Disparities in mental health service use in Australia: Analysis of the National Study of Mental Health and Wellbeing (2020–2022)

Abstract

Objectives:

We aimed to quantify inequalities in mental health service use in Australia and identify the sociodemographic, economic, and geographic factors associated with access.

Methods:

We conducted a secondary analysis of the National Study of Mental Health and Wellbeing (NSMHWB, 2020–2022). We evaluated 12-month mental health service use before the survey among people who reported a mental disorder. We fitted multivariable logistic regression models adjusted for sociodemographic covariates and tested the joint (intersectional) effects of remoteness and socioeconomic area disadvantage on service use. Jackknife replicate weights accounted for the complex survey design.

Results:

The study included 15,893 respondents (weighted population: 19,828,348). Lifetime prevalence of any mental health disorder was 42.9% (95% CI = [42.5, 43.3]), and 21.5% (95% CI = [20.7, 22.3]) reported symptoms in the past 12 months. Among those with a 12-month disorder, 45.2% (95% CI = [39.9, 50.5]) had accessed care, and 16.5% (95% CI = [15.8, 17.2]) reported using digital or telephone services. Service use was higher in university graduates and residents of advantaged areas and lower in remote/outer regional residents and the most disadvantaged quintiles. Intersectional analysis revealed compounded inequities, with the lowest service use observed among socioeconomically disadvantaged populations in remote areas and the highest among advantaged major-city residents.

Conclusion:

Substantial and intersecting inequalities exist in access to mental health services, with the lowest coverage among remote, socioeconomically disadvantaged, older, and male populations. Targeted, place-based strategies – including expanded outreach and tailored digital and community-based models – are needed to reduce unmet need and improve equity.

Introduction

Mental health conditions are among the most prevalent and disabling health issues in Australia, accounting for 15% of the total burden of disease in 2024 (Australian Institute of Health and Welfare, 2025). Anxiety, depression, and substance use disorders affect millions of Australians each year, with rising trends in psychological distress particularly prevalent among younger populations (Australian Bureau of Statistics (ABS), 2023a). The societal and economic impacts are extensive, including reduced quality of life, increased healthcare use, and lost productivity (Mccallum et al., 2019). Despite heightened awareness and growing public discourse around mental health, population-level improvements in prevalence and distress indicators have remained limited (Crosland et al., 2025; Enticott et al., 2022).

Australia’s mental healthcare system is a multi-layered mix of universal primary care, specialist clinical services, psychosocial supports, and disability programmes. Medicare funds most access to psychologists and psychiatrists under a public insurance model, while state and territory governments deliver public hospital–based care, crisis services, and community mental health teams. General practitioners remain the main entry point, preparing mental health treatment plans and referring to psychologists, psychiatrists, and allied professionals. Alongside these clinical pathways, community-managed organisations, peer workers, housing and employment supports, and digital mental health services provide psychosocial care. Public mental health services offer intensive, multidisciplinary support for people with severe and persistent illness, complemented by the National Disability Insurance Scheme for those eligible for psychosocial disability supports (Pirkis et al., 2022).

A growing body of research highlights significant disparities in mental health service utilisation across socioeconomic, geographic, and demographic lines. Educational attainment has emerged as a key determinant of access: individuals with higher education levels are markedly more likely to use services (Black et al., 2025). Although Medicare aims to reduce financial barriers, access to psychotherapy and specialist care remains disproportionately higher among wealthier Australians (Bartram and Stewart, 2019; Meadows et al., 2019). Geographic remoteness compounds inequity; residents in rural and regional areas often face reduced access to psychological services, relying more heavily on pharmacological treatment alone (Amos et al., 2023; Bartram and Stewart, 2019; Meadows et al., 2019). For Aboriginal and Torres Strait Islander peoples, access is further shaped by structural disadvantage and distance from specialised facilities, while newly resettled refugees encounter cultural stigma and language barriers that hinder engagement with care (Byrow et al., 2019; Mazumdar et al., 2022). These intersecting barriers suggest that existing systems may not be adequately aligned with the needs of underserved groups, contributing to inequities in mental health outcomes.

However, the evidence on social determinants of mental health service use remains inconsistent across studies. Some report strong effects for specific determinants, such as education or remoteness, while others find weak or non-significant associations, particularly for income or socioeconomic status as determinants of mental health service use (Amos et al., 2023; Bartram and Stewart, 2019; Black et al., 2025; Meadows et al., 2019). For instance, Black et al. (2025) analysed the longitudinal Household, Income and Labour Dynamics in Australia survey and found substantial education-related inequality but no significant association with income or financial hardship, highlighting the complex and context-specific nature of these relationships (Black et al., 2025).

Moreover, although increasing attention has been given to individual-level determinants of mental health service use, few studies have adopted an intersectional approach to examine how multiple equity-related factors interact to shape access across service types and population subgroups (Black et al., 2025; Hashmi et al., 2023; Islam et al., 2023). To this end, we sought to address this gap by examining patterns of mental health service utilisation through an equity lens using data from the 2020–2022 National Study of Mental Health and Wellbeing (NSMHWB 2020–2022) – the most recent nationally representative survey of mental health in Australia – with a focus on key demographic, socioeconomic, and geographic variables.

Methods

Data source

We conducted a secondary analysis of the NSMHWB, 2020–2022, a nationally representative household survey conducted by the ABS. The NSMHWB was designed to provide comprehensive estimates of the prevalence, correlates, and service use patterns associated with common mental health-related disorders among people aged 16–85 years. The survey employed a cross-sectional design using stratified, multistage probability sampling to ensure representativeness across geographic regions, age groups, and sociodemographic strata (ABS, 2023b).

Sample and data collection

The overall response rate was 52%, with 15,893 fully responding households out of 30,586 approached. The survey excluded non-private dwellings (e.g. institutions and hospitals) and individuals unable to complete the interview due to language or cognitive barriers.

Trained interviewers conducted face-to-face computer-assisted personal interviews (CAPI), supplemented by self-completion modules for sensitive topics. Diagnoses of mental disorders were derived using the WMH-CIDI 3.0 algorithms, which apply criteria from two classification systems: the Diagnostic and Statistical Manual of Mental Disorders (4th ed; DSM-IV) and the World Health Organization’s International Classification of Diseases, Tenth Revision (ICD-10). We used the ICD-10 classification for the purpose of this study. The survey collected information on lifetime and 12-month prevalence of mental health-related disorders, which were aggregated to three main groups: anxiety, affective, and substance use disorders.

Variables and measurement

Outcome variable

To explore inequalities in mental health service use, we examined two primary outcomes: lifetime service use and service use in the 12 months before the survey. Twelve-month service use reflects contemporary patterns of help-seeking and availability and forms the primary focus of the analysis. Lifetime service use was retained as a complementary indicator capturing overall engagement with the mental health system over the life course. This allows us to identify groups who have never accessed care despite meeting diagnostic criteria, highlighting a distinct and policy-relevant form of unmet need. Mental health service use refers to whether respondents with a lifetime history of any mental health disorder had ever accessed services, classified as (1) consultation only – mental health care through professional consultations (e.g. general practitioners, psychologists, and psychiatrists) without hospital admission; (2) hospitalisation only – admission to hospital for mental health treatment without consultations outside hospital settings (ABS, 2023b); or (3) both consultation and hospitalisation. Provider-specific patterns of service use in the NSMHWB 2020–2022 dataset have been examined elsewhere, demonstrating that access varies across psychologists, GPs, and psychiatrists (Harris et al., 2025). Our analysis, therefore, focuses on overall service use, drawing on this prior work to contextualise provider-level differences.

Stratifying variables

The PROGRESS-Plus framework guided the systematic selection and categorisation of equity-relevant variables for mental health service access (Neill et al., 2025). Core PROGRESS dimensions included place of residence, race and ethnicity, occupation, gender, education, socioeconomic status, and social capital, while the “Plus” component captured additional characteristics such as age, disability, and life-stage factors. Consistent with this framework, we identified and included the following available variables: remoteness, sex at birth, Indigenous status, culturally and linguistically diversified (CALD) status, educational status, area-level socioeconomic position (Index of Relative Socioeconomic Advantage and Disadvantage, IRSAD), employment status, long-term health conditions, age, and marital status.

Remoteness was measured using the Australian Statistical Geography Standard Remoteness Structure, which classifies areas according to relative access to services based on the Accessibility/Remoteness Index of Australia Plus. Respondents were grouped into four categories: Major Cities, Inner Regional, Outer Regional, and Remote Australia (ABS, 2021a).

Socioeconomic conditions were measured using the IRSAD. IRSAD summarises information about the economic and social resources of people and households within an area, drawing on indicators such as income, education, and occupation. IRSAD scores were grouped into quintiles, from Quintile 1 (most disadvantaged) to Quintile 5 (least disadvantaged) (ABS, 2021b).

Education level was measured using the Australian Standard Classification of Education, with respondents grouped into three categories consistent with ABS reporting: Secondary or lower, Certificate/Diploma, and University degree (ABS, 2001). We also included a range of sociodemographic and economic variables: age group, sex at birth, Indigenous status, culturally and linguistically diverse background, marital status, employment status, and severity of mental health disorder. Severity of mental health disorder was classified using the Australian version of the WMH-CIDI 3.0 severity measure, which was applied to respondents with a 12-month disorder. Three levels were derived. Severe required at least one of a diagnosis of bipolar affective disorder, substance dependence with serious role impairment, a suicide attempt in conjunction with any mental disorder, at least two domains of severe role impairment on the Sheehan Disability Scale, or a Global Assessment of Functioning score of 50 or less. Moderate was assigned to those not meeting the severe threshold who reported at least moderate interference on any Sheehan Disability Scale domain or had substance dependence without substantial impairment. Remaining respondents with a 12-month disorder were classified as mild.

Statistical analysis

Analyses were undertaken within the secure ABS DataLab environment using the Person Level Integrated Data Asset (PLIDA). Survey weights supplied by the ABS were applied to account for the complex survey design. Jackknife replicate weights were applied to correct for unequal probabilities of selection and clustering, ensuring that estimates were representative of the Australian population and that variance was robustly estimated. Geographic differences in mental health status and service use across Australian states and territories were illustrated using choropleth maps.

To examine inequities in mental health service use during lifetime and 12-month before the survey, we fitted multivariable logistic regression models adjusting for the following covariates: age group, sex at birth, Indigenous status, culturally and linguistically diverse background, marital status, employment status, engagement in education or employment, presence of a long-term health condition, highest level of educational attainment, area-level socioeconomic position (IRSAD), and remoteness of residence.

We further analysed the interaction effect to test whether the effect of remoteness varied by area-level socioeconomic position, with particular focus on the joint influence of remoteness and IRSAD, informed by evidence that geographic and socioeconomic disadvantage operate jointly to compound barriers to mental health service access (Amos and Coleman, 2023; Australian Institute of Health and Welfare, 2026). These variables capture distinct structural constraints – remoteness limits physical access to services, while socioeconomic disadvantage restricts the financial and social resources needed to navigate care – and their intersection represents a policy‑priority population in national mental health frameworks. Other interactions, such as sex by remoteness and age by remoteness, were not the primary focus of this analysis but remain important directions for future research. All analyses were conducted using Stata (version 18) and R.

Results

Prevalence of mental health disorders

A total of 15,893 respondents aged 16–85 years completed the survey, representing a weighted population of 19,828,348 Australians. Overall, 42.9% of Australians (95% CI = [42.5, 43.3]) reported ever experiencing a mental health disorder, with 21.5% (95% CI = [20.7, 22.3]) reporting symptoms in the past 12 months. Among those with a 12-month disorder, 31.4% (95% CI = [27.8, 35.3]) reported experiencing mild symptoms, 41.6% (95% CI = [39.7, 43.6]) moderate symptoms, and 27.0% (95% CI = [23.4, 31.0]) severe symptoms.

Prevalence varied by demographic characteristics. Lifetime prevalence was highest among those aged 15–24 years (48.5%, 95% CI = [46.8, 50.3]) and lowest among those aged ⩾65 years (29.5%, 28.8–30.2), with a similar age-related pattern for 12-month prevalence. Females had a higher 12-month prevalence than males (24.6% vs 18.3%), while lifetime prevalence was similar (43.6% vs 42.2%). Across jurisdictions, lifetime prevalence ranged from 40.2% in South Australia to 52.7% in the Northern Territory and 12-month prevalence from 19.5% in New South Wales to 29.4% in the Northern Territory. Socioeconomic differences were modest, with lifetime prevalence ranging from 46.0% in the most disadvantaged areas to 40.8% in the most advantaged (Table 1).

Table 1.

Prevalence of lifetime and 12-month mental health disorders by sociodemographic characteristics.

Variable	Lifetime		12 months preceding the survey
	Total	Percent with mental disorder [95% CI]	Total	Percent with mental disorder [95% CI]
Age groups
15–24	2,710,838	48.5 [46.8, 50.3]	2,710,838	39 [37.3, 40.6]
25–44	7,134,126	46.9 [46.6, 47.1]	7,134,126	23.4 [22.3, 24.6]
45–64	6,188,448	44.2 [42, 46.3]	6,188,448	19 [18.5, 19.6]
>65	3,794,936	29.5 [28.8, 30.2]	3,794,936	9.5 [9.4, 9.6]
Sex at birth
Male	9,761,056	42.2 [41.2, 43.2]	9,761,056	18.3 [17.8, 18.8]
Female	10,063,773	43.6 [42.3, 45]	10,063,773	24.6 [23.4, 25.9]
State/territory
New South Wales	6,261,319	40.6 [39.8, 41.3]	6,261,319	19.5 [19.2, 19.8]
Victoria	5,127,162	42.4 [42, 42.8]	5,127,162	21.3 [20.9, 21.7]
Queensland	4,021,011	45.2 [45.1, 45.3]	4,021,011	23.6 [22.2, 25.1]
South Australia	1,406,715	40.2 [37.2, 43.2]	1,406,715	21.5 [19.1, 24]
Western Australia	2,073,981	45.5 [43.6, 47.5]	2,073,981	23.1 [21.8, 24.5]
Tasmania	448,219	52.2 [51.3, 53]	448,219	19.8 [19.3, 20.3]
Northern territory	140,747	52.7 [49.9, 55.6]	140,747	29.4 [20.9, 39.8]
Australian Capital Territory	349,194	46.6 [40.3, 53.1]	349,194	25.4 [22.1, 29]
IRSAD
Quintile 1(most disadvantaged)	3,150,030	46 (40–46.9)	3,150,030	22.5 [21.5, 23.6]
Quintile 2	3,580,049	44 (40.6–47.5)	3,580,049	22.4 [20.1, 24.9]
Quintile 3	4,191,872	44.4 [42.7, 46]	4,191,872	21.9 [20.6, 23.4]
Quintile 4	4,205,501	40.6 [40.1, 41.2]	4,205,501	20 [19.4, 20.5]
Quintile 5 (least disadvantaged)	4,700,896	40.8 [40.8, 40.9]	4,700,896	21.1 [21, 21.3]

Mental health service use

Among individuals who experienced a mental health disorder at some point in their lifetime, 65.5% (95% CI = [65.0, 66.1]) accessed mental health services, with the majority (85.5%, 95% CI = [84.7, 85.5]) receiving consultation services from a health professional. In the 12 months preceding the survey, 45.2% (95% CI = [39.9, 50.5]) accessed mental health services, mainly through consultations. Among all mental health service users, 16.5% (95% CI = [15.8, 17.2]) reported receiving digital or telephone-based services (Supplemental Table 4).

Mental health service utilisation varied across demographic and socioeconomic groups (Table 2). Lifetime service use was similar across most age groups (approximately 66–67%) but lower among those aged ⩾65 years; a comparable pattern was observed for 12-month service use. Females were more likely than males to use services for both lifetime (74.4% vs 56%) and 12-month disorders (51.3% vs 36.7%). Differences by education, employment status, remoteness, and socioeconomic position were relatively modest.

Table 2.

Mental health service use among individuals with a lifetime and 12-month mental disorder by socioeconomic characteristics (weighted).

Variable	Category	Lifetime		12 months preceding the survey
		People with mental disorder	Proportion using mental health services [95% CI]	People with mental disorder	Proportion using mental health services [95% CI]
Age category	15–24	1,315,434	67.1 [66.2, 68.2]	1,053,466	43.7 [40, 47.4]
	25–44	3,338,872	67.2 [66.2, 68.2]	1,667,760	51 [47.1, 54.9]
	45–64	2,731,436	66.2 [61.5, 70.7]	1,177,135	41.2 [40.7, 41.6]
	>65	1,118,841	56.9 [56.2, 57.5]	357,495	35.5 [32.9, 38.1]
Sex at birth	Male	4,112,991	56 [54.1, 57.9]	1,781,679	36.7 [35.8, 37.7]
	Female	4,388,073	74.4 [74.2, 74.6]	2,470,657	51.3 [47.9, 54.7]
Indigenous status	Indigenous	257,743	67.1 [57.7, 75.3]	129,285	45.3 [39, 51.8]
	Non-Indigenous	8,246,840	65.5 [64.6, 66.4]	4,126,570	45.2 [42.2, 48.1]
CALD	No	8,401,559	65.7 [64.4, 67]	NA	NA
	Yes	103,024	49.2 [34.9, 63.6]	NA	NA
Marital status	Married	3,601,532	60.3 [57.7, 62.9]	1,399,839	41.5 [40.8, 42.2]
	Previously married	1,325,468	76.2 [74.3, 78.1]	639,023	53 [46.2, 59.7]
	Never married	3,577,583	66.8 [66.3, 67.4]	2,216,993	45.2 [42.3, 48.2]
Highest level of educational attainment	Secondary or lower	3,327,284	63.1 [62.5, 63.8]	1,757,655	41.5 [38.1, 45]
	Certificate/Diploma	2,547,288	66.3 [61.2, 71.1]	1,213,216	48.3 [44.8, 51.7]
	University degree	2,630,011	67.8 [67.3, 68.3]	1,284,984	47.3 [45.8, 48.8]
Employment	Employed	5,802,708	64.6 [64.4, 64.8]	2,890,305	43.2 [41.6, 44.8]
	Unemployed	371,380	69.6 [69, 70.1]	223,867	48.5 [44.2, 52.8]
	Not in the labour for	2,330,496	67.1 [63.3, 70.7]	1,141,683	49.4 [42.6, 56.3]
Remoteness of residence a	Major cities of Australia	6,105,407	64.6 [63.4, 65.7]	3,148,599	45.4 [44.3, 46.5]
	Inner Regional Australia	1,557,408	61.1 [55.3, 66.6]	746,905	47.5 [39, 56.1]
	Outer Regional Australia	788,933	66.1 [64, 68]	360,351	38.2 [36.6, 39.9]
	Remote Australia	52,835	65.3 [61.4, 69]
IRSAD	Quintile 1 (most disadvantaged)	1,447,438	64.6 [63.4, 65.7]	708,787	44 [41.2, 46.8]
	Quintile 2	1,571,428	61.1 [55.3, 66.6]	799,567	41.9 [30.1, 54.7]
	Quintile 3	1,859,269	66.1 [64, 68]	919,663	47.4 [43.6, 51.2]
	Quintile 4	1,707,564	65.3 [61.4, 69]	838,475	42.7 [36.3, 49.2]
	Quintile 5 (least disadvantaged)	1,918,884	69.6 [68.7, 70.5]	989,363	48.7 [43.8, 53.6]

Note. CALD: Culturally and linguistically diversified; IRSAD: Index of Relative Socioeconomic Advantage and Disadvantage; NA: Not applicable – sufficient data on CALD was not available for 12 months preceding the study.

^aOuter Regional Australia and Remote Australia have been combined for the past 12 months.

Geographically, the highest prevalence of lifetime mental health disorders was observed in the Northern Territory (52.74%, 95% CI = [49.88, 55.59]) and Tasmania (52.16%, 95% CI = [51.30, 53.01]). Despite high prevalence, mental health service use was lowest in the same states, at 60.62% (95% CI = [52.43, 68.26]) and 61.99% (95% CI [56.09, 67.57]), respectively.

Similarly, the Northern Territory had the highest prevalence of mental health disorders 12 months preceding the study (29.43%, 95% CI = [20.87, 39.75]), followed by the Australian Capital Territory (25.43%, 95% CI = [22.12, 29.04]). Self-reported mental health service use across the Northern Territory, Tasmania, and the Australian Capital Territory combined was 42.43% (95% CI = [41.55, 43.31]), which ranked the lowest. In contrast, Victoria and Western Australia reported among the highest rates of 12-month mental health service use at 48.74% (95% CI = [45.17, 52.32]) and 47.91% (95% CI = [46.86, 48.97]), respectively (Figure 1).

Figure 1.

Mental health disorder and mental health service use by states and territories (NSMHW, 2020–2022).

Socioeconomic correlates of mental health service use

Table 3 summarises crude and adjusted logistic regression results for 12-month service utilisation. Lifetime regression results are reported in Supplemental Table 5. Substantial sociodemographic differences were observed. After adjusting for covariates, adults aged 65 years and over had significantly lower odds of 12-month service use (adjusted odds ratio [AOR] = 0.31, 95% CI = [0.17, 0.56]) compared with those aged 15–24 years.

Table 3.

Factors associated with 12-month mental health service use among people who reported a mental health disorder.

Variable	Category	Mental health service 12 months preceding the survey
		COR [95% CI]	AOR [95% CI]
Age groups	15–24	1	1
	25–44	1.34 [1.33, 1.36]	0.99 [0.88, 1.1]
	45–64	0.9 [0.64, 1.27]	0.5 [0.31, 0.81]
	>65	0.71 [0.66, 0.77]	0.31 [0.17, 0.56]
Sex at birth	Male	1	1
	Female	1.81 [1.5, 2.19]	1.55 [1.49, 1.61]
Indigenous status	Indigenous	1	1
	Non-Indigenous	0.99 [0.47, 2.12]	1.45 [1, 2.11]
CALD	No	NA	NA
	Yes	NA	NA
Marital status	Married	1	1
	Previously married	1.59 [1.18, 2.15]	1.54 [0.95, 2.49]
	Never married	1.16 [1, 1.35]	0.97 [0.82, 1.16]
Level of education	Secondary or lower	1	1
	Certificate/Diploma	1.32 [1.3, 1.33]	1.58 [1.52, 1.65]
	University degree	1.27 [1.07, 1.49]	1.67 [1.63, 1.7]
Long-term health condition	Yes	1	1
	No	0.21 [0.13, 0.33]	0.2 [0.14, 0.27]
Engagement in education or employment	Fully engaged	1	1
	Partially engaged	1.19 [1.13, 1.26]	0.96 [0.83, 1.1]
	No study or employment	1.13 [1.09, 1.17]	0.63 [0.62, 0.64]
Employment	Employed	1	1
	Unemployed	1.24 [0.77, 1.99]	1.32 [1.07, 1.63]
	Not in the labour force	1.28 [0.84, 1.96]	1.79 [1.25, 2.56]
Remoteness of residence	Major cities of Australia	1.34 [1.31, 1.38]	1.31 [1.3, 1.32]
	Inner Regional Australia	1.46 [1.11, 1.93]	1.59 [1.57, 1.61]
	Outer Regional Australia	1	1
IRSAD	Quintile 1 (most disadvantaged)	1	1
	Quintile 2	0.92 [0.26, 3.25]	0.94 [0.42, 2.09]
	Quintile 3	1.15 [0.67, 1.96]	1.43 [1.02, 2.01]
	Quintile 4	0.95 [0.44, 2.02]	1.2 [0.65, 2.22]
	Quintile 5 (least disadvantaged)	1.21 [1.03, 1.42]	1.66 [1.51, 1.84]
Severity	Severe	1	1
	Moderate	0.45 [0.38, 0.53]	0.45 [0.4, 0.51]
	Mild	0.13 [0.13, 0.14]	0.15 [0.14, 0.16]
Constant			0.97 [0.51, 1.85]

Note. CALD: culturally and linguistically diversified; IRSED: Index of Relative Socioeconomic Advantage and Disadvantage; COR: crude odds ratio; AOR: adjusted odds ratio.

Significant sex differences were observed. Females had substantially higher odds of mental health service use than males in 12-month use (AOR = 1.55, 95% CI = [1.49, 1.61]).

Socioeconomic area of residence was strongly associated with mental health service use, with higher odds observed among individuals living in more advantaged areas. Individuals in the most advantaged quintile (Quintile 5) had significantly higher odds of 12-month service use (AOR = 1.66, 95% CI = [1.51, 1.84]), compared with those in the most disadvantaged quintile (Table 3).

Mental health service use varied across remoteness categories. Service use in the past year was markedly higher among major-city residents compared with those in remote and outer regional areas (AOR = 1.31, 95% CI = [1.30, 1.32]). Furthermore, relative to residents of remote and outer regional areas, individuals in inner regional areas had higher odds of 12-month service use (AOR = 1.59, 95% CI = [1.57, 1.61]).

Higher educational attainment was associated with greater odds of both lifetime and past-year mental health service use. Compared with individuals with secondary education or lower, those with a certificate or diploma had higher odds of 12-month service use (AOR = 1.58, 95% CI = [1.52, 1.65]). University graduates had the highest likelihood of engagement (AOR = 1.67, 95% CI = [1.63, 1.70]).

Interaction of area-level socioeconomic position and remoteness

Patterns of mental health service use in the past 12 months showed similar variation. Individuals in inner regional areas reported the highest overall use, particularly those in middle socioeconomic areas (56.1%), followed by major cities (Quintile 4–5: 46.7%). In contrast, service use was lowest among those in outer regional and remote areas, especially in the high socioeconomic areas (35.8%). Notably, service use did not increase linearly with socioeconomic advantage across all regions, suggesting complex patterns of access (Figure 2).

Figure 2.

Mental health service use rate by socioeconomic status and remoteness of their residence (weighted).

In the adjusted model, the strongest association for 12-month service use was observed among individuals living in middle socioeconomic areas in inner regional areas (AOR = 2.82, 95% CI = [2.27, 3.51]). These findings indicate that geographic location and socioeconomic advantage jointly influence access, with compounded inequities for those living remotely and in disadvantaged circumstances (Table 4).

Table 4.

Controlled logistic regression model to test the interaction effect of IRSAD and remoteness for 12-month mental health service use.

Variable	Category	12-month AOR [95% CI]
Remoteness × IRSAD	Outer regional and remote × Quintile 1 and 2	1
	Outer regional and remote × Quintile 3	1.26 [1.21, 1.32]
	Outer regional and remote × Quintile 4 and 5	1.4 [1.18, 1.66]
	Inner regional Australia × Quintile 1 and 2	1.6 [1.5, 1.72]
	Inner regional Australia × Quintile 3	2.82 [2.27, 3.51]
	Inner regional Australia × Quintile 4 and 5	1.42 [1.23, 1.63]
	Major cities of Australia × Quintile 1 and 2	1.21 [1.16, 1.27]
	Major cities of Australia × Quintile 3	1.77 [1.54, 2.05]
	Major cities of Australia × Quintile 4 and 5	1.95 [1.82, 2.09]

Discussion

Our findings show that substantial gaps remain between the prevalence of mental health disorders and the use of services in Australia. Nationally, two in five Australians experience a mental health disorder in their lifetime, yet only around two-thirds have ever accessed care. In the 12 months preceding the survey, one in five experienced a mental health disorder, but fewer than half received care. Use of digital or telephone-based support was also limited. Service use was unequally distributed, favouring younger adults, women, those with higher education, and residents of advantaged or urban areas. Socioeconomic disadvantage and remoteness compounded these inequities, with the lowest utilisation among people living in disadvantaged remote areas.

Australia’s obligations under the United Nations Convention on the Rights of Persons with Disabilities (CRPD) (Office of the United Nations High Commissioner for Human Rights, 2007) and the National Mental Health and Suicide Prevention Agreement (Commonwealth of Australia; States and Territories, 2022) establish a shared commitment to ensuring equitable access to mental health care for all people, irrespective of disability, socio‑economic position, or place of residence (Commonwealth of Australia; States and Territories, 2022; Office of the United Nations High Commissioner for Human Rights, 2007). These frameworks position equity as a core system responsibility, making the distribution of access an important indicator of system performance.

However, our findings indicate that although mental health disorders are common, service use remains uneven, with clear socioeconomic and geographic disparities. Higher prevalence among younger age groups aligns with global trends of increasing mental health challenges amid social, economic, and pandemic-related stressors (Rickwood et al., 2015). Lower utilisation among older adults may reflect generational stigma, under-recognition of symptoms, or competing health priorities (Elshaikh et al., 2023). The higher likelihood of service use among women compared with men is also consistent with global evidence and may reflect gender differences in help-seeking behaviours, social norms, and patterns of symptom recognition (Judd et al., 2008). Effect sizes were generally small according to Rosenthal’s guidelines (Rosenthal, 1994), consistent with prior evidence that structural determinants exert modest but meaningful influences on service use. Notable exceptions were the sex difference in lifetime use and reduced use among adults aged ⩾65 years, both of which approached medium magnitude. The combined effect of socioeconomic advantage and remoteness produced the largest association in the analysis, falling in the small-to-medium range.

Utilisation was lowest in the Northern Territory and Tasmania, despite both reporting higher prevalence of mental health disorders. These patterns reflect structural, geographic, and demographic constraints: both jurisdictions rank lowest in proportion of people residing in urban centres (Capuano, 2025). Populations outside major urban centres mean that people often live far from available services, resulting in longer travel times and fewer local options for care. Providers are spread across large distances, which makes it hard to attract and keep a stable health workforce (Acil Allen, 2022; ABS, 2025). On the contrary, smaller jurisdictions face diseconomies of scale even though per‑capita health spending is relatively high, as can be observed in the Northern Territory (Australian Institute of Health and Welfare, 2024). The fixed costs of specialist times, inpatient units, and capital infrastructure are substantial (Hernandez-Villafuerte et al., 2017), yet small populations limit the feasibility of sustaining a comprehensive service mix. As a result, local availability is narrower and reliance on NGO or digital health provision is greater (Wakerman and Humphreys, 2013). This reliance, however, introduces its own set of limitations: NGO services often operate on short-term funding cycles that constrain continuity, while digital models, although valuable for reach, cannot fully substitute for in-person assessment, crisis response, or complex care (Lattie et al., 2019; Rosenberg and Hickie, 2013). In practice, this means that people with higher needs face longer waits, more fragmented pathways, and greater dependence on fly-in-fly-out clinicians or visiting specialists (Elshaikh et al., 2023). The cumulative effect is a system that can meet low-intensity demand but struggles to provide sustained, comprehensive support, reinforcing the very access gaps that drive lower uptake in these jurisdictions (Bradford et al., 2016).

The relatively low uptake of digital and telephone‑based services in our study also contrasts with expectations that telehealth would mitigate inequities (Wakeman et al., 2025). Evidence from the COVID‑19 period suggests that telehealth expansion did not fully overcome socio‑economic barriers, with significant disparities in psychiatric consultations persisting despite increased availability (Meehan et al., 2025). While digital service provision expands access, it does not, in itself, resolve persistent structural and socioeconomic barriers that continue to shape unequal uptake. In our study, strong and disproportionate socio‑economic differences were evident across area‑level advantage, education, and remoteness, with compounded disadvantage among people living in remote and socioeconomically deprived settings. These patterns indicate that equal access to mental health care, as mandated by national and international obligations, remains far from being realised (Commonwealth of Australia; States and Territories, 2022; Office of the United Nations High Commissioner for Human Rights, 2007). These disparities are consistent with wider evidence indicating that those with greater socioeconomic resources disproportionately benefit from the mental health system (Enticott et al., 2024; Hashmi et al., 2023; Meehan et al., 2025). Analyses of national survey data have shown that Australians living in the most disadvantaged areas experience the highest levels of mental distress yet have the least access to mental health services (Enticott et al., 2024). Our findings underscore this pattern: individuals in the most disadvantaged areas had lower odds of both lifetime and recent service use, with similar disadvantages observed among those living in remote regions and those with lower levels of educational status.

Such disparities are not unique to Australia (Epping et al., 2017; Evans-Lacko et al., 2017). Evidence from the WHO World Mental Health (WMH) surveys shows that in high-income countries, higher educational attainment is consistently associated with greater mental health service use, a finding that aligns with our results. In contrast, the WMH study reported a complex and non-linear association with income, with treatment rates not following a simple socioeconomic gradient. Our economic disadvantage variable serves as a composite measure, capturing cumulative barriers beyond individual income (Evans-Lacko et al., 2017). Our study extends this evidence by examining the interaction between area‑level socio‑economic disadvantage and remoteness, demonstrating how these factors compound barriers to care. Individuals living in highly disadvantaged outer‑regional and remote areas had the lowest lifetime mental health service use, whereas use was highest among those in advantaged major cities. This pattern echoes concerns raised in recent commentary that Australia’s universal health system is “for richer, but not for poorer,” with structural barriers preventing equitable access to mental health support (Enticott et al., 2024).

Taken together, these results indicate that important gaps remain in how equity is addressed within Australia’s mental health system. The persistence of socioeconomic and geographic disparities, despite policy commitments (Commonwealth of Australia; States and Territories, 2022; Office of the United Nations High Commissioner for Human Rights, 2007), digital mental health service expansion (Department of Health, 2021), and increased national attention, suggests that current reforms have not yet fully addressed the structural determinants of access. The clear socioeconomic and geographic gradients in service use show that access continues to be shaped by where people live and the resources available to them. Meeting international and national obligations under the United Nations CRPD or the National Mental Health and Suicide Prevention Agreement will require a more deliberate focus on structural fairness.

The limited reach of digital and telephone‑based care further indicates that the mere availability of technology is insufficient to close access gaps. Substantial effort is needed to promote these services and ensure that disadvantaged groups are able to use them and that they adequately ameliorate the lack of other services available. Addressing digital exclusion (Perimal-lewis et al., 2025) among populations with low access would help strengthen telehealth uptake, particularly in areas and situations where face‑to‑face care is limited and inaccessible. Overall, the results point to the need for a shift from broad availability to genuinely equitable access, one that recognises structural disadvantage as a central barrier to mental health care in Australia.

Study strengths and limitations

A major strength of this study is the use of a large, nationally representative dataset, which allows for robust examination of socioeconomic and geographic patterns in mental health service use across Australia. The analysis incorporates multiple dimensions of disadvantage, education, area-level socioeconomic status, and remoteness, and uses adjusted models to explore both independent and interacting effects. This provides a more nuanced understanding of how structural factors shape access to care.

However, several limitations should be noted. Self-reported measures of mental health conditions and service use may be subject to recall or reporting bias, and area-level indicators may not fully capture individual socioeconomic circumstances. The survey also lacks detailed information on service quality, continuity, or unmet need. Lifetime service use estimates – particularly among older respondents – may partly reflect care received under earlier mental health system configurations. Although all models adjust for age group, cohort-level differences in historical service availability cannot be fully eliminated. While models adjust for disorder severity – which correlates with disorder type – residual confounding arising from the differential distribution of disorder classes across population groups may remain. This analysis cannot account for local workforce availability or service configuration, both of which may independently influence access.

The NSMHWB 2020–2022 was conducted across two cohorts: one during COVID-19 lockdowns (December 2020–July 2021) and one during the subsequent recovery period (December 2021–October 2022). Pandemic conditions may have influenced prevalence estimates – through elevated psychological distress – and service use patterns, through disrupted face-to-face care and expanded telehealth and Better Access provisions. The survey reported slightly higher disorder prevalence in Cohort 1 than Cohort 2 (20.9% vs 19.7%), indicating a modest pandemic effect. These factors may limit the generalisability of findings to non-pandemic conditions, and separating pandemic-related influences from underlying structural patterns in service access would require longitudinal approaches beyond the scope of this study.

Conclusion

This study demonstrates that socioeconomic and geographic factors continue to shape mental health service use in Australia, with clear gradients across area-level advantage, education, and remoteness. Despite national commitments to equitable access, substantial gaps remain, particularly for people living in remote, socioeconomically disadvantaged areas. Addressing these disparities will require sustained, equity-focused policy action, including targeted investment, strengthened public services, and approaches that recognise structural disadvantage as a central barrier to care. Ensuring that access aligns more closely with need is essential if Australia is to meet its obligations under national agreements and international human rights frameworks.

Data availability statement

The detailed microdata used in this study can be obtained by seeking approval from the Australian Bureau of Statistics.