Attitudes and beliefs of the French public about schizophrenia and major depression: results from a vignette-based population survey

Persons aged 16–65 years old, of French nationality, were drawn from an established market research panel. The panel is recruited and managed under the highest standards (‘MRA Verified’ seal, adhering to CASRO and ESOMAR guidelines). Persons were contacted by email and invited to participate in an on-line survey between January and March 2012. If a person did not respond to the initial contact, he or she was contacted again three days later. Recruitment continued until 1600 interviews were obtained. Participants were aware that their responses would be used for scientific research, specifically commissioned by the authors. To ensure that the sample was representative of the general adult population of France, sampling was stratified for place of residence, gender, age, and family status. In total, 1,600 persons were interviewed. The socio-demographic characteristics of the sample reflect fairly well the socio-demographic composition of the general population in France [11]: 50% were men (France: 50.6%); 17.5% of respondents were 16–24 years old, 20.2% 25–34 years old, 22.1% 35–44 years old, 21.3% 45–54 years old, and 18.9% 55–65 years old (France: 17.1%, 19.1%, 20.9%, 21.1%, and 21.7%, respectively). 43.1% of respondents were single (France: 45.8%). The educational level of respondents was as follows: 17.3% CAP/BEP, 21.7% bachelor, and 61.0% superior or other (France: 23.4%, 17.1%, and 59.5%, respectively). Informed consent was considered to have been given when individuals agreed to complete the interview.

The fully-structured interview had originally been developed in Germany and had been successfully employed there in several surveys [4, 12]. For use in this study, it was translated into French following the guidelines of WHO [13]. At the beginning of the interview respondents were presented with a vignette of a diagnostically unlabeled psychiatric case history, depicting a case of either schizophrenia or major depressive disorder. The symptoms described in the vignettes that had originally been prepared for the German surveys fulfilled the criteria of DSM-III-R for the respective disorder [14]. Each vignette had been independently rated by five experts on psychopathology masked to actual diagnosis, providing confirmation of the correct diagnosis for each case history. In order to be able to compare our findings with those from Germany we decided to use the same vignettes. The sex of the individual presented in the vignettes was randomly varied. Respondents were randomly allocated to receive either the vignette depicting schizophrenia or the vignette depicting major depressive disorder.

After the assessment of respondents’ mental health literacy, questions on their attitudes towards the person depicted in the vignette were asked. According to Hayward and Bright [15], the four most important stereotypes about mental illness are that people with mental disorders are unpredictable, that they are dangerous, that they are responsible for their illness, and that mental illnesses are hard to treat. Agreement or disagreement with these stereotypes was recorded with the help of a five-point Likert scale, anchored with 1 = ‘fully agree’ and ‘5’ ‘totally disagree’.

Emotional reactions to the person described in the vignette were assessed by means of nine five-point Likert-scaled items, anchored with 1 = ‘applies completely’ and 5 = ‘does not apply at all’ , representing the main three dimensions fear, anger, and pro-social reactions: ‘I feel the need to help him/her’ , ‘I feel pity’ , ‘I feel uncomfortable’ , ‘He/she makes me feel insecure’ , ‘He/she scares me’ , ‘I feel sympathy for him/her’ , ‘I feel annoyed by him/her’ , ‘I react angrily’ , ‘I am amused by something like that’. Respondents who endorsed the two points on either side of the mid-point of the scale (‘undecided’) were grouped together to the categories ‘agree’ , or ‘disagree’ , respectively [16].

For the assessment of respondents’ desire for social distance we used the scale developed by Link et al [17]. This scale encompasses the following social situations: rent a room, work together, have as a neighbor, let take care of a little child, have marry into family, introduce to friends, recommend for a job. With the help of a five-point Likert scale respondents could indicate to what extent they were willing or unwilling to engage in the proposed relationships. Respondents who endorsed the two points on either side of the mid-point of the scale (‘undecided’) were grouped together into the categories ‘accept’ , or ‘reject’ , respectively.

In order to examine whether respondents reacted differently with regard to the schizophrenia vignette and the depression vignette, we calculated separate multinomial logit regressions for each item. To adjust for potential differences between samples for the influence of demographic factors, the regression analyses controlled for respondents’ gender, age, and educational attainment. To illustrate the magnitude of differences between the subsamples presented with the schizophrenia or the depression vignette, discrete probability changes were calculated for all items and each response category with control variables held at their means for the whole sample. A discrete change coefficient is the difference in the predicted probability of a given outcome between both subsamples; it serves as an indicator of the effect size of the change. Ninety-five percent confidence intervals were computed with the delta method. The calculation of probability changes was carried out using SAS 9.3 [18]. To make adjusted predictions comparable to unadjusted survey results, probabilities and discrete changes are multiplied by 100 and can be read as percentage of respondents choosing any answer category. Tables 1, 2, 3, 4, 5, 6, 7, 8 and 9 show, separately for schizophrenia and major depressive disorder, the raw percentage and the predicted probability for each item/answer category plus the difference in probabilities between both disorders and the confidence interval for this difference.

A clear pattern emerges from our findings: Schizophrenic symptoms are defined by the vast majority of respondents as expression of mental illness, caused by biogenetic factors and current stress. In contrast, depressive symptoms are less unanimously defined as mental illness, and current stress plays here a dominant causal role. The prognosis of major depressive disorder appears more favorable than that of schizophrenia. In the eyes of the public, psychiatrists represent the most appropriate helping source for people suffering from schizophrenic symptoms, and, apart from psychotherapy, medication is recommended most frequently for treatment. For depressive symptoms, general practitioners are considered most frequently as a source for help; taking a health cure or seeing a practitioner in complementary medicine are less likely to be rejected; apart from formal help, there is also a strong emphasis on self-help and social support. In the eyes of the public, relaxation techniques and ‘alternative’ methods are better suited for the treatment of major depression than psychotropic medication. Thus, although there exists a certain overlap between both disorders, the French public seems to draw a clear line between schizophrenia and major depressive disorder: on one hand schizophrenia as an illness with a strong biological component, demanding psychiatric treatment, on the other hand major depression which is seen mainly as a consequence of the exposure to psycho-social stress, benefitting also from alternative treatments.

In 2011, a similar survey had been conducted in Germany, using the same interview [12]. Since the interview mode was different (in France on-line, in Germany face-to-face) a direct comparison of results for schizophrenia and depression between both studies seems problematic [19]. However, it is legitimate to examine within each study the differences between both disorders and then contrast them across studies [20]. Apart from many similarities, there were some interesting differences. The divide between both disorders regarding the endorsement of current stress as a cause was more pronounced in France than in Germany. For instance, in France work-related stress was seen as a cause of depression by 80% and as a cause of schizophrenia by only 47%, while in Germany the respective percentages were 79% and 62%; thus, the difference between both disorders was larger in France than in Germany. Moreover, the difference between both disorders in relying on self-help was more pronounced in France (depression 50%, schizophrenia 29%) than in Germany (depression 59%, schizophrenia 47%). Finally, while in France ‘alternative’ methods like natural remedies or acupuncture were more frequently recommended for the treatment of depression (41% and 32%, respectively) than for the treatment of schizophrenia (22% and 21%, respectively), in Germany practically no difference was found between both disorders (27%/24% and 18%/17%) [12]. All these results suggest that the distinction made between schizophrenia and depression tends to be more marked in France than in Germany.

Irrespectively of the type of disorder, psychotherapy was the clear favorite. The acceptance of it was even more pronounced regarding the treatment of schizophrenia, which appears somewhat counterintuitive. This is not a French specialty. In the afore-mentioned German survey, 73% of respondents recommended psychotherapy for the treatment of depression and 84% for the treatment of schizophrenia. Similar findings have also been reported from previous studies (e.g. [21‐23]). The public’s reservation against psychotropic medication has also been observed in Germany. However, the contrast to the acceptance of psychotherapy was not as marked there as in France. For instance concerning the treatment of schizophrenia, 79% of respondents opted in France for psychotherapy and only 37% for medication, whereas in Germany the corresponding figures were 84% and 51%. A similar pattern was also found for depression (70%/21% versus 73%/36%). Interestingly, the difference between both countries in attitudes does not correspond with the actual use of psychotropic medication which is higher in France than in Germany [24]. The reasons for the aversion against psychotropic medication expressed by the French public are certainly manifold. One may be that while in both countries current stress was in sum more frequently endorsed than biogenetic causes, the preponderance of the first was slightly more pronounced in France. Regarding depression, for instance, the ratio between the sum of stress factors and the sum of biogenetic causes was twice as high in France as in Germany. Another reason may be that medication still tends to be seen by the public less as treatment of the real causes of an illness than psychotherapy, particularly psychoanalysis, which in France still enjoys relatively great popularity [25]. Finally, the widespread fear of getting addicted to medication may play a role, as the public is not able to sufficiently distinguish between drugs that have this unwanted effect, such as benzodiazepines, and others, such as antidepressants or antipsychotics, which do not have it [26].

Across all components studied, attitudes towards people with schizophrenia were more unfavorable than those towards people with major depressive disorder. People with schizophrenia more frequently were perceived as unpredictable and dangerous, there was a stronger need to separate ‘us’ from ‘them’ , they were more frequently met with fear and less frequently reacted to with pro-social feelings, and they faced more rejection. As with beliefs, the comparison with the German survey yields some interesting differences. For instance, concerning the endorsement of dangerousness, the discrepancy between both disorders was greater in France than in Germany (ratio schizophrenia/major depression in France 36%/14%, in Germany 23%/19%). While in France, 28% of respondents agreed with the statement that ‘we are all sometimes like this person’ in case of schizophrenia and 58% in case of major depression, in Germany, the difference between both disorders was smaller with 26% and 44%, respectively. While the difference between schizophrenia and major depression in the amount of fear expressed by respondents was in both countries the same, the difference in pro-social feelings in favor of major depression was more marked in France. Taken together, the discrepancy between attitudes towards people suffering from schizophrenia and major depressive disorder appears to be slightly more pronounced in France than in Germany. As already observed with illness beliefs [6], French people seem to distinguish between both disorders even more clearly than their German counterparts. While depression is perceived as something most people are familiar with through personal experience and as something which more or less belongs to normal life, people with schizophrenia are perceived as strange (by 60% of respondents as compared to only 16% with depression) and their behavior appears as incomprehensible (45% versus 21%). Another indication of the great familiarity of the French people with the notion of depression may be that almost all respondents (88%) identified the depressive symptoms depicted in the vignette as an expression of some sort of depression [27].

Before concluding, some limitations of our study should be mentioned. First, the focus on attitudes may be looked upon as a limitation since it allows predicting behavior with less than ideal accuracy. However, rather than using them as a proxy for individual behaviors, public attitudes can also be conceptualized at a collective level as a reflection of cultural conceptions of mental illness. Such conceptions form a cultural context that influences the way we behave towards those suffering from mental illness. As Link et al. [28] have pointed out, ‘as a context this cultural conception becomes an external reality, something that individuals must take into account when they make decisions and enact behavior’ (p. 255). Second, due to selection processes typical for on-line surveys [19], the representativeness of the results for the whole of the French population may be considered uncertain. In this context it seems worth noting that the feasibility of this interview mode for exploring beliefs and attitudes about mental disorders has recently been shown in another survey in France [29]. Third, social desirability may have biased our findings in the sense that respondents may have responded to interview questions in a way that makes them appear socially desirable, and may not actually have responded according to their true reactions. However, this seems to be less a problem in online interviews than in face-to-face interviews [30]. Finally, we do not know how the terms used in the interview were understood by the respondents. For instance, what the lay public associates with this term ‘psychotherapy’ does not necessarily reflect how psychotherapy is defined by mental health professionals. A recently published study comes to the conclusion that lay people appear to have ‘a modestly realistic but somewhat naïve view of the process and efficacy of psychotherapy’ [31]. It is not unlikely that many people simply associate with psychotherapy talking with patients [32].