There can be long delays between the onset of psychotic symptoms and the receipt of adequate treatment. It has been argued that this duration of untreated psychosis causes unnecessary suffering of patients and has adverse effects on their social networks and on the achievement of their educational and occupational goals (1). Recent meta-analyses have confirmed an association between duration of untreated psychosis and poorer outcome (2,3). Although this correlational evidence cannot show that duration of untreated psychosis is the cause of poorer outcome, it supports further investigation into the benefits of early intervention and into the barriers to earlier help seeking.

One potential barrier is knowledge and attitudes about sources of professional help and treatments (4). In many countries the first step in seeking professional help for a psychotic disorder is to consult a general practitioner. However, attitudes toward general practitioners as a source of help for psychosis or schizophrenia vary from country to country, with generally favorable attitudes in Australia (5,6), Switzerland (7), and Germany (8) but less favorable attitudes in Japan (9) and Austria (10). When members of the public are asked about health professionals, psychiatrists are seen fairly favorably in many countries as a potential source of help (6,7,8,9). However, counselors are rated more highly than psychiatrists as a potential source of help in Australia and Japan (6,9), and psychotherapists are rated more highly than psychiatrists in Germany (8).

Antipsychotic medication is a cornerstone of treatment of psychosis, but surveys in many countries show that members of the public generally have negative attitudes toward it (6,7,10). These negative attitudes are associated with concerns about side affects and a belief that medication deals with symptoms rather than causes (11,12). In contrast, counseling and psychological treatments tend to be viewed very favorably (6,8,9,10). In-patient admission is often a component of the management of psychotic disorders, but surveys in several countries have found that it is seen as harmful by most members of the public (7,9,13).

These survey data suggest that there may be some major gaps between public and professional beliefs about the appropriate interventions for psychotic disorders. In Australia, there has been a direct comparison between public beliefs and those of general practitioners, psychiatrists, clinical psychologists, and mental health nurses (14,15). Although agreement was found about some interventions, the public respondents were much more negative than the professionals about antipsychotic medication and admission to a psychiatric ward. On the other hand, the public respondents were much more in favor of taking vitamins, following special diets, and reading self-help books. A similar study in Switzerland compared the public with psychiatrists, psychologists, nurses, and other mental health professionals (7). There was agreement about some interventions, but again the public was more negative about medication and psychiatric hospitals. On the other hand, public respondents were more positive about psychotherapy and the benefits of "fresh air."

Most of the previous research on public beliefs about interventions for psychosis has involved adults. However, psychotic disorders often have first onset during adolescence and early adulthood (16). This is an age where knowledge of appropriate sources of help and interventions is likely to be the poorest. Parents will be important mediators of help seeking, particularly for adolescents, so their knowledge and attitudes are important as well (17). Reported herein are data from national surveys of the beliefs of Australian health professionals. The surveys covered beliefs about a wide range of potential interventions for mental disorders affecting young people. The beliefs of these health professionals are compared with those of Australian young people and their parents as elicited in a recent national survey (18). We hypothesized that there would be some major gaps between the beliefs of professionals, youths, and parents and that these gaps may have an impact on help seeking and treatment adherence. On the basis of previous research comparing adult members of the public with clinicians, we expected that young people and parents would be less in favor of medical interventions for psychosis and more in favor of psychological and lifestyle interventions.

Professional samples

Clinicians were sent a questionnaire based on a vignette describing psychosis, depression, depression with alcohol misuse, or social phobia. We report here on the subsample of professionals who received the vignette about psychosis. Ethics approval was given by the University of Melbourne Human Research Ethics Committee. Details of the entire sample have been published elsewhere (19). Surveys were mailed in 2006 to 428 psychiatrists listed in the Medicare Provider File (Medicare is Australia's national government-funded health insurance program), a random sample of 500 general practitioners listed in the file, 407 Australian members of the Australian and New Zealand College of Mental Health Nurses, and a random sample of 500 psychologists listed in the Victorian Psychologists Registration Board's online database of registered psychologists. Surveys were completed anonymously, and separate response cards with identification numbers were used to determine participation or refusal. General practitioners, psychiatrists, and psychologists were sent one reminder letter to encourage participation. Completed surveys were received from 105 general practitioners, 155 psychiatrists, 106 mental health nurses, and 183 psychologists. Response rates were 21% (general practitioners), 37% (psychiatrists), 26% (mental health nurses), and 41% (psychologists). Consent was implied by returning the questionnaire.

Youth and parent samples

As with the professionals, the youths and parents received an interview based on one of four vignettes. Again, we report on the subsample who received psychosis vignettes. Details of the entire sample have been published elsewhere (18,20). In 2006 a telephone survey was carried out with a national sample of young Australians ranging in age from 12 to 25. If the young person lived at home with a parent, then one parent was randomly invited to be interviewed as well. Interviews were completed for 968 young people and 531 coresident parents. The response rate for the entire sample of young people was 62%, and that of the sample of parents was 69%. It was not possible to determine separate response rates for the sample subset reported here. Consent was given orally. Parents also gave consent for youths under 18 years old.

Survey questions

The survey was based on a vignette of a young person with early psychosis (20). The vignette was written to meet the minimum requirements for DSM-IV criteria for schizophrenia. Professionals were randomly given a vignette describing a 15- or 21-year-old. Vignettes were matched to the gender and age group of youths and their parents, whereas professionals received vignettes portraying only males because of the smaller sample. After reading the vignette, professionals were asked a series of questions to assess their recognition of the disorder described in the vignette, beliefs about first aid (that is, actions that family or friends could take to help the person), interventions, and prevention. They also provided information on sociodemographic characteristics.

This study focused on beliefs about the helpfulness of different interventions, so these are described in detail here. We assessed the following interventions: consulting with a general practitioner or family doctor, a teacher or lecturer (the former for 12- to 17-year-old respondents and the latter for 18- to 25-year-old respondents), a counselor, a telephone counseling service such as Kids Helpline or Lifeline (the service varied by age group), a psychologist, a psychiatrist, other mental health professionals (occupational therapist, social worker, or mental health nurse), a close family member, or a close friend; dealing with the problem on his or her own; taking vitamins, St. John's wort, antidepressants, tranquilizers, antipsychotics, or sleeping pills; becoming more physically active; getting relaxation training; practicing meditation; having regular massages; getting acupuncture; getting up early each morning and out in the sunlight; receiving counseling by a health professional, receiving cognitive-behavioral therapy; searching a Web site for information about the problem; reading a self-help book on the problem; joining a support group of people with similar problems; going to a local mental health service; being admitted to a psychiatric ward of a hospital; using alcohol to relax; smoking cigarettes to relax; using marijuana to relax; cutting down on use of alcohol; cutting down on smoking cigarettes; and cutting down on use of marijuana. These interventions were selected on the basis of published surveys (5,6) and included complementary and self-help interventions that have some evidence base for treatment of depression or anxiety (21,22).

Participants assessed interventions as "helpful," "harmful," "neither," "depends," or "don't know."

Statistical analysis

Ratings of each intervention were dichotomized, with "helpful" responses scored 1 and other responses scored 0. For descriptive purposes, the percentage of respondents giving "helpful" ratings was calculated for each type of intervention by age of the subject in the vignette and by professional group. To examine differences in ratings according to clinician characteristics and age in the vignette, binary logistic regressions were carried out. The predictors were profession (with general practitioner used as the reference category because practitioners are often the first point of professional contact), gender of clinician, age of clinician (in decades from 20—29 up to 60 years and over), clinician experience with young people ages 12—25, clinician experience with adults ages 26—64, clinician experience with patients age 65 and older (for each variable, the clinician rated how frequently he or she treats this age group on a 4-point scale: exclusively, frequently, sometimes, and never or rarely) and age in the vignette (with 21 years as the reference category). Because of the number of comparisons examined, only effects significant at the p<.01 level are reported here.

Data on the youth and parent samples have been reported previously (18) and are presented here in a form that allows direct comparison with the clinician data. The major interest is in how youths and parents rate interventions about which there is clinician consensus on helpfulness and, conversely, how clinicians rate interventions that are widely endorsed by youths or parents. Consensus varied in degree, but we selected interventions where the mean helpfulness rating across the four professions was 70% or higher. Because of the large sample of youths and parents, even very small differences in ratings of helpfulness could be statistically significant. Therefore, comparisons of the public with professionals were made in terms of effect sizes, with medium and large effect sizes noted. These were defined according to Cohen's h (23), which represents the difference between two proportions after they have been subjected to an arcsine square-root transformation. A medium effect corresponds to h=.5 and a large effect to h=.8 (23).

Differences between clinical professions

When logistic regression was used to examine predictors of helpfulness ratings, several significant differences were found. Compared with general practitioners, psychiatrists were less likely to recommend a counselor (OR=.31, p<.001), telephone counseling (OR=.44, p=.005), a psychologist (OR=.38, p=.004), meditation (OR=.24, p<.001), massages (OR=.15, p=.002), counseling by a health professional (OR=.41, p=.003), or a support group (OR=.24, p<.001) or to recommend reducing use of alcohol (OR=.12, p<.001) or cigarettes (OR=.41, p=.004). [Tables that present the percentages of each clinical profession endorsing each intervention as likely to be helpful are available as online supplements to this article at ps.psychiatryonline.org.]

Compared with general practitioners, psychologists were less likely to recommend use of a general practitioner (OR=.15, p=.005), teacher or lecturer (OR=.33, p=.006), other mental health professional (OR=.34, p=.002), family (OR=.31, p<.001), friend (OR=.37, p=.001), tranquilizers (OR=. 24, p=.001), antipsychotics (OR=.15, p<.001), or support group (OR=.35, p=.003) or to recommend reducing use of alcohol (OR=.07, p=.001) or cigarettes (OR=.38, p=.002).

Compared with general practitioners, mental health nurses were more likely to recommend vitamins (OR=3.78, p=.006) and less likely to recommend a counselor (OR=.38, p=.002) or support group (OR=.39, p=.010).

Differences by clinician experience, age, and gender

The frequency of contact that clinicians had with young people had no association with youths' intervention beliefs. However, clinicians who had greater contact with adults were more likely to advocate getting up early and out in sunlight (OR=2.44, p=.004).

Clinician age had more influence. Older clinicians were less likely to believe in the helpfulness of psychologists (OR=.68, p=.001), other mental health professionals (OR=.34, p=.002), friends (OR=.77, p=.004), relaxation training (OR=.77, p=.004), and getting up early and out in sunlight (OR=.76, p=.004). Clinician gender had no associations with intervention beliefs.

Differences by age in vignette

For the 21-year-old compared with the 15-year-old portrayed in the vignette, clinicians gave higher ratings of helpfulness to use of antipsychotics (OR=1.88, p=.009) and hospitalization in a psychiatric ward (OR=1.65, p=.005) and lower ratings to use of a teacher or lecturer (OR=.26, p<.001).

Differences between public beliefs and clinicians' consensus

Although there were some statistically significant differences between the clinical professions, most of these were between interventions that had low endorsement. The major interest was in the interventions that were widely accepted as likely to be helpful. Table 1 shows the interventions that were endorsed as helpful by an average of 70% or more of clinicians or by 70% or more of youths or parents. The table also indicates where there were medium or large differences between clinicians and youths or parents.

There were a number of interventions for which there was high professional endorsement but lower endorsement by youths or parents. Both youths and parents gave lower ratings to seeing a psychiatrist, using antipsychotics, and seeking mental health services.

There were many interventions that had high youth or parent endorsement but lower clinician endorsement. Compared with youths, clinicians gave lower endorsement to counseling from a health professional, phone counseling, a counselor, a close friend, close family, a support group, physical activity, relaxation training, or meditation; getting up early and out in sunlight; reading a self-help book; seeking Web site information; and reducing use of cigarettes. Compared with parents, clinicians gave lower endorsement to counseling by a health professional or professional counselor, phone counseling, a close friend, close family, a support group, relaxation training, physical activity, and meditation; getting up early and out in sunlight; and reducing use of cigarettes.

This study shows some major gaps between professional, youth, and parent beliefs about the potential helpfulness of interventions for early psychosis. In general, the professionals favored medical and specialist mental health services. They showed general agreement that a young person with psychosis would benefit from consulting a general practitioner, a mental health specialist, or mental health services. They believed such a person would also benefit from taking antipsychotic medication and reducing use of marijuana and alcohol.

Young people and their parents agreed with some of this professional consensus but were much less likely to believe in the benefits of some specialist mental health interventions (taking antipsychotics, seeing a psychiatrist, and using mental health services). On the other hand, there was a wide range of general and informal sources of support that young people and their parents believed in more than professionals. The public particularly endorsed informal social supports (help from family, friends, and support groups), generic counseling, and general stress reduction methods (relaxation training, meditation, and increased physical activity). Young people also more often endorsed informal sources of information (such as Web sites and self-help books).

The survey presented vignettes portraying either an adolescent or a young adult, and the young people who participated covered a broad age range (12—25 years). However, when the interventions most often rated as helpful were examined, neither the clinicians nor the young people and their parents differentiated their responses by the vignette subject's age. These findings are relevant to the controversy about whether adolescents should be treated separately from young adults (the child and adolescent service model) or whether both age groups should be offered common services (the youth model) (24). These findings show little basis for splitting the two age groups for early psychosis intervention.

The beliefs of professionals can be compared with the recommendations in clinical practice guidelines (25,26). The interventions most endorsed are all consistent with the guidelines. However, psychologists were less likely to endorse antipsychotic medication. Although some of these psychologists may not have been working in the mental health field, all were in a position to influence help seeking and should have had a basic familiarity with recommended practice. Recommending antipsychotic medication needs to be a target for continuing professional education. A more widespread problem was the low level of endorsement of cognitive-behavioral therapy, which is recommended by practice guidelines (25,26) but was endorsed as likely to be helpful by only about half the professionals. Indeed, the professionals showed slightly higher support of generic counseling, which is explicitly not recommended by one set of guidelines on schizophrenia (26). Although cognitive-behavioral therapy had a similar low level of endorsement by young people and their parents, this finding is likely to be due to lack of knowledge about the treatment label. If this type of intervention were more widely available, it would not only increase the use of evidence-based treatment but also fulfill the desire of young people and their parents for psychological interventions.

The youths and parents endorsed the likely helpfulness of stress reduction methods. These are not recommended in clinical practice guidelines, but there is some limited evidence to support their usefulness (27,28). Given the high acceptability of these interventions, they merit greater investigation by researchers.

What can be done to reduce the gap between professional and public beliefs about interventions for psychosis? Public attitudes toward psychosis treatment can be changed, as indicated by substantial historical changes found in both Australia and Germany in recent years (29,30, 31). The reasons that these changes have occurred is unclear. Although public information campaigns to reduce the duration of untreated psychosis have focused on things other than intervention beliefs (4), their success shows that mental health information campaigns can be effective. There also is evidence that individual training programs to increase the mental health literacy of the public can change beliefs about interventions for psychosis (32).

Given that informal help from friends and family is seen as so important by young people and their parents, these people can play an important role in facilitating earlier help seeking. However, many members of the public lack appropriate first-aid skills (20,33). There are several helpful actions that a member of the public can take to assist someone who is becoming psychotic (34), and training in first-aid skills is known to improve helping behavior (32). Such training needs to be more widely available.

Given that the Internet and self-help books were rated highly by young people, these media could be further used to promote earlier help seeking. There has already been work in this area with Web sites (35).

This study has a number of limitations that must be acknowledged. The survey content was designed for the public and was not optimal for professionals. For example, it did not tap into the staging of interventions, gave no detail on the content or quality of interventions, did not allow the respondent to record the intensity of his or her agreement, and omitted describing more specialized interventions that were difficult to convey to the public with simple questions (such as cognitive remediation and social skills training).

The study showed that there are major gaps between professional and public beliefs, with young people and their parents less in favor of some specialist mental health interventions and more in favor of general and informal sources of support. These gaps could be barriers to early and appropriate treatment of psychosis. Closing these gaps is a two-way process, requiring community education about psychosis treatment and a broadening of services offered so as to better meet consumer expectations.

Funding was provided by grant 179805 from the National Health and Medical Research Council Program, the Colonial Foundation, and Beyondblue: The National Depression Initiative. Claire Kelly, Ph.D., Robyn Langlands, M.A., Betty Kitchener, M.Nurs., and Len Kanowski, M.Sc., had input into the survey content. Staff of the Social Research Centre provided advice on the methodology of the youth and parent survey. Eoin Killackey, D.Psych., provided advice on the interpretation of clinicians' beliefs. The Australian Government Department of Health and Ageing and the Australian and New Zealand College of Mental Health Nurses assisted with the sampling of professionals.

The authors report no competing interests.