes in addition to effects of the intervention. Conducting and evaluating interventions in other target groups that already are, or will be in frequent contact with individuals with skin diseases due to their occupation, like for example hairdressers or swimming pool attendants, would be an important addition to anti-stigma efforts in general and the current findings in particular.
In line with this, future investigations should include other skin diseases and/or more than one skin disease to ascertain the effectiveness of the intervention beyond psoriasis. Several student remarks in the free-text feedback option box addressed this constraint, mentioning a great interest in and wish for inclusion of different skin diseases as well as more than one patient in future seminars.
Another limitation that needs to be considered is the time frame of the current evaluation, which makes causal interpretations of the current results difficult if not impossible. While 12 weeks allow for a discussion of short-to midterm effects, there is still a need for longitudinal studies to properly estimate the longterm effectiveness of the short intervention as well as potential causal associations between outcomes and control variables.

| CONCLUSION
The current short intervention to reduce stigmatising attitudes Natascha-Alexandra Weinberger and Claudia Luck-Sikorski interpreted the study results. Natascha-Alexandra Weinberger and Sonja Mrowietz wrote the first manuscript. All authors discussed the results and findings and also discussed and edited the final manuscript.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.