• Open Access

Falling through the cracks: the gap between evidence and policy in responding to depression in gay, lesbian and other homosexually active people in Australia

Authors


Abstract

Objective: To examine the evidence for a national policy response to depression among gay, lesbian and other homosexually active people in Australia.

Methods: A literature review using database searches on depression among non-heterosexual people then a web-based search of national policy investigating how mental health needs in this population are addressed in Australia.

Results: The literature review found that non-heterosexual people experience depression at higher rates, but the literature on interventions was sparse.

The policy analysis found no mention of depression or the broader mental health needs of non-heterosexual people in key national mental health policy documents. These documents outline a policy approach for population groups with a higher prevalence of mental health problems, and stigma and discrimination are relevant associated factors, but only the National Suicide Strategy considers non-heterosexual people an ‘at-risk group’.

Conclusions: The results suggest that the evidence on higher rates of depression in non-heterosexual people is strong, but that this is not recognised in current national policy.

Implications: Defining non-heterosexual people as an ‘at-risk’ group is appropriate, as is prioritising access to mental health services that are socially and culturally appropriate. Addressing homophobia as an associated factor would require a strategic policy approach across a range of sectors.

Mental health has become increasingly prominent in national policy discussion in Australia since the adoption of the first National Mental Health Policy in 1992. With the election of the federal Labor government in 2007, mental health was identified as a key issue and re-confirmed as a second-term priority after the 2010 election. A new Ministry for Mental Health was established and substantial funding initiatives followed.

This study examines the evidence for a national policy response to depression among gay, lesbian and other homosexually active people in Australia. The study includes a review of the growing body of evidence, both international and Australian, showing greater risk of depression among non-heterosexual people. A systematic review and meta-analysis by King et al.1 investigating epidemiological evidence around prevalence of mental health issues in this population found significantly higher rates of suicide attempts, depression and anxiety disorders and alcohol and other substance dependence.

The review presented here focuses specifically on depression as a way to shed light on mental health in non-heterosexual people more broadly. A detailed investigation of the literature for one particular condition has allowed a focus on methodological and theoretical issues, differences within non-heterosexual populations, risk and protective factors, the influence of stigma, discrimination and homophobia, and on interventions to address these issues. The study investigates the national policy response to depression among non-heterosexual people in Australia through an analysis of national mental health policy, as this sets the framework for coordinating and funding mental health programs and initiatives.

The standard term for the populations reported on here will be ‘non-heterosexual’. However, phrases such as ‘lesbian and other homosexually active women’ and ‘gay and other homosexually active men’ and other terms (‘gay men’, ‘lesbians’, ‘bisexuals’ or the acronym GLB to refer to gay, lesbian and bisexual people as a group) will be used when discussing studies that used those specific terms.

Methods

Literature review

A systematic search was conducted of five databases: Medline, Current Contents, PsycINFO, CINAHL, and Sociological Abstracts. Figure 1 presents a flow diagram of the methodology for the search. Search terms were selected according to two primary domains: ‘depression’ and ‘sexual orientation’. The search terms were crossed in both domains [(depress* OR mental) AND (gay OR lesbian* OR bisexual* OR queer OR transgender* OR transsex* OR homosexual* OR same-sex attract* OR GLB OR MSM)]. Searches were limited to titles, abstracts and keywords, to published works with stated authorship, to studies in English, to human beings, and to studies published between 2000 and 2008 (July). Of the 2,917 references identified, 1,121 were removed as duplicates. The abstracts of the remaining 1,806 references were assessed according to whether the consideration of depression in non-heterosexual populations was ‘core’, ‘peripheral’ or ‘irrelevant’.

Figure 1.

Flow diagram for literature review.

Of the 369 references identified as core, studies were excluded if: they contained no outcome data; depression or sexual orientation were thematically peripheral; non-standardised measures of depression were used; depression was subsumed under other categories of analysis; or non-heterosexual sample sizes were less than 50 (except for those that sampled specific and harder-to-reach populations such as non-heterosexual parents, or those studies that typically have smaller sample sizes, such as small-group intervention studies). The exclusion criteria relating to non-standardised measures for depression and small sample sizes were not applied to qualitative studies. This left 59 studies. Of these, 54 were empirical, three were literature reviews and two were intervention studies. Despite relaxing the criteria in relation to qualitative studies, only one qualified for inclusion.

Policy analysis

A review of current Australian national health policies and plans for mental health investigated how these deal with depression and any other mental health needs of non-heterosexual people. A desktop search via the internet was conducted in three phases to locate documents that were then subjected to a content analysis. A search was conducted to locate publicly available Commonwealth Government policies or strategies that address mental health. An initial search of the Department of Health and Ageing websites (http://www.mentalhealth.gov.au and http://www.health.gov.au) located eight documents that were profiled as key policies or strategies. Another key national report on mental health was also located. A search was then conducted of the same government websites to locate any other publicly available national policies, strategies, reports or research that specifically address non-heterosexual people. Searches were conducted using the terms ‘gay’, lesbian’, ‘bisexual’, ‘homosexual’, ‘same sex’. Each search generated between 140 and 200 items and all of these were scanned for relevance. Items dated before 2000 were excluded. This produced only 12 items. Finally, a search was conducted for Commonwealth Government strategies where a possible policy linkage could be made to the mental health of non-heterosexual people. These included policies addressing men's health, women's health, sexually transmissible infections and HIV.

Results

The literature review revealed a number of methodological limitations and clear gaps in the literature.

Methodological limitations

Measuring sexual orientation

Reviewing and comparing empirical studies of non-heterosexual groups becomes complex due to different ways of measuring sexual orientation, based on identity, behaviour or attraction. Of the studies that measured sexual orientation based on self-identification, several included a question designed to identify participants as heterosexual, homosexual or bisexual, or predominantly heterosexual, homosexual or bisexual. Five studies aimed to capture more complex levels of self-identification by exploring how people situated themselves between categories.2–6 Some studies did not determine the self-identified sexual orientation of research participants through questionnaires but, rather, through pre data-collection screening or recruitment advertising. It was often unclear how sexual identity was measured.

Measuring sexual behaviour may increase the likelihood of identifying harder-to-reach, same-sex-attracted, sexually active populations that do not identify as GLB. However, it can lead to misclassification, wherein women or men are categorised based only on their sexual behaviour, which may result in selection bias: over-representation of people who are distressed by their sexuality and, consequently, who may exhibit higher rates of psychiatric morbidity.7,8 Another problem is variation in how sexual behaviour is defined. For instance, Gilman et al. determined same-sex or other-sex sexual behaviour by questioning participants whether they had had ‘sexual intercourse’.9 A variety of time frames have also been used (i.e. within 12 months, five years, or over one's lifetime). Finally, measuring sexual orientation through behaviour can distort sampling (e.g. young people not yet engaged in sexual activity).

There is a shortage of research that measures sexual orientation in diverse ways, partly due to limited theoretical understandings of sexuality, and also because small sample sizes reduce the possibility of exploring differences between non-heterosexuals.10 In the small number of studies that utilised multiple measures of sexual orientation, this was most often a way of capturing broader and larger samples rather than exploring the differences in levels of depression within non-heterosexual groups. The majority of studies collapsed participants into firm categories. Bisexual or both-sex-attracted men and women were often collapsed into gay or lesbian sub-samples, or were excluded.7,11,12 Those people unsure of or questioning their sexual orientation were mostly excluded.7,9

Measuring depression

Difficulties in comparing studies and producing general empirical conclusions across studies were compounded because a variety of terms and definitions were used for depression. It was often unclear whether it was defined as an emotional state, a mood, a cluster of symptoms, or a clinical diagnosis. Only one study gave an explicit definition of depression and the variety of different types, drawing from the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) criteria.13 Diagnostic interviews that assess mental disorders based on DSM criteria also have methodological issues, such as the variety of different interview schedules utilised or clinical diagnoses based on different versions of the DSM. The most popular scale used to measure non-clinical depression was the Center for Epidemiologic Studies Depression Scale (CES-D), but it was used in different ways. The CES-D only measures short-term prevalence of depressive symptoms.14

Comparison with heterosexual controls

An important methodological limitation relates to comparators employed by studies reviewed. Twenty-seven studies compared non-heterosexual samples with heterosexual samples, either from corresponding representative population studies, or matched heterosexual controls. Five studies did not use heterosexual samples or control groups, but compared rates of depression in non-heterosexual people with the general population. Twenty-three studies did not make any comparisons between non-heterosexual samples and heterosexual populations as they were interested in examining the variability within non-heterosexual groups, or identifying particular risk and protective factors for depression in non-heterosexual people. The two intervention studies lacked comparison groups.15,16

Prevalence of depressive symptoms and clinical depression

To capture the potential variability between different groups, rates of depressive symptoms and clinical depression in non-heterosexual people were explored separately according to the following categories: (1) young people; (2) lesbian and other homosexually active women; (3) gay and other homosexually active men; (4) bisexual people; and (5) general GLB populations.

The literature on depressive symptoms in young people is robust and reveals that gay/lesbian and same-sex-attracted young people exhibit significantly more depressive symptoms than heterosexual and other-sex-attracted young people. This was consistently found across studies from the United States, Canada, New Zealand, the Netherlands, Hong Kong and Australia.3,10,17,18 Both the international and Australian literature strongly suggests that both-sex-attracted and bisexual young people, and those questioning their sexual orientation, are at higher risk of suffering depressive symptoms.18,19

Lesbian and other homosexually active women consistently experienced higher rates of depression than heterosexual women, however, the literature reports mixed results in relation to time frames. Five studies that measured past-year major depression indicated that rates were higher for lesbian and other homosexually active women, however, some showed statistically insignificant results.7,9,12,20,21 The results for longer prevalence of major depression were stronger. A longitudinal study in New Zealand found that rates of major depression over a four-year period increased significantly according to increasing non-heterosexual identity.22 Moreover, all the studies that measured lifetime prevalence of major depression showed significantly higher rates for lesbian and other homosexually active women compared with heterosexual women.

Studies on depression in gay and other homosexually active men show a trend toward higher rates of depression compared with heterosexual men that also vary according to prevalence time frames. Across four representative population studies, 12-month prevalence rates of major depression were higher for homosexually active men compared with heterosexual men. Three of the four studies showed significant differences.7,9,20,21 Although gay and other homosexually active men still show higher lifetime prevalence rates of major depression than their heterosexual counterparts, these differences were not always significant.22,23

Although HIV infection may lead to higher rates of depression and depressive symptoms in both heterosexual and non-heterosexual men, five international studies in the United States, the Netherlands and Australia suggest that higher rates of depressive disorders in gay and other homosexually active men cannot be accounted for simply by HIV status.9,13,20,21,24 Nevertheless, one study did find a significant difference in depressive symptoms between HIV positive and HIV-negative, same-sex-attracted men, highlighting the need for more research.25

The few studies that did specifically explore bisexual people separately from gay, lesbian and same-sex-attracted categories consistently showed that they have higher rates of depression or depressive symptoms than heterosexual people and, in some cases, are at the same or higher risk of depression than homosexuals.4,22,26

The literature on general GLB populations confirms these trends. A key finding extracted from the Australian Bureau of Statistics National Survey of Mental Health and Wellbeing in 2007 was that homosexual/bisexual respondents had higher levels of affective disorders (which include bipolar disorder, major depression and dysthymia) than heterosexual respondents.27

Risk and protective factors for depression

The literature on risk and protective factors for depression in non-heterosexual people was reviewed to explore the psychosocial processes that make non-heterosexual people more vulnerable to depression. A key problem was a lack of research methodologies that allowed for the identification of causal factors. Several studies included multivariate and correlation analyses, both of which can establish an association or relationship between variables but not causality. Therefore, the findings reported below relate only to potential predictors of depression and depressive symptoms.

Residential context (living in metropolitan, suburban or rural areas) predicts depressive symptoms in some non-heterosexual samples, particularly when factoring in other variables such as sex. Two studies suggested that non-heterosexual young people (particularly young women) and HIV-positive gay/bisexual men are more vulnerable to depression when living in rural or suburban contexts.18,25

Being in a relationship appears to be a significant protective factor.5,12,28 However, relationship status is also reported as an influential variable for heterosexual people. The results of some studies suggest that higher rates of depressive symptoms in non-heterosexual people persist after factoring in demographic variables such as relationship status. Social support from peers, friends and family emerges as a fairly robust protective factor against depressive symptoms in non-heterosexual samples, particularly for young people. There is some evidence to suggest that bisexual people may receive less social support than both heterosexuals and homosexuals, which may contribute to poorer mental health outcomes.26,28 Feeling that one belongs to a community is a protective factor against depression in both heterosexual and non-heterosexual people. In some studies, non-heterosexual people reported a lower sense of belonging to the general community than heterosexuals which was linked to significantly more depressive symptoms.28 Feeling a sense of belonging to the GLB community can be an important protective factor against depression in GLB people, particularly for gay men.11,29

The literature that examined disclosure of sexual orientation and its effects on depressive symptoms reveals contradictory results. Some studies show no relationship, while others suggest that disclosure to family and friends predicts fewer depressive symptoms.3,14 Disclosure may be a protective factor by lowering a sense of isolation from the GLB community. However, concealing one's sexual identity might also be a coping mechanism or even a protective factor.30

Victimisation is a significant risk for non-heterosexual people, particularly young people. D'Augelli found that 81% of an international sample had been the victims of verbal abuse because of sexual orientation, 15% had suffered serious physical assault and 16% had experienced sexual assault.3 A history of verbal, sexual or physical victimisation and abuse is associated with higher levels of depressive symptoms in non-heterosexual people.24,29

In terms of stigma and discrimination, the literature notes that depressive symptoms in non-heterosexual people are associated with both ‘distal’ discriminatory events (objective social acts or events in which non-heterosexual people are subjected to prejudice and discrimination), and ‘proximal’ processes (which include the subjective interpretation of discriminatory events, and the internalisation of discrimination and negative stereotypes).8 All the studies that measured ‘distal’ events showed that discrimination based on sexual orientation was associated with higher rates of depressive symptoms. Lewis et al. found that general discrimination, which included lack of access to mental health, housing and social services, was significantly associated with higher depressive symptom scores, as was discrimination at work (which included actual and potential job loss).29 The literature on ‘proximal’ processes focuses predominantly on the effects of ‘internalised homophobia’, and associates this with more depressive symptoms.31,32 Overall, the literature on ‘proximal’ processes greatly outnumbers the literature on ‘distal’ discriminatory acts and events, reflecting an overall tendency to focus on the psyches of non-heterosexual people rather than exploring homophobia as a sociological problem associated with depression.

Problems associated with depression

Just as it is difficult to determine a causal link between depression and risk factors, it is difficult to establish outcomes of depression due to a lack of appropriate research methodologies. Thus, it was unclear whether depression was the cause or consequence of behaviours often connected to poor mental health. Nevertheless, the literature suggests that depression is linked to three key outcomes among non-heterosexual people: sexual risk; alcohol and drug use; and suicidal behaviours.

Although depression and depressive symptoms in gay and other homosexually active men have been consistently associated with high-risk sexual behaviours in the literature, this review found mixed results. Some studies found that sexual risk was not associated with depressive symptoms, while others suggest a connection.13,24,30

Although non-heterosexual people (both men and women) tend to have higher rates of alcohol and drug use and dependency than heterosexual people, these differences are not always significant; however, the evidence that bisexuals suffer from substance use and dependency disorders at higher rates is more robust.8,26,33 The literature notes that substance use may be a coping strategy to deal with the stress related to living in homophobic environments.

Depression has been recognised as an important predictor of suicidal behaviour (e.g. suicidal thoughts, plans and attempts) in heterosexual populations and the same pattern is true for non-heterosexual people, particularly young people. It appears that higher prevalence rates of depression and depressive symptoms in non-heterosexual people are likely to predict higher rates of suicidal behaviour compared with heterosexual people; however, these findings are not always significant.6,9,22 Again, bisexuals appeared to be at similar or higher risk.4

Interventions

Surprisingly, few intervention studies aimed at reducing levels of depression in non-heterosexual people were found. There are many examples of intervention studies on stress management in HIV-positive people, AIDS-related bereavement, amphetamine use in gay men, and the effects of interventions on levels of depression.34–36 These interventions, however, were not concerned with sexual orientation per se. Several studies that explored the risk factors for depression in non-heterosexual people noted the potential for interventions. Nevertheless, there were only two intervention studies that targeted depression in these people. One study involved a psychosocial intervention drawn from cognitive behavioural therapy models and including content specific to ‘coming out’ and internalised homophobia.15 The other investigated the role of social support and ‘confidants’ in reducing depression by tracking this over time among young people attending gay and lesbian holiday camps.16 Both studies reported some positive results in terms of reducing depressive symptoms, but were uncontrolled and inconclusive in terms of their implications for future interventions. There were no social intervention studies targeting victimisation, stigma, discrimination and homophobia.

Literature gaps

A clear gap in the research literature is an understanding of the depth and complexity of non-heterosexual experiences and the social determinants of non-heterosexual depression and mental health disorders. Several qualitative studies were identified that focused on developmental issues and mental health concerns for non-heterosexual people; however, depression consistently appeared as a peripheral issue or was largely absent.37–40 There was also a dearth of theoretically informed literature relating specifically to depression in non-heterosexual people. The exception was a literature review and meta-analysis by Meyer, which provided a theoretical review of minority stress models in non-heterosexual research.8 There is also a significant gap in terms of assessing the efficacy of interventions, and it is unclear whether this reflects a lack of existing interventions.

Results from the policy analysis

The national policy response in Australia was analysed to deepen an understanding of whether, where and how these findings on depression in non-heterosexual people are addressed within a specific policy context. There is no specific national policy response to depression. It is dealt with as part of the broader policy response to mental health. This framework sets the basis for programs conducted at federal and state level, and the prioritisation of funding for broader initiatives implemented through non-government bodies.

Key national policies/strategies

There is no mention of gay, lesbian, bisexual or non-heterosexual people in the current National Mental Health Policy, the National Mental Health Plan, the Council of Australian Governments’ National Action Plan on Mental Health, or the National Mental Health Report.41–44 Two other key Commonwealth policy/strategy documents for mental health involve Aboriginal and Torres Strait Islander (ATSI) communities and veterans, and so do not address non-heterosexual populations.45,46 However, both the National Mental Health Policy and National Mental Health Plan outline an approach to population groups where the prevalence of mental health problems is higher, and where stigma and discrimination is a relevant associated factor. The policy sets out a population health framework that recognises the importance of social determinants of mental health and illness. Risk factors are identified as occurring at an individual and a community level, with the latter involving social exclusion, discrimination and bullying. Population groups in need of particular attention and coordinated efforts to achieve social inclusion are identified as ATSI peoples, people who are homeless, disadvantaged children, unemployed, newly arrived or refugees.41 A similar list can be drawn from the Council of Australian Governments’ National Action Plan on Mental Health, which outlines major funding commitments for each State and Territory for 2006–2011. Populations targeted in initiatives include children and young people, ATSI peoples, culturally and linguistically diverse populations, the homeless, and people in correctional facilities.43

The National Mental Health Policy outlines the need for a multi-sectoral response, with a range of prevention, treatment and support services to be provided by the mental health sector, and other sectors as appropriate. The policy argues that services should be integrated, tailored and evidence-based.41 It advocates specific, community-controlled services for at-risk groups, e.g. delivery of culturally appropriate services to ATSI populations. In other cases, cultural and linguistic diversity is advocated to form part of workforce training. The policy highlights the need to draw on the existing evidence base to identify incidence, prevalence and improve the capacity of the mental health sector to respond. The implication overall is that evidence in relation to mental health in non-heterosexual populations is not considered strong enough to justify definition as a high-risk group. This is despite government-funded research, uncovered through keyword searches of government websites, from 2000 and 2004 that notes the increased risk of mental health issues due to stigma and discrimination for non-heterosexual people and particularly young non-heterosexual people.47,48

In only two other key national policy documents are non-heterosexual people considered: the National Suicide Prevention Framework and National Suicide Prevention Strategy recognise ‘gay and lesbian’ populations as an at-risk group.49,50 Specific actions suggested include improving access to services, and education and information for service providers and those in contact with at-risk groups. Documents located through the keyword searches of government websites show that a Senate enquiry in 2010 into suicide in Australia recommended that a national prevention and awareness campaign should target recognised high-risk groups, including: young people, people in rural and remote areas, men, Indigenous populations, lesbian, gay, bisexual, transgender and intersex people, and the culturally and linguistically diverse communities.51 However, the government response to this recommendation was to accept it with qualification, noting that there was insufficient evidence to support a targeted awareness campaign, as well as the need to balance resource allocation to ‘upstream population health approaches’ and more targeted efforts for at-risk groups.52 Nevertheless, the Federal government funded a $22.6 million national initiative of community prevention activities for high-risk groups (including gay, lesbian and bisexual people) from 2011.

Policy links

In terms of other Commonwealth policies where a link could exist in addressing mental health in non-heterosexual populations, the results showed that attention to the issue was limited, included as one issue among many or with no practical effect on interventions.

The National Male Health Policy notes that some men have poorer health outcomes, including gay, bisexual or transgender males, and intersex people, as well as men with mental health issues.53 As such, these groups are considered within the policy priorities, with actions including making information culturally appropriate, improving access to healthcare services and promoting research. The Women's Health Policy recognises sexuality in terms of social determinants of health, noting that lesbian and bisexual women are at greater risk of poor health outcomes and mental health issues.54 The policy promotes the need to reduce barriers to accessing services, in terms of cultural appropriateness and addressing discrimination within services based on sexuality or the assumption of heterosexuality. However, neither policy includes any specific initiatives to address these issues. The National HIV Strategy considers the mental health needs of those at risk of or living with HIV and how to address these with mental health services, but has no specific mention of mental health needs of non-heterosexual people.55 The National STI Strategy does not consider mental health issues.56

The keyword searches of government websites uncovered a clear policy link being drawn in terms of the mental health of non-heterosexual populations and drug use. The National Drug Strategy outlines the link between disadvantage and marginalisation and drug use, including among gay, lesbian, bisexual, transgender and intersex populations.57 However, no specific interventions are proposed. Other reports of government-funded research documented the risks of methamphetamine and injecting drug use in non-heterosexual populations, and the link between higher rates of drug use and higher rates of depression and greater risk-taking behaviour.58,59

A number of documents were located that considered the mental health of non-heterosexual people in specific contexts. These included issues such as dealing with bereavement grief in HIV/AIDS, supporting GPs to deal with those presenting with sexual problems with sensitivity to sexuality, addressing the issue of homophobia in sport and the negative consequences on health and wellbeing (particularly among same-sex-attracted young people), and dealing with dementia in non-heterosexual people.60–64 However, none of these was specifically related to policy in terms of addressing the mental health of non-heterosexual populations.

Conclusion

The results suggest that the evidence base on higher rates of depression in non-heterosexual people is strong. However, the weight of this evidence is not currently recognised within national policy in Australia. Most major policy documents do not consider this population, and where the needs of this population are recognised, this does not then feed back into national mental health policy. The exception to this trend is the inclusion of non-heterosexual people as a risk group in the National Suicide Prevention Policy and Action Framework, and this is consistent with the evidence that higher prevalence rates of depression and depressive symptoms in non-heterosexual people are likely to predict higher rates of suicidal behaviour in this population compared with heterosexual people. However, the policy response to suicide deals with an outcome linked to depression in this population, and is not focussed on relevant associated and underlying social factors in terms of homophobia. Similarly, the place of non-heterosexual people within the National Drug Strategy is consistent with the evidence that suggests non-heterosexual people (both men and women) have higher rates of alcohol and drug use and dependency, along with depression. However, if the evidence suggests that substance use is associated with the stress of living in homophobic environments, then this policy response does not deal with underlying social factors either.

The National Mental Health Policy framework does include a focus on at-risk groups where stigma and discrimination is a described as a relevant factor in determining higher rates of mental illness. It includes a focus on developing culturally appropriate and targeted interventions in these populations, while maintaining a broad population approach. This sets the precedent for the specified inclusion of the mental health needs of non-heterosexual people in the Australian policy context. In general, policy links that might support responses to depression and other mental health needs in non-heterosexual people are weak, and demonstrate the lack of a coherent policy approach. There is also no national policy focus on sub-groups that the evidence suggests are at particular risk: same-sex-attracted young people, and bisexual people.

Implications

A national policy approach of defining non-heterosexual people as an ‘at-risk’ group for depression is consistent with the evidence. Prioritising access to mental health services that are socially and culturally appropriate for this population should be included in the national policy response to mental health. More research and consideration is needed of how non-heterosexual people use mental health services in order to determine the right mix of specific services, targeted interventions and sensitisation of general mental health services to cater for non-heterosexual people. The literature on protective factors specific to this population (e.g. social support and feelings of community belonging) could be drawn upon to support practical initiatives. However, a broader and more difficult question is how to formulate an effective national policy response that deals with underlying social factors in groups identified as at greater risk of depression because of stigma and discrimination. In other populations recognised as being at risk of poorer health outcomes (e.g. Aboriginal and Torres Strait Islanders), specific government offices, policies, strategies and funding streams have been put in place to ensure prioritisation, co-ordination and implementation of interventions. However, there are currently no mechanisms of this type that are specifically relevant to non-heterosexual people. Addressing homophobia, stigma and discrimination as underlying social factors associated with higher rates of depression would require a strategic policy approach across a range of sectors.

The findings presented here are consistent with the broader literature that analyses the link between research evidence and policy practice, and has found this link to be particularly weak in the area of health services.65 This has led some to focus on the need to promote partnerships and ‘knowledge exchange’ between researchers and policy makers.66 As an indication of the need for a two-way flow of information in determining research and policy directions, this study also found that there is an important gap in the academic literature in terms of analysing homophobia as an associated or causal factor for depression in non-heterosexual people, thereby weakening the evidence base to be drawn upon in arguing for effective policy and interventions. Addressing this gap in the literature, as a starting point, would build the evidence base and assist in identifying options for an appropriate policy response. One practical step to bridge the gap between evidence and policy practice would be to convene a national working party involving relevant government sectors, practitioners, researchers and community representatives to consider effective policy responses and program initiatives to address depression and other mental health needs of non-heterosexual people.

Acknowledgements

Some analyses were supported as part of project conducted by a research partnership between beyondblue, the national depression initiative, and the Australian Research Centre in Sex, Health and Society. The result of this partnership was the report Feeling queer and blue.67 Anne Mitchell, Murray Couch, Paul Agius and Marian Pitts also contributed to this project.

Ancillary