Parents’ Stigmatizing Attitudes Toward Psychiatric Labels for ADHD and Depression

Participants were 225 parents of children in elementary school living in Queensland or New South Wales, Australia, who were recruited through advertisements in school newsletters, community billboards, and word of mouth. The parents’ mean age was 39.17±5.42 years, and 200 (89%) were female. They had 1.73±.80 children aged 7.44±1.88 years. Participants’ ethnicity was Australian (N=207, 92%), New Zealander (N=6, 3%), Indian (N=4, 2%), North American (N=4, 2%), English (N=2, 1%), and Pacific Islander (N=2, 1%). Annual family income ranged from under $20,000 to over $150,000, and education ranged from less than high school to postgraduate.

Each participant read three vignettes (control, symptoms only, and symptoms plus label) and then completed three measures of stigma. The vignettes were modeled after prior research (10,11) and contained 130–144 words. The control vignette described a child who displayed developmentally typical depression (mild, nonimpairing sadness) or ADHD (mild, nonimpairing inattention). It was intended to assess parents’ attitudes toward a typically developing child, not an unrealistically perfect child.

The vignettes describing symptoms only and symptoms plus label depicted a child who met DSM-IV-TR symptom criteria (13). The ADHD vignette described a child who met criteria for ADHD, combined type (six or more symptoms of inattention and six or more symptoms of hyperactivity-impulsivity). The depression vignette described a child who displayed four symptoms required for diagnosis of major depressive episode. The vignettes describing symptoms plus label also included the statement, “[Child’s name] has recently been diagnosed with [attention-deficit hyperactivity disorder (ADHD) or major depressive disorder (depression)].”

Participants rated the child in each vignette on a stereotypes subscale, a prejudice subscale, and a social distance subscale. The subscale items were derived from existing measures of stigma (10,11) and developed on the basis of theory and a literature review. The stereotypes subscale asked participants to evaluate the likelihood that the child was aggressive, was dangerous, tried to avoid hard work, functioned poorly, and lived in a stressful home. Possible ratings ranged from 1, not at all likely, to 7, extremely likely. Items had good internal consistencies for each of the three vignettes (Cronbach’s α=.70–.74).

The prejudice subscale asked participants to rate their likelihood of feeling anxious, compassionate (reverse scored), and irritated toward the child. Possible ratings ranged from 1, not at all likely, to 7, extremely likely. Internal consistencies were very good for the three vignettes (Cronbach’s α=.80–.83).

The social distance subscale asked participants to rate their willingness to move next door to the child, encourage a friendship between the child and their own child, spend an evening socializing with the child’s family, have a child in the same classroom as the child, invite the child to play at their house, and involve a child in a group project with the child. Possible ratings ranged from 1, definitely unwilling, to 7, definitely willing. Scores were reversed so that higher scores would reflect greater social distance, which keeps interpretation of the subscales consistent (higher scores indicate more stigma). Internal consistencies were excellent for the three vignettes (Cronbach’s α=.92–.93).

This study was approved by our university’s ethics committee. A Web-based software program was used to distribute the survey online from 2010 to 2011. Participants were presented with an information sheet that described the study, along with buttons to provide written consent or decline participation. The three vignettes were presented randomly by the child’s gender (only boys or only girls), the order of the control and symptoms-only vignettes (vignettes with a label attached always came last to avoid sensitizing participants’ ratings for symptoms-only vignettes), and the diagnostic label (the label was placed on each symptom vignette half the time to avoid the effect of label being due to a particular symptom vignette). Participants were free to skip questions if they desired. Upon completion, participants could enter their e-mail address for a prize drawing.

A 3×2×2 (vignette type × disorder type × gender) repeated measures analyses of variance was run for each subscale. Vignette type is a within-participants factor, and disorder type and gender are between-participants factors. A Bonferroni correction yielded an alpha level of .017.

Results for each subscale are shown in Table 1. For stereotypes, there was a significant main effect of vignette type (F=530.60, df=2 and 217, p<.001, η²=.83). Participants held significantly more stereotypes toward the vignettes describing symptoms only (t=31.22, df=1,223, p<.001, d=2.69) and symptoms plus label (t=29.00, df=1,221, p<.001, d=2.76) than toward the control vignettes. Stereotypes were significantly greater toward the vignettes describing symptoms plus label versus symptoms only (t=2.54, df=1,221, p<.012, d=.17).

For prejudice, there was a significant main effect of vignette type (F=154.11, df=2 and 219, p<.001, η²=.59). Prejudice ratings were significantly higher for vignettes describing symptoms only (t=15.20, df=1,223, p<.001, d=1.19) and symptoms plus label (t=17.14, df=1,223, p<.001, d=1.43) than for the control vignettes. Prejudice was significantly greater toward the vignettes describing symptoms plus label versus symptoms only (t=4.67, df=1,223, p<.001, d=.23). There was a significant main effect of disorder type (F=5.73, df=1 and 220, p<.017, η²=.025), with higher mean scores for prejudice for vignettes describing children with ADHD (3.09±1.50) versus depression (2.74±1.43).

For social distance, there was a significant main effect of vignette type (F=134.88, df=2 and 217, p<.001, η²=.55). Social distance ratings were significantly higher for vignettes describing symptoms only (t=14.05, df=1,221, p<.001, d=.97), and symptoms plus label (t=14.61, df=1,221, p<.001, d=1.16) than for the control vignettes. Social distance was significantly greater toward the vignettes describing symptoms plus label versus symptoms only (t=2.94, df=1,223, p<.04, d=.12). There was a significant effect of disorder type (F=24.87, df=2 and 217, p<.001, η²=.10), with higher mean scores for social distance for vignettes describing children with ADHD (3.20±1.39) versus depression (2.48±1.27). There was also a significant interaction of vignette type and disorder type (F=11.93, df=2 and 217, p<.001, η²=.10), indicating there was no difference in levels of social distance toward children with ADHD and depression compared with children in the control vignettes. However, social distance among children with ADHD was greater than among children with depression for vignettes that described symptoms only (t=4.65, df=1,222, p<.001, d=.61) and symptoms plus label (t=5.65, df=1,222, p<.001, d=.70).

The primary aim of this research was to evaluate the extent to which a psychiatric label would worsen stigma associated with ADHD and depression, as is often feared (1–5). As expected, even when a diagnosis was not disclosed, parents held high levels of stigma toward children with ADHD or depression symptoms. Including a diagnostic label with a description of symptoms led to small but consistent additional increases in parents’ stereotypes, prejudice, and social distance. This implies that parents experience highly stigmatizing reactions when they encounter a child with symptoms of ADHD or depression; telling the parents that the child “has ADHD” or “has depression” has a small effect in furthering these stigmas. These findings suggest that the small increase in stigma arising from attaching a psychiatric label to symptoms may well be outweighed by the benefits that psychiatric diagnoses can provide, namely, treatment access, validation, and feelings of relief (6).

The consistent but relatively small magnitude of the labeling effect invites speculation about its origins. One likely option is that parents correctly guess a child’s diagnosis when confronted with just the symptoms (10). This process may be sufficient to trigger their preexisting beliefs and facilitate categorization almost as much as presenting the label directly (14). Put in real-world terms, many parents may assume that a child with particularly high levels of hyperactivity “has ADHD,” thus triggering associated negative stigma.

Another interesting aspect of our results is that parents reported more prejudice and social distance toward children with ADHD than depression. Past studies of adults in the general community have shown the reverse, although those depression vignettes intimated suicidal ideation (10,11), whereas our vignettes did not. It may be that suicidality is a key symptom that engenders stigma. Parents may be more concerned about keeping a child away from peers who have difficulty controlling their activity (ADHD) than from those who feel sad (depression) but are not suicidal.

One limitation of our study was that we did not include a “label-only” vignette and, therefore, were unable to observe the effect of diagnostic status, in the absence of any symptom information, on parents’ stigma. In addition, the sample consisted primarily of mothers of European descent. However, we did not find significant differences on the basis of participant gender, and past research in this area has shown that demographic variables are weak, often nonsignificant, predictors of stigma (11). Another concern is social desirability, the tendency to present one’s self in an unduly positive light, which could result in underreporting stigma to look more tolerant (15). If social desirability influenced our results, it indicates that stigma is even greater than reported here. Finally, although aligned with previous work, the vignette methodology may have limited ecological validity. This must be weighed against the benefits of vignettes, which provide control over extraneous variables, such as symptom severity and gender.

Despite fears about the potential for children to be stigmatized by a label arising from psychiatric diagnosis (1–5), little empirical research has addressed whether a psychiatric label adds to the stigma associated with symptoms alone. The effect of a label was reliable across all measures of stigma (stereotypes, prejudice, and social distance), although the size of the effect may not be large enough to justify the high degree of concern about labeling that is sometimes expressed. These results should not be interpreted to mean that psychiatric labels do not lead to stigma; rather, the stigma that parents attach to a child with a psychiatric diagnosis is primarily due to the symptoms of the disorder, not to the label. Therefore, parents may find that the advantages of seeking help for a child—in the form of validation, support, and treatment access—outweigh the disadvantages of stigma due to disclosing the diagnosis to other parents. Further work to replicate these findings and extend them to other populations and disorders will help to address fears of labeling.

This research was supported by a grant from Women in Research to Dr. Ohan.

The authors report no competing interests.