Only about a third of those with a personal history of treatment for mental illness recognized depression, a proportion similar to that observed among persons with no such history. However, attitudes concerning psychological and medical interventions for a person with depressive symptoms were more positive among those with a personal history. Due to the cross-sectional design of our study, we cannot assume that these more positive attitudes are a direct result of the treatment experience. It is probable that persons with positive attitudes towards mental health care would be more likely to seek help in the first place. Before results are discussed further, some comments concerning methodology are warranted.
Strengths and limitations
Strengths of the study include a study design with a well-defined population. It is advantageous to carry out such a study in a country like Sweden where every resident has a unique personal identification number. This facilitates sampling and respondents may be more representative of the underlying population than those who are selected by random telephoning and household sampling. While there was considerable non-participation at the questionnaire stage, our response rate is higher than in some other studies in the field [22, 24]. Still, it is probable that persons with psychiatric symptoms and negative attitudes to help-seeking are overrepresented among those who did not respond to the postal questionnaire.
The method of postal questionnaire was chosen with the specific purpose of reaching also a non-clinical group, namely persons with symptoms of mental disorder but no treatment contact. Most of the subjects classified as cases had an active clinical picture. As expected there were somewhat fewer subjects with active symptoms among the cases with contact, probably due to successful treatment. It is notable that even though cases without contact had not sought help for their symptoms, as many as two out of three of these opted to participate in an interview about their mental health and personal problems, when an opportunity was offered.
Due to the skewness in participation rate at the second stage, significant differences among the groups may have been missed. As expected, cases with mental health care contact had the highest participation rate. It was also fairly easy to recruit cases without mental health care contact. However, it was more difficult to motivate mentally healthy to participate in the interview. One might assume that those who are sceptical to mental health care and what it offers are less inclined to participate in an investigation like this. Another limitation is that the study did not have the power to detect group differences in attitudes to interventions that were chosen by a small number of persons. For example, while the proportion of cases with contact who were positive about ECT was thrice that of the group without contact, the difference in proportions was not significant.
The study did not focus on particular diagnostic entity, but rather on mental illness within a much broader context. This is a weakness, when it comes to comparability with other research, but could also be seen as strength as those sampled with this approach mirror the mental health status in the community.
Less than one third correctly recognized that the vignette depicted a depression. This study was set in a country without a national depression awareness campaign, and it is thus not surprising that many respondents (60%) described the problem within the broader context of mental illness (including depression, mental illness, stress and emotional problems) rather than specifically identifying depression. Our findings can be compared to results from a country such as Australia, where public recognition of depression increased from 39% in 1997 to 67% in 2003–2004 . Efforts have been made on many levels in the Australian society to enhance public knowledge about the ubiquity of mental disorders, particularly depression, but it is still difficult to attribute the improvement of mental health literacy about depression to any one factor .
Recognition of a depression from a vignette did not differ among our three interview groups. Neither was recognition influenced by the fact that the individual actually had a current episode of major depression. It has previously been shown by Goldney and co-workers , that those with a major depression were no more likely than others to recognize a depression. One explanation for this might be cognitive impairment due to depression.
Female sex, young age and a higher degree of education were associated with recognition. Greater mental health literacy has been shown in younger persons than in older persons in Australia . In our study, young women were the group with the best mental health literacy; more than fifty per cent recognized depression. Consequently, males and less educated groups have difficulties in recognizing depression which may prevent them from seeking help from the mental health care system. This may result in unmet needs, which has recently been pointed out as a motive for strategies that target these groups .
In response to the open-ended question regarding the best form of treatment, one third of the participants suggested counselling and there was no difference among groups. A large majority of the respondents favoured psychotherapy. This is in line with other reports based on case vignettes . A review of preferences among depressed patients in primary care reveals that a majority prefer counselling or psychotherapy .
Significantly more respondents among cases with contact were in favour of consulting a GP, compared to both cases without contact and mentally healthy. Only 13% of the total group recommended contact with a GP, a proportion considerably smaller than reported from Australia, where half of the respondents indicated that the person in the vignette should contact their GP . While this may be due to different attitudes towards GPs, it may also reflect differences in how primary care is structured in Australia and Sweden. It is not always easy to get an appointment with a GP in Sweden, and there is often a lack of physician continuity which might make persons less willing to seek help for mental health problems. The proportion of respondents in the current study who indicated that work-related interventions would provide the best help was similar to the proportion that suggested contact with a GP. During the interview many respondents made note of the boss' concern regarding lower productivity. Stressful work environments resulting in sick leave due to "burn-out" have been a recent focus in the Swedish mass media, and this might in part explain the finding that as many as 15 percent considered that the best form of help would be a work related intervention.
Cases with contact were less positive towards the lay support system; this was the case both in the open-ended and in the forced responses. It is possible that cases with history of mental health contact had previously elicited help from family and friends and had found that insufficient. This could have played a role in their decision to seek treatment. Cases without contact preferred the lay support system. Previous research from Germany shows that the public opinion favours the lay support system for depression . In Australia, however, only one fifth suggested lay support [43, 45]. As mentioned above, the most common suggestion was the GP, which might reflect the strong role of the family doctor.
Concerning pharmacological interventions, such as antidepressants, cases with contact were more positive than both cases without contact and mentally healthy. This was even more pronounced in the subgroup that was presently using antidepressants; nine out of ten rated this medication as helpful. In light of previous research, it was somewhat unexpected that half of those without contact rated antidepressants as helpful. Other research shows that most lay people have negative views of antidepressants [16, 50] and less than one third of general population or depressed primary care samples are positive about treatment with antidepressants . In the present study 20% rated antidepressants as harmful, a figure similar to that reported from Australia . Reasons for public scepticism include worry about side effects and the belief that antidepressants may cause dependency . Also, it has been shown that psychotropic drugs provoke fear of losing control to a larger extent than drugs indicated for physical illness . A question may be to what extent the individual's experience may affect treatment preferences? Jorm and co-workers showed that people who had sought help for depression were more likely to believe in medical interventions .
Outcome with and without professional help
As in other research  our participants perceive the course of depression more optimistically with appropriate treatment than in the absence of treatment. Studies from Europe, Asia and Australia report that only about 5% of the population believe in full recovery for depression without help [53, 54]. In our study one fourth of the cases without contact believed in full recovery without intervention. Is this optimism based on own experience of recovery without treatment or is it due to an underestimation of the problem?