Role of Stigma and Attitudes Toward Help-Seeking from a General Practitioner for Mental Health Problems in a Rural Town

Abstract

Objective: To examine the role of perceived stigma and attitudes to seeking care in predicting help-seeking from a general practitioner (GP) for mental health problems.

Method: Across-sectional surveyin 2002 with self-repor t questionnaires assessing current levels of symptomatology, disability, attitudes towards mental illness, knowledge of prevalence and causes of mental illness, contact with mental illness and help-seeking behaviour and preferences and attitudes toward seeking professional psychological help.

Results: No significant relationship was found between symptom measures and measures of disability and help-seeking. Variables positively associated with general attitudes to seeking professional psychological help were: lower perceived stigma, and biological rather than person-based causal attributions for schizophrenia. Willingness to discuss mental health issues with a GP was predicted by the perceived helpfulness of the GP and by no other variable.

Conclusions: Causal attributions and perceivedstigma rather than participants' levels of symptomatology and disability influence attitudes to help-seeking for mental health issues. Efforts to improve attitudes to help-seeking should focus on reducing stigma and improving mental health literacy regarding the causes of disorders.

Keywords

attitudes to help-seeking general practitioner mental health literacy mental illness stigma

The National Survey of Mental Health and Wellbeing (NSMHWB) found that 18% of Australian adults met the criteria for at least one mental disorder and 29% of these individuals were identified as experiencing moder-ate or severe disability [1]. However, only around 40% of people who met the criteria for a mental disorder had sought professional help for a mental health concern in the previous 12 months [2]. Parslow and Jorm [3] conducted a series of analyses on these data in order to identify the sociodemographic and ‘need’ variables associated with the use of various health providers for mental health problems, finding that irrespective of the health provider selected, Composite International Diagnostic Interview (CIDI)-diagnosed depression and anxiety, sex and being separated were generally associated with help-seeking. Other studies have consistently found sex to be one of the strongest predictors of attitudes toward help-seeking for mental health issues [4–6] with women seeking help more frequently than men [7–10].

Other important components which may influence help-seeking include perceived helpfulness of serviceproviders and the treatments offered by them, knowledge of risk factors and causes, attitudes toward seeking help for mental health problems and attitudes toward mental illness [11], [12]. It is found that mental health professionals and the therapies they espouse are often not perceived as being particularly helpful in the treatment of psychological problems [11], [13], [14]. Moreover, individuals who hold negative views about the effectiveness of mental health services are unlikely to express an intention to access such services [13], [14]. However, several studies have shown that people who have sought professional help at some time have more positive attitudes toward helpseeking than those who have not [6],[15–17], although this is not a universal finding [18].

Mental illness continues to incur negative attitudes, often characterized by fear, stigma and rejection [19], [20]. Such attitudes have been identified as contributing to a reluctance to seek help [4], [6], [21]. Attitudes toward helpseeking may be particularly relevant to help-seeking within the context of rural settings with Fuller et al. [22] arguing that the stoic and self-reliant attitudes found in rural Australian residents make it difficult for people to acknowledge that they are experiencing problems and/or distress. US research also indicates that rural residents display less reliance on professional health services and more on family and friends than urban dwellers, and place high emphasis on self-responsibility for health [23].

A recent Australian study sought to understand how mental health is defined in rural and remote communities [22]. Semistructured interviews were undertaken with a range of informants knowledgeable in the area of mental health who provided services in rural and remote areas of South Australia. These informants included mental health nurses, general practitioners (GPs), priests etc. Most informants felt that people living within these areas generally equate mental health issues with severe psychiatric disability or ‘insanity’, which requires hospitalization and is largely irremediable. Such a conceptualization may result in a reluctance to acknowledge mental health problems and/or an inability to conceive their own problems/distress as mental health issues. This is likely to have a direct bearing on help-seeking, as in order for people to seek help, they must first identify themselves as requiring help. This suggests that people living within rural areas of Australia may fail to identify their own symptoms as indicators of mental ill-health, thereby preventing or at least hindering any form of help-seeking behaviour.

In summary, published work supports a range of psychological variables as being important in help-seeking for mental health issues. In Australia, much of the work in this area has focused on mental health literacy [11]. This study aimed to add to the research by examining the role of a number of other psychological variables in people's willingness to seek professional help for mental health issues in a rural setting. It was hypothesized that after controlling for need and demographic variables, psychological variables would significantly predict attitudes toward help-seeking. Specifically, we sought to examine the role of perceived stigma in addition to aspects of mental health literacy, such as contact and experience with mental illness and respondents' causal attributions about mental illness.

Method

The study was conducted in 2002 in Echuca, a rural community in Victoria, Australia. Echucais 220 km north of Melbourne and has a population ofapproximately 10 000.

Sampling

The study used a cluster-sampling approach in which a series of streets within residential areas were randomly selected. The streets selected were each within an approximately 30 km radius of the local Community Mental Health Service. This proximity was chosen in an effort to exclude physical access(i.e. distance) to mental health services as a potential barrier to help-seeking. Occupants of homes with odd street numbers were approached. One occupant per household over the age of 18 years was given a brief verbal summary of the study and requested to participate by completing and returning the questionnaire package. A reply-paid envelope was provided for return of the questionnaires. In the event that the individual declined to participate, no further attempts were made to involve other members of the household. Homes that were unoccupied, inaccessible or deemed as unsafe to access were not included. Questionnaire delivery took place in two phases due to a poor response rate following initial distribution. The first phase, in which 400 questionnaires were delivered, occurred over a 4-day period including both weekend and weekdays, with deliveries occurring between the hours of 11 am and 7 pm. In the second phase, which took place approximately 2 months later, 100 questionnaires were delivered over a 2-day period, both weekdays, between the hours of3 pm and 8 pm. Delivery times were selected in an effort to both maximize the number of residents who would be home and also to sample a cross-section of the community.

Measures

Self-report questionnaires were used to assess current symptomatology, disability, attitudes toward mental illness, knowledge of prevalence and causes of mental illness, contact with mental illness, help-seeking behaviour and preferences and attitudes toward seeking professional psychological help.

Current symptomatology was measured by the Centre for Epidemiological Studies Depression Scale (CES-D: [24]) and the state component of the State–Trait Anxiety Inventory (STAI-Form Y: [25]). Disability was measured using the Medical Outcomes Studies Short Form (SF-12: [26]). The 12-item version was used to assess limitations or disability due to both physical and mental health issues.

A new self-report measure, the Perceived Stigma Scale (PSS), was developed to assess perceived stigma about people with mental illness within a community. The PSS was based on the 12-item Devaluation and Discrimination Scale (DDS: [27], [28]). However, the item wording was changed from American wording or simplified and four items added to total 16 items. The four extra items were intended to gauge the responses of people within the specific local community to someone with a mental illness, for example, ‘people with a mental illness would be treated poorly in this community if people found out about it’. Respondents indicate the extent to which they agree with each of the 16 items with responses assessed on a 1–4 Likert scale ranging from ‘strongly agree’ to ‘strongly disagree’. Higher scores indicate lower perceivedstigma.

Help-seeking attitudes were measured using the Attitudes Towards Seeking Professional Psychological Help (ATSPPH: [4]) Scale.

The Health Beliefs Questionnaire (HBQ) is a self-report measure that was developed for this study and was designed to assess respondents' knowledge of the prevalence and causes of mental illness. Items within this scale are largely based on a series of studies by Jorm et al. on mental health literacy [11]. To assess causal attributions, respondents were asked to indicate how important they believed a range of factors to be in the cause of schizophrenia and depression. A four-point Likert scale was used, ranging from ‘very important’ (1) to ‘not at all important’ (4). In fact, each of the items represents an independent score and does not produce a summary score.

Contact with mental illness, and help-seeking behaviour and preferences, were assessed with a new questionnaire developed for this study, the Contact and Experience Questionnaire (CEQ). The questionnaire asks the following questions: (i) whether the respondent had ever sought help themselves for a mental health issue (yes/no); (ii) how they found the help-seeking experience (if pertinent); (iii) whether they were currently receiving treatment for a mental health issue (yes/no); (iv) whether theyknew ofothers who had experienced a mental health issue (yes/no) and (v) for which condition; (vi) whether they knew of others who had sought help for a mental health issue (yes/no); (vii) whether they had ever wanted to or felt they needed to seek help for a mental health issue but did not do so (yes/no); and (vii) their reasons for not seeking help. Finally, (ix) participants were asked how helpful they believed a GP would be in the treatment of a mental health problem. As with the HBQ, each of the items represents an independent datum and does not produce a summary score.

Statistical analysis

Exploratory and summary statistics were obtained for all variables within the dataset. Data were examined for the presence of missing values, influentialvalues and outliers, and skewness and kurtosiscoefficients were checked. A range of statistical analysis techniques were applied to the data using the SPSS V 9.0 statistical package. Reliability analyses were conducted for all measures. Independent measures t-tests were used to compare mean scores for men and women where appropriate. Pearson's χ² analyses were also used to examine sex differences for the contact with mental illness items and perceived helpfulness of help. A multiple simultaneous logistic regression analysis and hierarchical regression were used to test the role of various psychological variables in help-seeking.

Results

Questionnaires were returned by 142 individuals, a response rate of 28.4%. The sample comprised 50 (35.2%) men and 92 (64.8%) women.

Current symptomatology and disability

The mean score on the CES-D scale was 11.78±7.96 and the Cron-bach's α of 0.85 compares favourably with the 0.90 figure of Radloff [24]. The mean score on the STAI for the sample was 32.65 9.95 and the Cronbach's α of 0.92 compares favourably with the 0.91–0.93 figures reported by Spielberger et al. [25]. The mean for the Mental Component Summary (MCS) scale of the SF-12 was 48.11±9.71 and 50.11±8.29 for the Physical Component Summary (PCS) scale. In the current study, Cronbach's α was 0.71 for the MCS and 0.67 for the PCS.

Attitudes toward help-seeking and perceived stigma

The mean score on the ATSPPH was 49.88±8.99 and Cronbach's α was 0.89. This compares favourably with the reliability of 0.84 reported by Fischer and Turner [4]. The mean score on the PSS was 37.51±4.93. We obtained a Cronbach's α of 0.80 and this compares favourably to the α coefficient for the DDS of 0.82 [28].

Knowledge of mental illness – prevalence and causes

Thirty-five per cent (n =50) of the sample believed the chance of experiencing a mental disorder was approximately one in five, and 35% (n =49) said one in 10.

Respondents were asked to indicate how important they believed a number of factors to be in the cause of schizophrenia and also depression, using a 4-point Likert scale (1 =‘very important’, 2=‘moderately important’,3=‘somewhat important’ and 4=‘not at all important’). For schizophrenia, social/environmental, genetic, personality, illegal drug use, stress and upbringing, were each endorsed as very or moderately important by more than 60% of the sample. Of note, weakness of character was believed to be very or moderately important by just over 40% of respondents and not at all important by just 33%.

For depression, social/environmental factors and stress were rated as very or moderately important by more than 90% of respondents, with genetic factors, personality, illegal drug use and upbringing being endorsed by more than 70% of people. Weaknesses of character was believedto be very or moderately important in the cause of depression by over 60% of those sampled and was considered not at all important by less than 20%.

Contact with mental illness and help-seeking behaviour

Thirty-six per cent (n =52) reported having sought professional help for a mental health issue at some time. Significantly more women than men reported previous help seeking (44.6% vs 20%; p< 0.05). Among those who reported seeking help, 29 responded to the item asking how they had found the help-seeking experience. Over onethird (41%, n =12) described the experience as difficult, 31% (n =9) said it was embarrassing and 17% (n =5) found it to be a frightening experience. Only seven people (5%) were currently receiving treatment for a mental health issue.

Two-thirds (n =94) of the sample reported they knew someone who had experienced a mental health issue. The problem most commonly identified was depression. Fifty-nine per cent (n =84) of the sample reported that they knew someone who had sought help for a mental health issue. Again, more female participants reported knowing someone who had sought help than male participants (70.7% vs 38%). Forty-three per cent of the sample (n =61) indicated that they had at some time wanted to, or felt they needed to, seek help for a mental health issue but did not do so. The most common reasons for not seeking help were embarrassment (43%, n =26) or not knowing whom to see for help (34%, n =21). Sixty-eight per cent (n =97) would feel comfortable discussing a mental health issue with a GP, with a significant difference between men (n =37, 38%) and women (n =60, 62%; p< 0.05).

The role of psychological variables in help-seeking for mental health issues

A simultaneous multiple logistic regression analysis was used to measurethe association between 12 predictor variables and affirmative responses to the question: ‘Would you feel comfortable discussing mental health issues with your GP?’ The predictor variables included sex, measures of symptom and disability (need), four contact-with-mental-illness items, the ATSPPH Scale and the PSS and whether they believed a GP would be helpful in treating a mental health problem (from the HBQ). These variables are displayed in Table 1.

Table 1.

Logit coefficients, standarderrors, Wald statistics and odds ratios (95% CI) for associations between demographic, need and psychological variables and reporting a willingness to discuss mental health (MH) issues with a general practitioner (GP)

Variable	Description	Logit	SE	Wald	OR (95% CI)
Sex (female)	Female/male	0.05	0.59	0.01	1.05 (0.33–3.31)
GP-helpful	‘In treating MH problem?’	−2.46	0.63	15.40	0.09 (0.03–0.29)∗
Contact 1	‘Have you sought help for a MH issue?’	0.15	0.61	0.06	1.16 (0.35–3.84)
Contact 2	‘Are you currently receiving treatment for a MH issue?’	0.09	1.45	0.01	1.10 (0.06–18.69)
Contact 3	‘Know anyone who experienced MH problem?’	0.19	0.86	0.05	1.21 (0.22–6.47)
Contact 4	‘Know anyone who sought help for a MH problem?’	1.14	0.81	1.96	3.12 (0.63–15.35)
Depression	Scores on the CES-D	−0.05	0.05	0.95	0.96 (0.87–1.05)
State anxiety	Scores on the STAI	0.06	0.04	1.97	1.06 (0.98–1.16)
Attitudes towards help seeking	Scores on the ATSPPH	0.08	0.05	2.67	1.08 (0.99–1.18)
Perceived stigma	Scores on the PSS	0.03	0.07	0.21	1.04 (0.90–1.20)
Mental disability	Scores on the MCS – SF-12	−0.01	0.04	0.09	0.99 (0.92–1.07)
Physical disability	Scores on the PCS – SF-12	0.06	0.03	3.45	1.06 (1.00–1.13)

∗ p< 0.05. CES-D, Centre for Epidemiological Studies Depression Scale; STAI, State–Trait Anxiety Inventor y; ATSPPH, Attitudes Towards Seeking Professional Psychological Help; PSS, Perceived Stigma Scale; MCS – SF-12, Medical Outcomes Studies Short Form; PCS –SF-12, Physical Component Summary Short Form.

As shown in Table 1 only one of the predictor variables was found to significantly predict a positive response to the question. Believing that a GP would not be helpful in the treatment of mental health issues decreased the likelihood of a positive response by91%. Surprisingly, sex and contact variables were neither predictive of seeking help from a GP nor were attitudes toward help-seeking, perceived stigma or disability scores.

A hierarchical regression analysis was conducted in order to assess the relationship between demographic need and psychosocial variables, and general attitudes towards help seeking as measured by the ATSPPH Scale. Demographic variables were entered in step 1, followed bysymptom variables in step 2 and the PSS in step 3. In the final step, responses to the seven causal attribution questions for depression, were entered. The final results for depression are shown in Table 2. It should be noted that the table shows the adjusted R² and R² change statistics for each block of variables, but the beta weights displayed in the fourth and final step thereby include all 16 variables. This was also the case for the regression statistics displayed in Table 3.

Table 2.

Hierarchical regression analysis – final results for demographic, symptom, disability, perceived stigma and causal attributions for depression as predictors of attitudes to help seeking

Regression step	Adjusted R2	R2 change	Predictor	Stand. β
1	−0.03	0.00	Sex	−0.01
			Age	−0.09
			Education	−0.00
			Income	0.01
2	−0.04	0.02	CES-D	−0.10
			STAI	−0.13
			MCS	−0.13
			PCS	−0.07
3	0.21	0.24	PSS	0.43∗
4	0.33	0.15	HBQ-Cause – upbringing	0.13
			HBQ-Cause – stress	−0.07
			HBQ-Cause – social/environmental	−0.12
			HBQ-Cause – genetics	−0.06
			HBQ-Cause – drug use	−0.02
			HBQ-Cause – personality	−0.02
			HBQ-Cause – weakness of character	0.35∗

∗p<0.05. ATSPPH, Attitudes Towards Seeking Professional Psychological Help; CES-D, Centre for Epidemiological Studies Depression Scale; HBQ, Health Beliefs Questionnaire; MCS, Mental Component Summary of the SF-12; PCS, Physical Component Summary of the SF-12; PSS, Perceived Stigma Scale; STAI, State–Trait Anxiety Inventory – Form Y.

Table 3.

Hierarchical regression analysis – final results for demographic, symptom, disability, perceived stigma, and causal attributions for schizophrenia as predictors of attitudes to help seeking

Regression step	Adjusted R2	R2 change	Predictor	Stand. β
1	−0.03	0.00	Sex	0.05
			Age	−0.02
			Education	0.04
			Income	−0.03
2	−0.03	0.03	CES-D	−0.17
			STAI	−0.03
			MCS	−0.11
			PCS	−0.02
3	0.27	0.29	PSS	0.47∗
4	0.46	0.20	HBQ-Cause – upbringing	0.27∗
			HBQ-Cause – stress	−0.10
			HBQ-Cause – social/environmental HBQ-Cause – genetics	−0.10 −0.21∗
			HBQ-Cause – drug use	−0.07
			HBQ-Cause – personality	−0.16
			HBQ-Cause – weakness of character	0.43∗

Table 2 shows that scores on the PSS were predictive of scores on the ATSPPH Scale, with less perceived stigma associated with more positive attitudes toward seeking help for a mental health issue. The causal attribution items for depression accounted for 15% of the observed variance, with weakness of character being the only individually significant item in predicting ATSPPH scores; greater perceived importance of weakness of character as a cause of depression was associated with less positive help-seeking attitudes.

For schizophrenia, the first three steps of the regression shown in Table 3 are identical to Table 2. The PSS was significant. The casual attribution questions for schizophrenia comprised the fourth step and accounted for 20% of the variance, with three of the items being individually significant: genetics, weakness of character and upbringing. More positive attitudes toward help-seeking was associated with higher perceived importance of genetics and lower perceived importance of weakness of character and upbringing.

It might be argued that the PSS is also highly associated with causes for both depression and schizophrenia. In fact, the correlations for the PSS and the seven causal attributions for depression were 0.04, 0.01, 0.19, 0.26 (personality), 0.03, −0.03 and 0.02. With the exception of the personality variable, we believe that the stigma scores can be considered to be independent. In fact, for weakness of character the beta weight was 0.35, whereas the correlation was 0.19. However, the beta weight for personality was −0.02, whereas the correlation was 0.26.

For schizophrenia, the correlations between the PSS and the seven causal attributions were −0.02, 0.04, 0.21, 0.29, −0.10, 0.03 and 0.15. The correlation for upbringing was 0.15 and the beta weight was 0.27; the correlation for genetics was 0.04 but the beta weight was-0.21; and for weakness of character the correlation was 0.21 and the beta weight was 0.43.

Discussion

The current findings support the mental health literacy work of Jorm et al. [11]. A significant proportion of the sample indicated that they had experience with mental health problems. More than one-third of respondents indicated that they themselves had previously sought professional help for a mental health problem, with more than half the sample reporting that they knew someone who had. Around two-thirds of the respondents said that they knew someone who had experienced a mental health issue. In each case, more women than men reported such contact. Almost half the respondents indicated that they had at some time wanted to, or felt that they needed to, seek help for a mental health issue but had not done so. The most frequently cited reason for not seeking professional help was embarrassment, followed bynot knowing who to seek help from.

Around two-thirds of the respondents indicated that theywould feel comfortable discussing mental health issues with their GP. Of those who indicated that theywould not feel comfortable discussing mental health issues in this context, almost half indicated that this was because they did not believe that GPs dealt with such issues. More women than men reported being willing to discuss mental health issues with a GP.

Ratings for two causal factors were found to be different for depression and schizophrenia. Both social/environ-mental factors and stress were considered to be more important in the cause of depression than schizophrenia. The observed difference between causal attributions for depression and schizophrenia may reflect differences in perceptions regarding the seriousness of the two conditions [13] and is likely to influence help-seeking.

Nevertheless, the major focus of the paper was on two key questions: the first was whether participants would feel comfortable discussing mental health issues with their GP; the second was assessing general attitudes toward seeking help for psychological problems. No significant relationship was observed between sex, symptom and disability measures and attitudes to help-seeking. This indicates, perhaps surprisingly, that sex, increases in symptomatology and/or disability do not correspond with increased willingness to discuss such problems with a GP nor lead to more positive attitudes toward helpseeking in general. The lack of sex differences in the three regressions could be either due to the low response rate which reflected selection bias or true differences in rural settings. It could also be because of the greater contributions of other variables such as perceived stigma or perceptions of causation of mental illness.

As regards the first key question, one variable predicted willingness to discuss mental health issues with a GP; those respondents who believed that a GP would be unhelpful in the treatment of a mental health problem were 91% less likely to be willing to discuss mental health concerns with their GP than those who felt a GP would be helpful. This finding is very consistent with previous research [13], [14].

As regards the second key question, there were two major findings. First, it was found that higher perceived stigma was associated with more negative attitudes toward seeking help, although it should be emphasized that it was not found to be associated with less willingness to discuss mental health issues with a GP. The finding for perceived stigma in predicting attitudes to help-seeking is consistent with previous studies [21], [29]. The observed association between perceived stigma and attitudes toward help-seeking in this study may, in part, reflect the similarities between the two scales. The PSS was designed to assess the level of stigma respondents perceived to exist in their own community. However, the ATSPPH Scale also contains a number of items designed to measure perceived community stigma, particularly in relation to help-seeking. As such, the relationship between these two scales could be because the constructs that they are measuring overlap to a degree. Despite this measurement issue, it seems reasonable to conclude that people who hold stigmatized attitudes about mental illness or who perceive their community to do so, are more likely to display a reluctance to seek professional help for such issues.

Second, the most important finding was that causal attributions were predictive of scores on the ATSPPH. In particular, there was increased perception that weakness of character was associated with more negative attitudes toward help-seeking for both schizophrenia and depression, whereas upbringing was associated with more negative attitudes toward help-seeking for schizophrenia but not depression. However, the greater perceived importance of genetics for schizophrenia was associated with more positive attitudes. This is consistent with the findings of Angermeyer et al. [13]. Biological factors such as genetics were more likely to lead to formal help-seeking preferences. These findings indicate that person-based causal attributions, that is, those which implicate the ill person and/or their family, are more likely to be associated with negative attitudes toward professional helpseeking.

The question could be posed as to why there is a difference between general attitudes toward help-seeking and seeking help from a GP. The most obvious reason is that the question about seeking help from a GP is directing respondents toward making a specific decision about help-seeking. In this case, perceived stigma, which is about stigma in general, was trumped by respondents' perceptions of the helpfulness of a GP. However, attitudes toward help-seeking in general, may be affected by a range of decisions about acceptability, availability and helpfulness of a range of providers. In this case, perceived stigma was a significant predictor irrespective of the nature of the psychiatric condition.

A major problem with this study is the poor response rate to the survey (28%). We did not keep records about the relative percentages of dwellings that were unoccupied, inaccessible, or deemed unsafe and thus not approached. Anecdotally, the vast majority of people who were approached to participate took the questionnaires but failed to return them. A total of 500 questionnaires were given out and the response rate uses that as the denominator figure. Although not an epidemiological survey, the result is nevertheless disappointing. One can speculate on the reasons for the poor response. It could be that people with more negative attitudes toward mental illness may not have participated, or those who had accessed treatment for mental health problems and were dissatisfied wished to put the experience behind them.

The response rate may have also been affected by the lack of anytangible rewards for participation. In addition, the study was a ‘stand-alone’ initiative, with no link to service providers, state or national initiatives. Without advertising, it may have been unattractive to prospective participants. Finally, the length of the questionnaire may have acted as a disincentive. Thus, the study sample must be regarded as a convenience sample and may not be representative of the population at large.

In summary, the key findings of this study are that factors other than need and demographic variables are influential in both general attitudes toward seeking help for mental health issues, as well as willingness to discuss such issues with a GP. Specifically, the variables observed to be associated with general help-seeking were causal attributions and perceived stigma. Willingness to discuss mental health issues with a GP was predicted by the perceived helpfulness of the GP and by no other variable.

The findings of this study support the current emphasis on improving mental health literacy of the general community in efforts to improve help-seeking. In regard to mental health service provision to rural areas of Australia, the results emphasize the need to educatethe public on the role of GPs. It was observed that a considerable number of respondents indicated that they would not feel comfortable discussing mental health issues with a GP and many did not believe that GPs were an appropriate help-source for such issues. This obviously has important ramifications for rural communities as access to specialist services is limited, and GPs are often the only help-source available.

References

Australian Bureau of Statistics . Mental Health and Wellbeing: profile of adults, Australia, 1997. Australian Government Publishing Service, Canberra 1998.

Andrews

Henderson

Hall

. Prevalence, comorbidity, disability and service utilisation. Overview of the Australian National Mental Health Survey. British Journal of Psychiatry 2001; 178: 145–153, [erratum appears in British Journal of Psychiatry 2001 Dec; 179:561–562.].

Parslow

R A

Jorm

A F

. Who uses mental health services in Australia? An analysis of data from the National Survey of Mental Health and Wellbeing. Australian and New Zealand Journal of Psychiatry 2000; 34: 997–1008.

Fischer

E H

Turner

J L

. Orientations to seeking professional help: development and research utility on an attitude scale. Journal of Consulting and Clinical Psychology 1970; 35: 79–90.

Leong

F TL

Zachar

. Gender and opinions about mental illness as predictors of attitudes toward seeking professional psychological help. British Journal of Guidance and Counselling 1999; 27: 123–132.

Surgenor

L J

. Attitudes towards seeking professional psychological help. New Zealand Journal of Psychology 1985; 14: 27–33.

Good

G E

Wood

P K

. Male gender role conflict, depression and help seeking: do men face double jeopardy?. Journal of Counseling and Development 1995; 74: 70–75.

Gove

. Gender differences in mental and psychiatric illness. Social Science and Medicine 1984; 19: 77–91.

Greenley

J R

Mechanic

. Social selection in help seeking for psychological problems. Journal of Health and Social Behavior 1976; 17: 249–262.

10.

Kessler

R C

Brown

Broman

. Sex difference in psychiatric help seeking. Journal of Health and Social Behavior 1981; 22: 49–64.

11.

Jorm

A F

Korten

A E

Jacomb

P A

Christensen

Rodgers

Pollitt

. “Mental health literacy”: a survey of the public's ability to recognise mental disorders and their beliefs about the effectiveness of treatment. Medical Journal of Australia 1997; 166: 182–186.

12.

Jorm

A F

Korten

A E

Jacomb

P A

Christensen

Rodgers

Pollitt

. Public beliefs about causes and risk factors for depression and schizophrenia. Social Psychiatry and Psychiatric Epidemiology 1997; 32: 143–148.

13.

Angermeyer

M C

Matschinger

Riedel-Heller

S G

. Whom to ask for help in case of mental disorder? Preferences of the lay public. Social Psychiatry and Psychiatric Epidemiology 1999; 34: 202–210.

14.

Bayer

J K

Peay

M Y

. Predicting intentions to seek help from professional mental health services. Australian and New Zealand Journal of Psychiatry 1997; 31: 504–513.

15.

Cash

T F

Kehr

Salzbach

R F

. Help-seeking attitudes and perceptions of counselor behaviour. Journal of Counseling Psychology 1978; 25: 264–269.

16.

Halgin

R P

Weaver

D D

Edell

W S

Spencer

P G

. Relation of depression and help-seeking to attitudes toward seeking professional psychological help. Journal of Counseling Psychology 1987; 34: 177–185.

17.

Lin

Parikh

S V

. Sociodemographic, clinical and attitudinal characteristics of the untreated depressed in Ontario. Journal of Affective Disorders 1999; 53: 153–162.

18.

Bosmajian

C P

Mattson

R E

. A controlled study of variables related to counseling center use. Journal of Counseling Psychology 1980; 27: 510–519.

19.

Link

B G

Struening

E L

Rahav

Phelan

J C

Nuttbrock

. On stigma and its consequences: evidence from a longitudinal study of men with dual diagnosis of mental illness and substance abuse. Journal of Health and Social Behavior 1997; 38: 177–190.

20.

Nunnally

J C

. Popular conceptions of mental health. Holt, Rhinehardt, Winston, New York 1961.

21.

Kushner

A G

Sher

K J

. The relation of treatment fearfulness and psychological service utilization: an overview. Professional Psychology: Research and Practice 1991; 22: 196–203.

22.

Fuller

Edwards

Procter

Moss

. How definition of mental health problems can influence help seeking in rural and remote communities. Australian Journal of Rural Health 2000; 8: 148–153.

23.

Weinert

Long

. Understanding the health care needs of rural families. Family Relations 1987; 36: 450–455.

24.

Radloff

L S

. The CES-D scale: a self-report depression scale for research in the general population. Applied Psychological Measurement 1977; 1: 385–401.

25.

Spielberger

C D

Gorsuch

R L

Lushene

R E

. Manual for the State–Trait Anxiety Inventory. Consulting Psychologists, Palo Alto, CA 1983.

26.

Ware

J J

Kosinski

Keller

S D

. SF-12: how to score the SF-12 Physical and Mental Health Summary Scales. Quality Metric Limited, Boston 1998.

27.

Link

B G

. Understanding labeling effects in the area of mental disorders: an assessment of the effects of expectations of rejection. American Sociological Review 1987; 52: 96–112.

28.

Link

B G

Mirotznik

Cullen

F T

. The effectiveness of stigma coping orientations: can negative consequences of mental illness labeling be avoided. Journal of Health and Social Behavior 1991; 32: 302–320.

29.

Amato

P R

Bradshaw

. An exploratory study of people's reasons for delaying or avoiding helpseeking. Australian Psychologist 1985; 20: 21–31.