Students 18 years or older, who were enrolled in a university, were included in the study. The second and fourth year medical students were recruited during seminar series devoted to the teaching of mental health topics. The fourth year students were engaged in the clinical aspects of psychiatry and addiction medicine. Ethnic Chinese students were contacted and invited to the study through Chinese student organisations. The survey was also distributed to students attending a college of The Australian National University (ANU, Canberra) using a convenience sampling approach. The study was approved by The University of Sydney Human Research Ethics Committee.

Version 1.1 of the survey was divided into seven sections [see Additional file 1]. At the end of each section, subjects are given the option of continuing the survey or terminating. The survey covers a range of topics, including:

The survey utilised the concepts of knowledge, awareness and attitudes that have been more generally used in mental health literacy research and the evaluation of other mental health promotion campaigns [⁶]. However, it made additional use of those items that were developed to assist with the assessment of the impact of beyondblue in Australia [¹²]. Additional elements drew on previous research assessing the impacts of stigma in relation to persons with mental disorders [¹⁴] as well as the specific development of a brief questionnaire to assess experiences of mental health service use [¹³].

As the aim was to develop a basic self-report format for use in the countries and cultures of the Asia Pacific Region, a draft survey was sent to SEBoD committee members and country chapter heads. They were asked to comment particularly on the suitability and comprehensibility of the survey in their local circumstances. The draft was modified according to their comments and suggestions. The items were simplified and pre-tested among a group of native English speakers to ensure the survey was understandable for people with various levels of English literacy. It was also tested and commented upon by culturally diverse community members, including Chinese, Japanese and Korean, for cultural appropriateness.

For the items regarding major health problems, specific illness or injuries and mental health problems, only those categories nominated by over 10% of students are reported. Additionally, only those categories nominated by over 20% of students within the items concerning typical signs, behaviours and experiences of persons with depression are reported in this article. For the questions on help seeking, only the combined percentages of the answers "probably likely" and "definitely likely" are presented. Likewise, only the percentages for combined positive answers are described, such as "probably likely" and "definitely likely" for the discrimination item, and "agree" and "strongly agree" for the stigma item. These combined categories were also used for the purposes of statistical analysis.

For the K10, the response options "none of the time", "a little of the time", "some of the time", "most of the time" and "all of the time" were coded as 1, 2, 3, 4 and 5 respectively. The sum of the scores for each item were calculated and recoded into three levels of psychological distress: 10?15 representing low or no, 16?29 representing medium, and 30?50 representing high distress. The last two categories were combined both for the purposes of presentation and for statistical analysis.

The response options in the SPHERE questionnaire, "never or some of the time", "a good part of the time" and "most of the time" were coded as 0, 1, and 2 respectively. The total scores of the six psychological questions and the six somatic questions were calculated separately. The students with a total score of two or more for the six psychological items were categorised as having strong psychological symptoms (PSYCH-6); the students with a total score of three or more for the six somatic items were categorised as having strong somatic symptoms [¹⁶]. The percentages of students who showed significant psychological and/or somatic symptoms were reported and compared among the groups.

Two sets of analyses were conducted, the first comparing second year medical students with fourth year medical students (to examine the effect of mental health training), the second comparing Chinese students with ANU students (to examine the effect of culture). Associations between group membership (type of student) and item responses were made using the Chi-square test, the continuity correction value being considered for two by two tables. Age was significantly skewed, and this could not be rectified, so the non-parametric Mann-Whitney U test was used for the two group comparison. All calculations and analyses were carried out using the Statistical Package for the Social Sciences (SPSS) version 12. Where appropriate, Bonferroni corrections were applied to alpha to account for multiple comparisons.

The mean age of students was 24.4 years (SD = 4.0 years; range = 15 ? 43 years), with Chinese students being significantly older than ANU students, and fourth year medical students significantly older than second year students (see Table 1). A small proportion of the ANU students had a medical/health care background (8%), as did the ethnic Chinese students (9%). Conversely, a small proportion of medical students reported Chinese nationality (second year: 4%, fourth year: 1%), as did the ANU students (3%).

There were significant differences between the various student groups with regard to the recognition of major health problems in Australia (see Table 2). For example, approximately half of the Chinese students nominated "heart disease and stroke" as a main cause of death or disability, compared to the majority of ANU students. By contrast for "brain, behavioural and mental health disorders", just over half of the ANU students rated this key area as a major health problem compared with only a quarter of ethnic Chinese students.

There also appeared to be a training effect, with significantly more fourth year medical students nominating "brain, behavioural and mental health disorders" as a major health problem than second year medical students.

When asked about specific illnesses and injuries, rather than broad disease categories, a similar pattern of responses was observed with significant group differences observed in both sets of analyses (see Table 3). For example, just over half of the Chinese students nominated "heart attack or other heart disease" as a main cause of death or disability, compared to more than 90% of ANU students. There was a trend towards a difference between the medical student groups in nominating "depression" as a specific illness causing death or disability, the proportion increasing with mental health training.

There was a substantial degree of variation in recognition of common symptoms of depression (see Table 5). A significantly greater proportion of ANU students than Chinese students recognised "feeling overwhelmed" as a typical sign of depression. It is of interest to note that there were also some shifts in responses between the second and fourth year medical students, for instance a greater recognition following mental health training of fatigue as a typical symptom of depression.

When asked about the typical behaviours and experiences of people with depression, there were some differences among the student groups (see Table 6). Most notably, withdrawal was recognised as a typical behaviour by over two-thirds of the ANU students, but less than a third of the Chinese students. Otherwise the responses in these two groups were largely similar. In terms of possible training effects, the only significant difference between the medical student groups was for "lack of self care", fewer fourth year students nominating this behaviour. Suicidal ideation or behaviour was also less frequently nominated as typical of people with depression following training.

When asked to estimate the proportion of people who will experience depression in their lives, there was a significant difference between the non-medical student groups in the proportion correctly nominating "one in five" (51% of the ANU students and 23% of the Chinese students), but not between the medical students (49% second year and 60% of fourth year students; see Table 7).

Students were asked to rate how likely they would be to seek help from a list of health professionals (see Table 8). There were no significant differences between the medical student groups. The most notable difference between the ANU and ethnic Chinese students was the proportion reporting they would be probably or definitely likely to seek help from a pharmacist, the proportion being seven times higher in the Chinese group.

Students were then asked about seeking help from non-professional sources. Again, there were no significant differences between second and fourth year medical students. A greater proportion of Chinese students than ANU students indicated they were likely to seek help from non-professionals (such as acupuncturists, religious persons, naturopaths or herbalists, personal trainers and traditional healers; see Table 9).

Students were asked to rate the perceived helpfulness or harmfulness of various common treatments for depression (see Table 10). There were no significant differences between the medical students in terms of the proportion nominating treatments as helpful. However, differences emerged between the non-medical students (ANU and ethnic Chinese). Over two-thirds of the Chinese students (72%) rated "reading self-help books" as helpful, compared with only 39% of ANU students. More ANU students (70%) thought antidepressant medications would be helpful than did Chinese students (46%). Similarly, there was trend towards a difference regarding the helpfulness of "having an occasional alcoholic drink", with 33% of Chinese students considering this helpful, compared with only 13% of ANU students.

When asked about the perceived likelihood of discrimination if they, or someone close to them, experienced depression, the only difference in response amongst the medical students was regarding "a government or other public welfare agency", with a quarter of second year medical students believing this was definitely or probably likely, compared with 42% of fourth year medical students (see Table 11). There was greater disparity in the perceived likelihood of discrimination between the non-medical groups. Only 7% of ANU students felt that discrimination was definitely or probably likely within the context of a public or private hospital, compared with 44% of Chinese students. Similarly, only 4% of ANU students felt that discrimination was likely from a "doctor or other health professional" compared with 37% of the Chinese students.

Students were asked whether they agreed or disagreed with statements concerning people with severe depression (see Table 12). There were no significant differences between attitudes towards people with severe depression between the medical student groups. In contrast, 48% of Chinese students agreed with the statement "have themselves to blame", compared with 7% of ANU students; and 39% of Chinese students agreed with the statement that people with severe depression "shouldn't have children in case they pass on the illness" compared with none of the ANU students.

As shown in Table 13, more than half of the students were experiencing medium or high levels of psychological distress (as measured by the K10) during the 30 days prior to participation in the study.

A significantly greater proportion of the Chinese students showed medium or high levels of psychological distress on the K10 than the ANU students, whereas there was no difference between the medical student groups. Consistent with this, the Chinese students indicated experiencing substantial psychological and somatic symptoms as measured by the SPHERE, with substantial psychological symptoms being significantly more prevalent than in the ANU students.

We would like to thank all the students who participated in the survey.

¹ Mann-Whitney U test.

*There are missing data for different items (age, gender: 2^nd year medical n = 1, 4^th year medical n = 4, ANU none missing, Chinese n = 8 for age and n = 1 for gender; area inhabited: 2^nd year medical none missing, 4^th year medical n = 4, ANU none missing, Chinese n = 5).

^aAdjusted alpha = 0.05/9 = 0.006.

* Data were missing for 6% of the valid sample (n = 25/407).

^a Adjusted alpha = 0.05/15 = 0.003.

* Data were missing for 7% of the valid sample (n = 27/407).

^a Adjusted alpha = 0.05/8 = 0.006.

* Data were missing for 6% of the valid sample (n = 23/407).

^a Adjusted alpha = 0.05/10 = 0.005.

* Data were missing for 3% of the valid sample (n = 13/407).

^aAdjusted alpha = 0.05/9 = 0.006.

* Data were missing for 3% of the valid sample (n = 13/407).

* Data were missing for 4% of the valid sample (n = 16/407).

^a Adjusted alpha = 0.05/8 = 0.006.

* Data were missing for 8% of the valid sample (n = 29/374).

^a Adjusted alpha = 0.05/7 = 0.007.

* Data were missing for 6% of the valid sample (n = 24/374).

^a Adjusted alpha = 0.05/10 = 0.005.

* Data were missing for 9% of the valid sample (n = 32/374).

^a Adjusted alpha = 0.05/8 = 0.006.

* Data were missing for 12% of the valid sample (n = 41/339).

^a Adjusted alpha = 0.05/10 = 0.005.

* Data were missing for 12% of the valid sample (n = 41/339).

K10 = Kessler Psychological Distress scale, SPHERE = Somatic and Psychological HEalth REport, PSYCH-6 disorder = a total score of two or more for the six SPHERE psychological items indicating strong psychological symptoms, SOMA-6 = a total score of three or more for the six SPHERE somatic items indicating strong somatic symptoms.

* Data were missing for 7% of the valid sample (n = 24/333).