The authors have declared that no competing interests exist.
Conceived and designed the experiments: AF LT JK AM KL OT. Performed the experiments: AF AM. Analyzed the data: AF LT JK. Wrote the paper: AF LT JK AM KL OT.
In general, assessment tools for stigma in mental disorders such as attention deficit hyperactivity disorder (ADHD) are lacking. Moreover, misbeliefs and misconceptions about ADHD are common, in particular with regard to the adult form of ADHD. The aim of the present study was to develop a questionnaire measuring stigma in adults with ADHD and to demonstrate its sensitivity.
A questionnaire initially containing 64 items associated with stigma in adults with ADHD was developed. A total number of 1261 respondents were included in the analyses. The psychometric properties were investigated on a sample of 1033 participants. The sensitivity of the questionnaire was explored on 228 participants consisting of teachers, physicians and control participants.
Thirty-seven items were extracted due to exploratory factor analysis (EFA) and the internal consistency of items. Confirmatory factor analysis (CFA) revealed good psychometric properties of a 6-factor structure. Teachers and physicians differed significantly in their stigmatizing attitudes from control participants.
The present data shed light on various dimensions of stigma in adult ADHD. Reliability and Social Functioning, Malingering and Misuse of Medication, Ability to Take Responsibility, Norm-violating and Externalizing Behavior, Consequences of Diagnostic Disclosure and Etiology represent critical aspects associated with stigmatization.
The core symptoms of attention deficit hyperactivity disorder (ADHD), namely inattention, hyperactivity and impulsivity, result in highly externalized behaviors. This externalized behavior can be easily recognized by the environment and may induce misperceptions and misunderstandings about the condition. Moreover, public perceptions concerning ADHD have been shown to be foremost tied to the impression that ADHD is a disorder mainly seen in white middle-class boys suffering preeminent from symptoms of hyperactivity
Empirical research on stigmatization revealed that not only physical deviances can set individuals apart and trigger stigmatization. Intrinsic characteristics of the individual such as behavioral deviance have also been found to provoke stigma
Research demonstrated that stigmatization of individuals with ADHD has adverse consequences leading to diminished self-esteem and self-efficacious beliefs and ultimately a reduced quality of life
Referring to courtesy-stigma, families, relatives or the social network of the target individual with ADHD can also be affected by stigma. For example, more than three-fourths of parents of children with ADHD reported to have encountered stigmatizing situations
Considering the impact of stigmatization on the various facets of an individual’s life, the sporadic empirical investigation and the shortness of valid assessment tools measuring stigmatization in ADHD is surprising. This is in accordance with findings by Angermeyer and Dietrich
Only a few studies made use of survey responses, most likely because of a lack of psychometrically proven questionnaires for the measurement of stigmatization in ADHD. Kellison and colleagues
In conclusion, despite stigmatization in ADHD being an important issue, there is a considerable lack of knowledge. Available measures primarily focus on children or do not distinguish between children, adolescents and adults with ADHD. Since it has only recently been acknowledged that ADHD is a condition that can continue from childhood to adulthood, it can be assumed that the public’s knowledge on ADHD in adults is more limited than their knowledge about childhood ADHD. Therefore studies on and measures for the assessment of stigmatization in adult ADHD are of particular significance. The aim of the present study is to enhance knowledge and conceptual clarity on stigmatization in adults with ADHD. A new questionnaire consisting of statements specifically designed to meet public beliefs and perceptions of ADHD in adulthood is developed. Psychometric properties of the questionnaire are explored and the sensitivity of the questionnaire is assessed by investigating differences between stigma responses of individuals with specific knowledge on ADHD (i.e. teachers, physicians) and individuals without specific knowledge.
Data were obtained from a total of n = 1261 participants who completed a questionnaire measuring stigma responses towards adults with ADHD. Psychometric properties were explored on a sample of 1033 respondents. 439 of the 1033 respondents were first-year undergraduate psychology students of the University of Groningen, The Netherlands. The remaining participants were recruited via public announcements, word-of-mouth and through contacts of the researchers involved. Respondents’ age ranged from 17 to 79 years with a mean age of 31.3 years (SD = 14.8 years). Mean level of education was 16.4 years (SD = 1.2 years). The sample consisted of 66.0% female and 31.3% male participants with 2.7% missing information. Only 1% of the total sample claimed of never having heard about ADHD and 62.6% of participants stated to know an adult diagnosed with ADHD. On a scale of self-rated knowledge about ADHD ranging from 0 (no knowledge at all) to 10 (expert knowledge), the average score was 4.5 (SD = 2.1). To perform an exploratory factor analysis (EFA) with a subsequent confirmatory factor analysis (CFA), the sample was split into two subsamples (i.e. Subsample 1 and Subsample 2). The allocation of participants to samples has been performed randomly by applying the option of a random selection of cases in SPSS 18. The two subsamples did not differ with regard to their descriptives (
Subsample 1 | Subsample 2 | Total sample | |
N | 516 | 517 | 1033 |
Sex |
344/158 | 338/165 | 682/323 |
Age |
31.3±14.9 | 31.3±14.7 | 31.3±14.8 |
Education |
16.4±1.2 | 16.4±1.2 | 16.4±1.2 |
A further set of data (n = 228) was collected to demonstrate the sensitivity of the questionnaire by exploring stigma responses of a group of teachers and a group of physicians in comparison to a control group. 77 teachers and 74 physicians were recruited via public announcements, word-of-mouth and through contacts of the researchers involved. All teachers had successfully completed a university study program for teachers and were currently working as teachers for primary or secondary schools. All physicians completed successfully a medicine study program and were currently working as physicians in the health service. A control group (n = 77) was recruited with similar characteristics with regard to age, gender and educational level (all respondents completed high school and received additional training or obtained a university degree) (
Teachers | Physicians | Controls | |
N | 77 | 74 | 77 |
Sex |
44/33 | 42/32 | 44/33 |
Age |
52.0±9.6 | 50.6±12.7 | 52.3±10.9 |
Based on (I) an extensive literature study on social perceptions, myths and stigma in ADHD, (II) the personal clinical experience with adults with ADHD of the researchers involved in this study and (III) patient interviews about their experiences, a deductive approach was applied for initial item generation
All participants were invited to take part in the study on a voluntary basis. Participants received no reward for participation with the exception of undergraduate students who were credited toward a research requirement. The time to complete the questionnaire was estimated to take around twenty minutes. Participants were informed about the aim of the study and it was emphasized that all data will be analyzed anonymously. The study was approved by the ethics committee of the University of Groningen, The Netherlands.
The study was approved by the Ethical Committee Psychology (ECP) affiliated to the University of Groningen, the Netherlands. Before participating, all participants were informed about the aims of the study. Participants were required to read and acknowledge an information sheet prior to completion of the questionnaire. Formal written consent was not sought for adult participants; submission of completed questionnaires was taken as implied consent. Since participants with an age of 17 years were included (n = 2), written informed consent was obtained from these participants as well as their parents prior to inclusion.
Exploratory factor analysis (EFA) was performed on
Cronbach’s α was calculated for the total scale and the subscales (factors) of the questionnaire as a measure of internal consistency.
To replicate the proposed factor structure model by the EFA on
The factor structure was examined by the following goodness-of-fit statistics: Chi-Square value with corresponding p-value, normed Chi-Square (χ/df), Root Mean Squared Error of Approximation (RMSEA), 90%-confidence interval of the RMSEA, Standardized Root Mean Square Residual (SRMR) and Comparative Fit Index (CFI).
The
The
The
The
The goodness-of-fit statistics of the factor model as proposed in EFA of the present study were compared to the cut-offs and recommendations as cited above. Additionally, fit statistics of the multitrait model were compared to the fit statistics of a single common factor model which served as a competing model. Once the overall fit of the model has been evaluated, each model coefficient was individually examined for its degree of fit. This was achieved by t-tests testing the null hypothesis that the true values of specified factor loadings are zero. All items with non-significant factor loadings (p≤.05) were deleted from further analysis
Stigmatization was analyzed in descriptive statistics on the sample (n = 1033) on the basis of the extracted factor structure. Subsequently, a dependent sample ANOVA with post-hoc pairwise comparisons was performed to assess differences in stigmatization between subscales. The effects of gender and age on stigma were assessed with t-tests for independent samples and Pearson product-moment correlations, respectively. Statistical tests were calculated separately on extracted stigma subscales which led to α-accumulation. To counteract the problem of multiple comparisons, the significance level α was adjusted by using a Bonferroni correction. Moreover, effect sizes (Cohen’s d) were calculated for all comparisons. Following Cohen’s guidelines for interpreting effect sizes
Stigmatization of teachers, physicians and control participants was explored on the basis of the extracted factor structure. A multivariate analyses of variance (MANOVA) with post-hoc pairwise comparisons (Scheffé) was calculated to compare stigmatization between the three groups on each subscale and on the total scale. Moreover, effect sizes (Cohen’s d) were calculated for all comparisons. Following Cohen’s guidelines for interpreting effect sizes
Psychometric properties were assessed by applying EFA on
Factor loadings on each factor | ||||||
Item | 1 | 2 | 3 | 4 | 5 | 6 |
15. Adults with ADHD care less about other’s problems. |
|
.038 | .07 | .04 | .080 | <.001 |
*17. Adults with ADHD are able to take care of a group of children in kindergarten. |
|
.07 | .30 | .15 | .16 | .06 |
25. You cannot rely on adults with ADHD. |
|
.20 | .21 | .26 | .08 | .33 |
27. Adults with ADHD are self-focused and egoistic. |
|
.29 | .16 | .26 | .03 | .31 |
*28. I would go on a date with someone with ADHD. |
|
.01 | .26 | <.01 | .11 | .06 |
*32. Adults with ADHD have no problems in making friends. |
|
.05 | .07 | .04 | .08 | <.01 |
33. Adults with ADHD are less successful than adults without ADHD. |
|
.24 | .16 | .30 | .30 | .17 |
* 35. Adults with ADHD are able to lead a group of people. |
|
.05 | .20 | .07 | .08 | .03 |
36. Under medication, adults with ADHD are less trustworthy. |
|
.33 | <.01 | .08 | .01 | .22 |
3. Many adults with ADHD simulate the symptoms. | .02 |
|
.13 | .04 | .02 | .18 |
4. Adults with ADHD misuse their medication (sell it to others, take too much…) | .16 |
|
.17 | .03 | .04 | <.01 |
5. ADHD is invented by drug companies to make profit. | .02 |
|
.13 | .11 | .04 | .12 |
7. Many adults with ADHD exaggerate their symptoms in order to be medicated. | .07 |
|
.11 | .03 | .09 | .19 |
9. ADHD is a childhood disorder and not seen in adults. | .20 |
|
.06 | .03 | .05 | <.01 |
10. Adults with ADHD lie more often than adults without ADHD. | .04 |
|
.22 | .37 | .07 | .11 |
11. Adults with ADHD have a lower IQ than adults without ADHD. | .17 |
|
.10 | .24 | .08 | .11 |
30. Many adults pretend to have ADHD just to get access to medication. | .28 |
|
.16 | .07 | .03 | .21 |
31. Adults with ADHD are less able to give advice. | .32 |
|
.29 | .34 | .14 | <.01 |
1. Adults with ADHD are bad parents and have problems with raising children. | .06 | .25 |
|
.18 | .09 | .29 |
2. I would mind if my investment advisor had ADHD. | .08 | .20 |
|
.06 | .15 | .13 |
*14. I would not mind if a doctor who has ADHD treated me. | .28 | .04 |
|
.05 | .02 | .18 |
26. If I had a business, I would not hire a person with an ADHD diagnosis. | .13 | .23 |
|
.22 | .04 | .05 |
29. I would mind if the teacher of my children had ADHD. | .21 | .17 |
|
.23 | .03 | .20 |
12. Adults with ADHD are more often involved in traffic errors. | .07 | .25 | .10 |
|
.20 | .05 |
18. I could tell when a person around me has ADHD. | .14 | <.01 | .07 |
|
.06 | .04 |
19. Adults with ADHD act without thinking. | .13 | .13 | .09 |
|
.06 | .04 |
20. Adults with ADHD have a different sense of humor than adults without ADHD. | .03 | .11 | .23 |
|
.14 | .08 |
37. Adults with ADHD cannot deal with money. | .02 | .34 | .18 |
|
.18 | .25 |
6. People’s attitudes about ADHD make persons with ADHD feel worse about themselves. | .21 | −.01 | <.01 | <.01 |
|
.04 |
8. Adults with ADHD are of lower social status. | .08 | .19 | .28 | .07 |
|
<.01 |
13. As a rule, adults with ADHD feel that telling others that they have ADHD was a mistake. | .11 | <.01 | .28 | .04 |
|
.08 |
21. Adults with ADHD have a lower self-esteem than adults without ADHD. | .25 | <.01 | .01 | .31 |
|
.12 |
24. Adults with ADHD feel excluded from society. | .15 | .08 | .05 | .12 |
|
.08 |
16. ADHD is caused by bad parenthood. | .37 | .31 | .02 | .05 | .05 |
|
22. Extensive exposure to video games and TV shows can cause ADHD. | .17 | .08 | .06 | .08 | .07 |
|
23. Adults with ADHD do not engage enough in sports. | .16 | .28 | .13 | .17 | .04 |
|
34. ADHD is a consequence of childhood trauma. | .13 | −.26 | .16 | .14 | <.01 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
N = 516; Cronbach’s α = 0.91; * inversed items; The item numbers reflect the relative position of the items in the original questionnaire.
The overall scale reliability (internal consistency) of the 37 items was high (Cronbach’s α = .91). The scale reliabilities of the six subscales (Cronbach’s α) ranged between.61 and.87, with.60 indicating the minimum acceptable and 0.80 indicating good reliability.
Subscales | Number of items | Cronbach’s α (range if items deleted) |
1. Reliability and Social Functioning | 9 | 0.87 (0.84–0.86) |
2. Malingering and Misuse of Medication | 9 | 0.81 (0.78–0.80) |
3. Ability to Take Responsibility | 5 | 0.74 (0.66–0.76) |
4. Norm-violating and Externalizing Behavior | 5 | 0.61 (0.52–0.61) |
5. Consequences of Diagnostic Disclosure | 5 | 0.65 (0.56–0.62) |
6. Etiology | 4 | 0.71 (0.60–0.65) |
Regarding the overall fit of the model, the Chi-Square statistics led to rejection of the model (χ2 (614) = 1763.68; p<.01). More crucial, the normed Chi-Square value was within a range indicating a good fit (χ2/df = 2.87). Both a RMSEA of.06 and its 90%-confidence interval [.057;.064] also pointed to a well-fitting model. The SRMR of.07 was below the recommended cut-off of.08 and therefore further supported the model fit. Finally, the incremental fit index (CFI = .93) also revealed an acceptable model fit. In summary, the CFA resulted in a satisfactory fit for the present 6-factor model. The proposed 6-factor model outperformed a single common factor model (χ2 (629) = 2680.78 p<.01; χ2/df = 4.3, RMSEA = .099, 90%-CI for RMSEA = [.096;.100]; SRMR = .082; CFI = .87) in all indices. To investigate each model coefficient individually, t-values for factor loadings were inspected. Significant loadings were found for each item (p<.01). Therefore, no items had to be excluded.
A dependent sample ANOVA indicated significant overall differences in stigmatization between the subscales (F(5;5095) = 295.2; p<.001). Post-hoc pairwise comparisons revealed significant differences between all pairs of subscales with the exception of the comparison between subscale 1 and subscale 4 and between subscale 3 and subscale 6 (
Subscale 1: Reliability and Social Functioning; Subscale 2: Malingering and Misuse of Medication; Subscale 3: Ability to Take Responsibility; Subscale 4: Norm-violating and Externalizing Behavior; Subscale 5: Consequences of Diagnostic Disclosure; Subscale 6: Etiology. Higher stigma scores indicate increased stigmatizing beliefs; All subscales differ significantly (p<.05) from each other with the exception of the comparison between 1 and subscale 4 and between subscale 3 and 6 (p>.05).
Subscale | 2 | 3 | 4 | 5 | 6 |
|
0.66 | 0.36 | 0.04 | 0.38 | 0.44 |
|
* | 0.27 | 0.67 | 1.10 | 0.21 |
|
* | 0.34 | 0.77 | 0.06 | |
|
* | 0.50 | 0.37 | ||
|
* | 0.75 |
Subscale 1: Reliability and Social Functioning; Subscale 2: Malingering and Misuse of Medication; Subscale 3: Ability to Take Responsibility; Subscale 4: Norm-violating and Externalizing Behavior; Subscale 5: Consequences of Diagnostic Disclosure; Subscale 6: Etiology.
For the analysis of the effects of gender and age, a Bonferroni adjusted significance level of p = .0083 was applied because of multiple comparisons/correlations (6 subscales and total score). Stigma scores of female respondents (n = 682; mean age = 30.1 years, SD = 14.3) were compared to the scores of male respondents (n = 323; mean age = 34.5 years, SD = 15.8) using t-tests for independent samples (
Stigma scale | |||||||
1 | 2 | 3 | 4 | 5 | 6 | Total scale | |
−0.76±1.11 | −1.35±0.92* | −1.02±1.16* | −0.81±1.03 | −0.28±1.11 | −1.15±1.22* | −0.91±0.79 | |
−0.85±1.22 | −1.57±0.78 | −1.31±1.01 | −0.86±0.92 | −0.23±0.95 | −1.36±1.18 | −1.03±0.68 |
Subscale 1: Reliability and Social Functioning; Subscale 2: Malingering and Misuse of Medication; Subscale 3: Ability to Take Responsibility; Subscale 4: Norm-violating and Externalizing Behavior; Subscale 5: Consequences of Diagnostic Disclosure; Subscale 6: Etiology.
Stigma responses of teachers, physicians and control participants are presented in
Subscale 1: Reliability and Social Functioning; Subscale 2: Malingering and Misuse of Medication; Subscale 3: Ability to Take Responsibility; Subscale 4: Norm-violating and Externalizing Behavior; Subscale 5: Consequences of Diagnostic Disclosure; Subscale 6: Etiology. a significant difference between teachers and control participants on p<.05. b significant difference between physicians and control participants on p<.05.
Teachers(n = 77) | Physicians(n = 74) | Controls(n = 77) | |
Subscale 1 | −1.27±0.87 | −1.11±0.82 | −0.85±0.56 |
Subscale 2 | −1.98±0.66 | −1.87±0.70 | −1.58±0.60 |
Subscale 3 | −1.07±1.27 | −1.14±1.24 | −1.11±1.10 |
Subscale 4 | −1.13±0.97 | −0.85±0.97 | −0.75±0.92 |
Subscale 5 | −0.07±0.90 | 0.03±1.15 | −0.21±1.12 |
Subscale 6 | −1.42±0.93 | −1.65±1.04 | −1.23±0.98 |
Total Scale | −1.16±0.64 | −1.10±0.70 | −1.03±0.59 |
Subscale 1: Reliability and Social Functioning; Subscale 2: Malingering and Misuse of Medication; Subscale 3: Ability to Take Responsibility; Subscale 4: Norm-violating and Externalizing Behavior; Subscale 5: Consequences of Diagnostic Disclosure; Subscale 6: Etiology.
Teachers vs. Physicians | Teachers vs. Controls | Physicians vs. Controls | |
Subscale 1 | 0.19 | 0.57* | 0.37 |
Subscale 2 | 0.16 | 0.63* | 0.44* |
Subscale 3 | 0.06 | 0.03 | 0.03 |
Subscale 4 | 0.29 | 0.40* | 0.11 |
Subscale 5 | 0.10 | 0.14 | 0.21 |
Subscale 6 | 0.23 | 0.20 | 0.42* |
Total Scale | 0.09 | 0.21 | 0.11 |
Subscale 1: Reliability and social functioning; Subscale 2: Malingering and misuse of medication; Subscale 3: Ability to take responsibility; Subscale 4: Norm-violating and externalizing behavior; Subscale 5: Consequences of diagnostic disclosure; Subscale 6: Etiology. * significant at p<.05.
Empirical research directly addressing stigma in ADHD is sparse. One difficulty in studying this topic is the lack of appropriate measures. Therefore, a new questionnaire on stigma in adults with ADHD has been developed in the present study. The aims of this development were (I) to identify dimensions that specifically meet stigmatizing issues persons with ADHD are confronted with, (II) to transfer the knowledge gained from research on childhood ADHD to the adult population, and finally (III) to examine the sensitivity of the questionnaire in measuring differences of stigmatizing attitudes between groups by comparing stigma responses of teachers, physicians and matched control participants.
The questionnaire consists of 37 items directly addressing stigmatizing beliefs on adults with ADHD. Exploratory factor analysis (EFA) revealed a 6-factor structure which was empirically confirmed by a confirmatory factor analysis (CFA). The devised 6-factor structure has been supported in overall fit indices and by inspecting each model coefficient individually. Furthermore, the 6-factor structure has been shown to be superior to a single common factor model.
Research demonstrated that the mere label of a psychiatric condition can trigger stigmatization
Participants’ responses to statements that adults with ADHD are not reliable, less trustworthy, self-focused, egoistic and careless about other’s problems have been found to load on a common factor which has been labeled
The factor
Another important issue in explaining the emergence of stigmatization in ADHD lies in the doubts concerning the existence of ADHD as a disorder
The factor labeled
Finally, a dimension has been identified concerning the causes of the condition. This factor has been labeled
The present study did not reveal a factor focusing on medication use. This might appear surprising, since misperception and stigmatizing beliefs about long-term effects of medication in ADHD were repeatedly reported
The overall level of stigma was found to be low to moderate as reflected in negative values for all subscales. Even though the absolute value is difficult to interpret due to lacking reference values, conclusions can be drawn on the basis of comparisons of stigmatizing responses between subscales. Except of two comparisons (between subscale 1 and subscale 4 and between subscale 3 and subscale 6), stigma responses on all subscales differ significantly from each other. Effect sizes between subscales ranged from negligible to large effects, supporting the notion that different dimensions of stigmatization towards adults with ADHD were measured. The lowest stigma responses could be shown on subscale 2 (
In line with previous research
When studying stigmatization towards a developmental disorder such as ADHD, people working in the educational or health care sector are of particular interest. Teachers, on the one hand, received a specialized training in educational sciences and have contact to a wide range of youngsters and accompany them in their development from childhood through adolescence into adulthood. Therefore, teachers can be assumed to be more sensitive towards developmental disorders such as ADHD and to be less prone to stigmatizing attitudes towards those affected by these disorders. Physicians, on the other hand, successfully completed a university study program in medicine and have a broad understanding of factors underlying human behavior. Moreover, stigma is a concept that is quite salient in medicine, since individuals with many different physiological, psychosomatic or psychological conditions experience stigmatization. Consequently, physicians have a higher chance to get in contact with people experiencing stigmatization than the general population. Furthermore, many physicians might even be informed by their professional associations, colleagues, conferences or any other additional training about the existence of ADHD in adulthood and the problems involved. Therefore, physicians are presumably also more sensitive with regard to stigmatizing beliefs and attitudes. However, analysis of the present data revealed that teachers, physicians and control participants did not differ on the overall level of stigmatization. Nevertheless, teachers and physicians could be found to show lower scores than control participants on certain aspects of stigma. Significantly lower stigma responses were found for teachers in statements focusing on
In the present study, a new questionnaire with six subscales assessing stigmatization in adults with ADHD has been described. The psychometric properties of these subscales were carefully evaluated and the factor structure was replicated as suggested by Hinkin