Comparison of Long-term Results of Computer-Assisted Anti-Stigma Education and Reading Anti-Stigma Educational Materials

Abstract

Background

Professionals working with psychiatric patients very often have negative beliefs and attitudes about their clients. We designed our study to investigate the effectiveness of anti-stigma interventions among university students who are trained to provide special education.

Objective

The objective of our study was to compare sustainability of the effect of two anti-stigma education programs.

Methods

We enrolled 91 college students from the School of Special Education at the Herzen Russian State Pedagogic University (St. Petersburg, Russia). Of those, 36 read two articles and World Health Organization brochure (reading group, RG) devoted to the problem of psychiatric stigma, and 32 studied an anti-stigma web-based program (program group, PG). Twenty-three students were in a control group (CG) and received no intervention. The second study visit in six months was completed by 65 students. To measure the level of stigma we used the Community Attitudes toward the Mentally Ill (CAMI) questionnaire. The web-based program was based on the Computer-assisted Education system (CO-ED) which we described previously. The CO-ED system provides self-paced interactive education driven by adult learning theories.

Results

At the time of their first visit the age of the study participants was 19.0±1.2 years; of them, 99% were females. After the intervention in PG, the level of stigma assessed by CAMI decreased from 24.0± 5.0 to 15.8±4.6 points (p<0.0001). In RG the level of stigma dropped from 24.1±6.1 to 20.3±6.4 points (p<0.0001). In six months after the intervention the analysis of CAMI scores showed that the level of stigma in PG was significantly lower than in CG and RG (20.2±6.2 in CG, 21.3±6.5 in RG, and 18.7±4.9 in PG, p<0.01).

Conclusions

Web-based education or reading anti-stigma materials could be effective in reducing psychiatric stigma among university students. The effect of interactive web-based education based on adult learning theories was more stable as assessed in six months.

Introduction

Stigma is a negative label that people frequently attach to groups or persons who are different from them in some respect, such as race, appearance, physical or mental health. Stigma is often connected with discrimination, i.e., treating the stigmatized group differently than other people, denying them personal and civil rights [1]. Numerous approaches are possible to reduce psychiatric stigma [2, 3], such as media campaigns [4], workshops for professionals in contact with people with psychiatric disorders [5], lectures [6,7], computer-assisted education [8], etc. The stigma of psychiatric diseases is a barrier to the rehabilitation of mentally ill patients and their effective functioning in society. Professionals working with psychiatric patients very often themselves have negative beliefs and attitudes about their clients. We designed our pilot study to investigate the effectiveness of anti-stigma interventions among university students who are trained to provide special education.

For computer-based education we used the Computer-assisted Education system (CO-ED) which we described previously [10,11]. The CO-ED system provides self-paced interactive education driven by adult learning theories [12–14]. Theoretical underpinnings of the CO-ED system design are summarized in Table 1.

Table 1.

Theoretical underpinnings of the design of Depression Stigma CO-ED program

Theory	Concept	Use in the program
Drive Reduction Theory (C. Hull) [15]	Response must be made in order for conditioning to occur	Interactivity software supports feedback loop based on the following sequence: educational message – multiple-choice question – explanatory feedback
Operant Conditioning (B.F. Skinner) [16]	The learner should make a response for every unit of learned behavior or information and receive immediate feedback	Providing immediate feedback on performance of a user
Operant Conditioning (B.F. Skinner) [16]	Information should be presented in small amounts so that responses can be reinforced	Information presented as a sequence of short messages
Information Processing Theory (G. Miller) [17]	Short-term memory can only hold 7±3 chunks of information where a chunk is any meaningful unit	Information provided to a user in short sections containing 7–10 educational messages
ACT – Architecture of Cognition Theory (J. Anderson) [18]	Provide instruction in the context of problem-solving	Providing case scenarios (asking users to select what they would do in problem situations)
Cognitive Flexibility Theory (R. Spiro, P. Feltovitch & R. Coulson) [19]	Instruction should be case-based	Providing case descriptions (personal stories) of real patients with depression.
Cognitive Flexibility Theory (R. Spiro, P. Feltovitch & R. Coulson) [19]	Learning activities must provide multiple representations of content	Information is presented using visual, audio, and textual format
Cognitive Load Theory (J. Sweller) [20]	Eliminate the working memory load by decreasing redundancy	Patient-tailored selection of information for the educational curriculum

Methods

We enrolled 91 graduate students from the Herzen Russian State Pedagogic University (St. Petersburg), School of Special Education. The study subjects were randomly allocated to one of three groups. They had two study visits with an interval of 6 months. The second study visit was completed by 65 students due to the turnover of the students and absence some of them from the classes. The samples at visit one and visit two were not statistically different in relation to age (with a correction for 6 months), gender, and psychosocial history.

At the time of their first visit the age of the participants was 19.0±1.2 years. Of them, 99% were females, and 1% (1 person) males, that reflected the gender distribution in the School. All the students in the sample had just completed a one-year long eight-credit psychiatry course taught by experienced faculty, mainly consisting of medical doctors. They were finishing their second year of a five-year university program for special education teachers, in training to provide education to children with mental retardation, speech, and hearing disorders. Of those students, 36 read two articles and World Health Organization brochure (reading group) devoted to the problem of psychiatric stigma, and 32 studied an anti-stigma computer program (program group) content for which was created at the University of Maryland in collaboration with the St. Petersburg State University.

The educational program driven by adult learning theories provided a sequence of short educational messages with each message followed by a multiple choice question. The educational curriculum was divided into a set of sections with a quiz after each section. The curriculum addressed three components of stigma: cognitive (lack of knowledge or untrue beliefs about psychiatric disorders), emotional (feelings toward people with these conditions and its treatment), and behavioral (behavior toward people with mental health problems). A stand-alone CD version of the computer-assisted system was used whenever access to Internet was limited. The functionality of the web-based and stand-alone versions was identical.

Twenty-three students were in a control group and received no intervention. The methods of anti-stigma education were chosen to avoid a personal influence of a lecturer and any fluctuations in the content of educational sessions across different institutions and groups of students.

We collected information on socio-demographic data and mental health history of the participants before the intervention. To measure the level of stigma before and after the intervention, we used the Bogardus Social Distance Scale (BSDS) and the Community Attitudes toward the Mentally Ill (CAMI) questionnaire. BSDS measures mainly emotional and partly behavioral component of stigma. It allows measurement of stigma toward people with different stigmatized conditions; the students were given three copies of the BSDS with the vignettes on severe heart, skin and psychiatric disease correspondingly. The CAMI scale was used to evaluate the cognitive component of psychiatric stigma.

Results

In the study sample, 42.9% (39 subjects) mentioned that they had experienced feelings of depression or anxiety or had other psychological problems for more than two weeks in the past; 40.7% (37) wanted to visit a psychiatrist (psychotherapist) because of their problems. Three subjects (3.3%) underwent psychiatric treatment, and 24.2% (22 subjects) thought they had at the time of study or in the past serious emotional or psychiatric problems, but nobody in the sample considered themselves as having a psychiatric disease. Twelve people (13.2%) had relatives with psychiatric diseases, 23 (25.3%) had friends or friends of their family who had some mental disorder. Socio-demographic parameters and mental health history did not statistically differ between the reading, computer and control groups.

The level of social distance was highest toward a person with a mental disorder (18.6±5 points), intermediate for a skin disease (14.7±5 points), and lowest for a heart disease (11.9±4.3). Levels of stigma toward all these three conditions were statistically different, with p<0.001 (according to ANOVA). The levels of the heart disease stigma and psychiatric stigma were highly correlated (Pearson r=0.92, p>0.001), while the skin disease stigma was not correlated with other stigma levels. To describe the sample, we divided the range of BSDS possible scores (from 0 to 21) into 3 equal parts. Most people in the sample had a moderate level of psychiatric stigma. The level of stigma was not connected with education or mental health history, maybe because of the low variability of these parameters in the sample. According to the CAMI scale, many students shared stigmatizing beliefs about mental disorders. The level of stigma toward mental conditions was not different across three study groups.

Both interventions were initially highly effective in reducing psychiatric stigma (See Table 2 and Table 3). In the program group the level of stigma decreased from 18.8±3.8 to 14.2±4.6 points (BSDS, t-test for paired samples p<0.0001) and from 24.0±5.0 to 15.8±4.6 points (CAMI, t-test for paired samples p<0.0001). In the reading group the level of stigma dropped from 18.5±3.9 to 15.3±4.4 points (BSDS, t-test for paired samples p<0.0001) and from 24.1±6.1 to 20.3±6.4 points (CAMI, t-test for paired samples p<0.0001).

Table 2.

The level of stigma (according to CAMI and BSDS) in three groups at baseline, post-test I (immediately after the intervention), and at post-test II (six months after the intervention).

Scales	Groups	Pre-Test	Post-Test I	Post-Test II
CAMI	Reading Group	24.1±6.1	20.3±6.4*	21.3±6.5
	Program Group	24.0±5.0	15.8±4.6*	18.7±4.9**
	Control Group	20.8±5.1	NA	20.2±6.2
BSDS	Reading Group	18.5±3.9	15.3±4.4*	16.8±4.4
	Program Group	18.8±3.8	14.2±4.6*	9.4±3.8**
	Control Group	19.7±4.7	NA	19.6±4.4

Numbers with one asterisk (*) means that the difference between pre-test and post-test I was significant at p<0.001,

two asterisks (**) - that the difference between pre-test and post-test II was significant at p<0.001

Table 3.

Differences between the groups in the level of stigma. Statistically significant differences are given in bold.

Statistical Tests	Groups	CAMI	BSDS
ANOVA between groups, pre-test	All	0.054	0.052
t-Test, paired samples, pre-test vs. post-test I	Reading	<0.00001	<0.00001
	Program	<0.00001	<0.00001
ANOVA between groups, post-test II	All	0.313	0.033
t-Test, two-sample assuming equal variances, pre-test vs. post-test II	Reading	0.092	0.117
	Program	<0.001	0.024
	Control	0.735	0.918

At the second visit the level of stigma in the control group remained the same as at the baseline. The level of stigma was lower than at the baseline in the program group according to BSDS and CAMI (t-test for the means p<0.01). There was no difference between CAMI scores in the computer, reading and control groups (according to ANOVA, p=0.31), however there was the difference in BSDS scores, with the lowest level of stigma in the program group (ANOVA p=0.03). Only in the program group the level of stigma in six months was significantly lower than at baseline (t-test, two-sample assuming equal variances CAMI p<0.001, BSDS p=0.02). So, only in the program group the results of the intervention were sustainable in six months.

Discussion

Even in this sample of relatively highly and professionally educated people, misconceptions and stigmatizing beliefs about psychiatric disorders were very widespread. Both computer-assisted and conventional education were effective in reducing psychiatric stigma short-term, however in six months only effect of computer-mediated intervention was sustained. We think that professionals who may interact with people with mental health problems or other stigmatized groups can benefit from such interventions. There is a need to provide anti-stigma interventions for students who will potentially work with psychiatric patients.

Whether the effect of the intervention is stable over longer period of time and if the intervention is effective in changing subjects behavior in real life (not a positive effect on students behavior) remains to be established.