Mental health services accessed by new patients: linked data analysis

Background

This report describes the types of mental health contacts accessed by new patients in 2018 and highlights differences in the broad types and combinations of services by sex, age group, remoteness area and socioeconomic disadvantage. The analysis uses linked national health datasets and state and territory public hospital data from the National Health Data Hub (NHDH) data asset (AIHW 2025b).

The analysis population was people identified as ‘new patients’, that is, who accessed their first, or first in at least 5 years, in-scope mental health-related service during 2018.

The in-scope services with available data were:

• Medicare Benefits Schedule (MBS) claims for mental health-related services.
• Dispensed Pharmaceutical Benefits Scheme (PBS) prescriptions for mental health-related medications.
• Public hospital emergency department (ED) presentations where there was a principal diagnosis of a mental health disorder.
• Public and private hospital admitted patient care (APC) episodes with specialised psychiatric care or a principal diagnosis of a mental health disorder.
• Public hospital outpatient (non-admitted) care (OPC) for select clinic service types.

This analysis included all in-scope NHDH mental health-related services accessed during the 12 months from, and including, the initial contact. The calendar year 2018 was selected as the most recent year of data available in the NHDH dataset when the analysis was initiated to ensure that the 12-month follow-up period for all new patients was completed before the end of 2019 and before the onset and impacts of the emergency phase of the COVID-19 pandemic.

Important caveats for this report include:

• People may have accessed other mental health-related services that are not measured in the available data. These include (but are not limited to): private medical consultations, private hospital emergency department presentations, community and residential mental health care, online/phone support services, or other MBS services such as GP attendances that were not billed as a mental health item.
• The NHDH does not include ED and APC data from Western Australia or the Northern Territory, nor OPC data from Western Australia, the Northern Territory, or South Australia.

The Methodology and technical notes section contains detailed information on the in-scope services, data caveats, limitations, and methodology of this analysis.

Key findings

Services

Around 962,500 people had their first, or first in at least five years, in-scope mental health service contact in 2018. This new patient cohort represented around 4% of the Australian population and accessed nearly 4.9 million services during the 12 months following, and including, their first service contact. A third (38%) of the cohort accessed a single service during this period, while 20% had 9 or more contacts, including 6% who had at least 16 contacts.

First mental health contacts comprised 53% MBS services, 42% dispensed PBS prescriptions, 2% ED presentations, 2% APC and 1% OPC services. The high proportion of PBS prescriptions as first contact indicates that many patients received their initial prescription in a consultation that was not captured as a mental health service in the data. These could include via a private specialist appointment or a GP consultation that was not billed as a mental health-related item.

Patient characteristics

Notable demographic differences in the type of first mental health contact include:

• New patients who were female or in younger age groups were more likely to access MBS services as the first contact and ongoing contacts than males or older age groups.
• Male patients, those in older age groups, or patients in remote areas were more likely to access PBS prescriptions than MBS services as their first or ongoing contacts.
• Patients in the 18–24 age group, and specifically males, were more likely to have an ED presentation as a first contact and during the 12-month follow up, compared with other sex and age groups.
• People residing in areas of greater socioeconomic disadvantage had higher hospital contact rates (ED, APC or OPC) and lower MBS rates during follow-up, compared with less disadvantaged areas.

By age group, overall patient and 12-month mental health contact rates peaked among the 18–24 age group (largely driven by MBS services) with a secondary peak among the 75 and over group (largely driven by PBS prescriptions). For the older group high rates of PBS prescriptions could possibly be related to mental health symptoms associated with neurodegenerative conditions such as dementia.

Service combinations and distributions during follow-up

The 38% of patients who had a single contact included 17% of patients who received a single MBS mental health service and 19% a single PBS mental health prescription. Around a quarter of patients accessed multiple MBS services without any PBS or hospital (ED, APC and/or OPC) contacts and a further 15% accessed multiple PBS without any MBS or hospital contacts. Eighteen percent of patients accessed a combination of MBS and PBS. Three percent of patients accessed hospital care only (single or multiple contacts) without any MBS or PBS, and a further 3% accessed hospital services in combination with MBS and/or PBS.

The number of monthly contacts dropped during the first six months and levelled off for the second half of the follow-up period. In the twelfth month following first contact, 16% of the cohort accessed 226,000 contacts, 61% of which were for dispensed PBS prescriptions. The sustained number of contacts indicates a portion of new patients who accessed ongoing mental health services, particularly medication.

Age cohorts

Differences in types of service accessed was observed between the age groups of Youth (0–17 years), Working age (18–64 years) and Older patients (65 years and over). Latent class analysis of contacts during the 12month follow-up indicates that the underlying patient groups with the largest population shares among each age cohort were:

• 70% of Youth patients were in a class characterised by primarily MBS contact
• 53% of Working age patients were in a class characterised by MBS and PBS contacts
• 78% of Older patients were in a class characterised by PBS contacts

Examination of service flows between different contact types by broad age group highlights some further differences in Hospital service pathways. Of patients who presented to ED (4% of Youth, 3% of Working age, 4% of Older patients), 43% of the Older patients had APC as their next contact, compared with 26% of Working age and 16% of Youth ED presentations.

New patient overview

Service summary

There were 962,500 people identified as new patients who had their first (or first in at least 5 years) in-scope mental health-related service in Australia during 2018. These patients, represented around 4% of the Australian estimated resident population and 18% of the 5.5 million people (new and continuing patients) who accessed in-scope mental health services in 2018. New patients accessed nearly 4.9 million services during the 12-month follow-up period from their first contact (Table 1). Most contacts were MBS services (53% of services, accessed by 62% of patients) followed by PBS prescriptions (44% of services, accessed by 53% of patients).

NOTE: The sum of patient percentages does not equal 100% as patients may access multiple types of services. The sum of service percentages may not equal the total due to rounding. The hospital subtotal comprises ED, APC, and OPC unique patients and services. Patient and contact rates for hospital services were calculated using the sum total populations of the states and territories that contributed data, not the national Australian total population.

Source: National Health Data Hub AIHW

Thirty eight percent of the cohort accessed a single in-scope service. The median number of contacts during the 12-month follow-up was 2. There were 191,000 patients (20%) who had 9 or more contacts, including 56,400 patients (6%) who had more than 16 contacts. The number and type of contacts and combinations are examined in more detail in the section Service type combinations during follow-up.

Demographic characteristics

The new patient cohort represented around 4% of the Australian resident population and was 53% female and 47% male.

Nearly half (48%) of the cohort was under 35 years of age. The 18–24 age group had the highest population rate of just under 5% (46 patients per 1,000 population), compared with 3%–4% for the other age groups. This was largely due to a higher rate for females in the 18–24 and 25–34 age groups (5%) compared to males (4%) in those age groups. The highest rate for males was for the 75 years and over age group (5% of the population), compared with 3%–4% for males in younger age groups, and 4% for females 75 years and over.

The median number of contacts was higher for patients in the 0–17 age group (4 services), and 18–24 or 25–34 groups (3), compared with the older age groups (2). The median number of contacts was the same for males and females overall, and between males and females of corresponding age groups.

By address of usual residence at the first mental health contact, patients in Remote and Very remote areas had the lowest patient rate, representing around 2% of the population, compared with around 4% for Major cities, Inner regional and Outer regional areas.

Patient rates were similar for all Socio-Economic Index for Areas (SEIFA) quintiles at around 4% of the population. Patients who resided in Quintile 5 (least disadvantaged) areas had a median of 3 contacts during the 12-month follow-up compared to 2 for patients who resided in areas of greater disadvantage.

Just under 20,300 (2%) of the cohort died during the 12-month follow-up. The proportion of patients who died was slightly higher among males (3%) compared with females (2%). The proportion of patients who died was also higher among the older age groups, particularly the 75 and over group at 18%, compared with the younger age groups (less than 1% of the age groups under 55 years, 2% of the 55–64 age group, and 5% of the 65–74 age group).

First mental health service contacts

The 962,500 patients had over 1 million ‘first’ mental health services due to nearly 71,500 (7%) of the new patient cohort receiving multiple services on the day of their first contact. MBS services made up over half (53%) of all first contacts, comprising 49% General Practitioner (GP) services, and the remaining 4% from Other medical practitioners, Psychologists, Psychiatrists, and Other allied health.

PBS prescriptions made up 42% of first contacts (20% Antidepressants, 10% Anxiolytics, 9% Hypnotics and sedatives, and 1% each Antipsychotics and Psychostimulants, agents used for ADHD and nootropics). The high number of PBS dispensed prescriptions as first contacts suggests that medications are often prescribed during consultations and treatment that are not billed as MBS mental health items (for example, using general GP items). ED and APC each accounted for 2% of first contacts, and OPC accounted for 1%.

Female patients had a higher proportion of first contacts that were MBS (56%, compared with 51% for males), and a lower proportion that were PBS (40% compared with 43% for males).

The proportion of first services that were MBS was highest for the 0–17 age group at 80%, and decreased with each older age group, down to 14% for patients 75 and over (Figure 1). The opposite was observed for PBS which increased from 13% of first services for the 0–17 age group, up to 77% for those 75 and over. Antidepressants, as the most frequently dispensed medication type, was a main contributor to this trend, making up larger proportions of first services for each older age group, from 5% for the 0–17 group, to 35% for patients 75 and over. The proportions of Anxiolytics and Hypnotics and sedatives also increased with each older age group, to a peak among the 65–74 age group (16% for Anxiolytics and 21% for Hypnotics and sedatives).

Figure 1: Proportions of first contacts by service type among each age group, remoteness area and SEIFA quintile of new patients in 2018

The 18–24 age group had the highest proportion with ED as a first contact (4,600 presentations, 4% of first contacts). By sex and age group, males 18–24 had the highest number (2,800) and proportion of first contacts (5%) that were ED, compared with 1%–3% for other groups. Patients 75 years and over had the highest proportion of first contacts that were APC (4,200 contacts, 5% of first contacts).

There were lower proportions of MBS and higher proportions of PBS as first contacts with each increasing category of remoteness (Figure 1). For patients in Major cities, first contacts comprised 56% MBS and 39% PBS, compared with 38% MBS and 56% PBS for those in Remote and Very remote areas. By SEIFA quintile, patients residing in Quintile 1 (greatest disadvantage) areas had a lower proportion of first contacts that were MBS items (49% compared with 51%–57% for patients in areas of lower disadvantage), and slightly higher proportions that were ED (3% compared to 1%–2%).

Patients who died during the 12-month follow-up period had higher proportions of first contacts that were PBS, ED, APC or OPC, and lower proportions that were MBS, compared with patients who did not die. This trend was evident within each age group, although noting that the number of deaths among the age groups under 65 years were too low to make reliable inferences about associations with first contact type.

Contacts during the 12-month follow-up

During the 12 months following the first in-scope mental health contact in 2018, for people overall (both sexes), patient and contact rates were highest for the 18–24 years age group at 46 patients and 257 contacts per 1,000 population (Figure 2). The rates were lower for each older age group, until increasing again among people aged 75 and over. By sex, the contact rates for females peaked among the 18–24 age group at 294 services per 1,000 population, while for males the peak contact rate was equal highest among the 18–24 and 75 and over age groups (221).

Figure 2 also shows that the peak in mental health contact rates for the 18–24 age group was mainly driven by MBS, while the high rates among the 75 and over age group was driven by PBS. People in the 18–24 age group also had slightly higher mental health related contact rates of 4 ED presentations per 1,000 population for males and 3 for females, compared with 1–3 for males in other age groups and 1–2 for females in other age groups.

Both sexes showed similar directions of trends over the age groups, but females 18–24 had considerably higher MBS and PBS contact rates compared with males. This was driven by higher contact rates for GP, Clinical Psychologist and Other Psychologist services and Antidepressants prescriptions in this age group compared with males. Males in the 75 and over age group had higher PBS, Psychiatrist, and hospital patient rates compared with females.

The higher contact and patient rates among the older population is likely related to the higher prevalence of dementia-related conditions and associated treatment for mental-health symptomology (AIHW, 2024a). The majority of people living with dementia in Australia are women, although men living with dementia tend to have higher rates of hospitalisation and are more likely to be prescribed Antipsychotics, compared with women (AIHW, 2024a).

Figure 2: 12-month patient and service population rates for new mental health services patients in 2018, by sex and age group

For people in Remote and very remote areas, the patient rates for PBS prescriptions were higher compared with MBS services (15 vs. 11 per 1,000 population, respectively), but were both considerably lower than the rates for patients in Major cities where the MBS patient rate (25) was higher than PBS (19) (Figure 3). Figure 3 also shows this trend where, as remoteness increased the MBS patient and contact rates decreased, and patients were more likely to access PBS than MBS. The MBS service rate was more than 3 times higher for people residing in Major cities at 111 per 1,000 population, compared with 33 in Remote and very remote areas.

The lower MBS rates for Remote and very remote areas are consistent with the patterns in general MBS services by remoteness area (AIHW 2023). This is also likely related to lower mental-health service availability in more remote areas. The 2018 full time equivalent rates per 100,000 people for psychologists and psychiatrists was lower with each category of increasing remoteness (Psychologists ranged from 25 for Very remote to 106 for Major cities, and Psychiatrists ranged from 3 to 16, respectively) (AIHW 2020).

Figure 3: 12-month patient and service population rates for new mental health service patients in 2018, by remoteness area and SEIFA quintile

By SEIFA Quintile, the patient and contact rates for MBS were lowest for people residing in SEIFA Quintile 1 (most disadvantaged) areas, with a gradient of higher service rates with each quintile of decreasing disadvantage (Figure 3). People residing in Quintile 1 (most disadvantaged) areas received 83 MBS services per 1,000 population, compared to 121 for people in Quintile 5 (least disadvantaged) areas. These results are consistent with patterns observed for MBS services in general (AIHW 2023) and may reflect barriers to accessing services for patients in more disadvantaged and remote areas and/or lack of service provision in these areas.

Hospital (ED, APC and/or OPC) contact rates also followed a gradient, with rates for people residing in the more disadvantaged quintiles, from 10 services per 1,000 population for people in the most disadvantaged areas, to 5 in the least disadvantaged areas. It should be noted that the ED and OPC hospitals data is for public hospital services only. Private hospital services are more likely to be accessed by patients with greater socioeconomic advantage and residing in urban areas where private psychiatric hospitals are predominantly located (AIHW 2024b).

Service type combinations during follow-up

Overall, during the 12-month follow-up, 38% of patients had a single in-scope contact, including 19% had a single PBS prescription and 17% had a single MBS service (Figure 4). Of all the new patient cohort, 27% had 2–4, 16% had 5–8, 14% had 9–15, and 6% had 16 or more mental health contact over the 12 months.

Forty two percent of patients accessed MBS only. This means that in addition to the 17% with a single service, 26% of new patients accessed multiple MBS services without any other service type (12% with 2–4 contacts, 8% with 5–8, 5% with 9–15 and 1% with 16 or more). The 33% of patients who accessed PBS only, included the 19% dispensed one PBS prescription, 8% with 2–4, 3% each with either 5–8 or 9–15 and less than 1% with 16 or more prescriptions. The PBS only group highlights that mental health services are often provided in consultations not billed using MBS mental health-related items, for example, other General Practitioner items.

In addition, 18% of patients had a combination of various numbers of MBS and PBS contacts. Three percent of patients accessed hospital services only (either single or multiple), and a further 3% accessed hospital services in combination with MBS and/or PBS.

Figure 4: Number and percentage of patients by service intensity and service combinations, during the 12-month follow-up

Age cohort analysis

The cohort overview shows different modes of service use according to age group and life stage. The following analysis has been separated into 3 broad age groups: Youth (0–17 years, 190,500 patients), Working age (18–64 years, 622,500 patients), and Older patients (65 years and over, 150,000 patients).

Service characteristics of age cohorts

To examine if the broad age cohorts were characterised by different types or combinations of services, a modelling technique, known as latent class analysis was used to identify unobserved (for example, ‘latent’) classes. Details of this method are outlined in Methodology and technical notes.

The model for Youth included 4 classes, while the models for Working age and Older patients each had 3 classes. The estimate class population shares and conditional probabilities of class membership for each service type are reported in Table 2, with darker shades highlighting the service types that have higher probabilities. For example, membership in the Youth MBS+PBS class (incorporating 22% of the Youth population) indicates that people in that class had a 60% probability of having had an MBS service and 100% probability of a PBS prescription.

Table 2: 2018 new patient age cohort classes, population share, and probabilities of indicator (service type) membership

Consistent with the results presented in the section Service type combinations during follow-up, each age cohort includes classes with large population shares characterised (probabilities ≥ 0.4) by either mainly MBS, PBS, or MBS+PBS. For example, 53% of the Working age group were estimated to fall into a class characterised by MBS+PBS and 41% fell into a class characterised by MBS only. Furthermore, the size of the population shares for the MBS and PBS dominated groups reflect the results of earlier sections of this report First mental health service contacts and Services accessed during the 12-month follow-up:

• Youth patients had the largest proportion of the cohort (70%) allocated to the class characterised by MBS
• The Working age patients had 53% allocated to a class characterised by MBS+PBS
• The largest class (78%) for Older patients was characterised by PBS

Within each age cohort, the classes with smaller population shares had different features when compared between the cohorts:

• Youth had two small classes of 4% of patients each, one characterised by APC+OPC, and the other by ED+MBS services
• 6% of the Working age cohort were in a class that was characterised by a combination of MBS+PBS+ED
• 8% of the Older patient cohort were in a class characterised by PBS+ED+APC

Service flows

The flows between the different mental health service types during the 12-month follow-up were examined by broad age cohort.

Conclusion

This analysis presents associations between MBS, PBS and hospital mental health service use patterns and demographic factors. Of note are the differences in service use by sex and age group, including the higher likelihood of younger people, and particularly females, to access mental health related MBS services compared to the propensity of older people to have PBS contacts.

The observed differences likely relate to a range of factors, including treatment guidelines, social factors and perceptions around mental ill health and appropriate types of mental health care, or differences in physical health and conditions over the life cycle. For example, a child with a mood disorder is less likely to receive an antidepressant prescription as the first line or sole treatment, compared with an elderly person with a mood disorder comorbid with a neurodegenerative condition. Differences in service use by remoteness area or area of socioeconomic disadvantage is also likely to relate to variations in service availability as well as geographic and economic barriers affecting access.

Statistical (latent class) analysis supported the results of the descriptive analysis around differing MBS and PBS service use by life stage, with Youth patients (0–17 years) characterised by MBS contacts, while Working age (18–64 years) were characterised by a combination of MBS and PBS contacts, and Older patients (65 years and over) were characterised by PBS contacts. Service flow analysis highlighted differences in hospital contacts, with larger proportions of Older patients who progressed from an ED presentation to APC as the next contact, compared with the Working age and Youth cohorts.

Future directions

Continuing from this high-level overview, further linked-data analyses could investigate questions concerning different age cohorts, or patients with specific mental health conditions, and interactions between more detailed service information such as MBS provider type, PBS medication type, or hospital setting.

The high proportion (42%) of PBS mental health related medication dispensing as a first contact highlights that many contacts that involve a mental health component and in which the medication was prescribed, are not captured in the data, such as general MBS GP consultations. Examination of non-mental health MBS items in proximity to PBS prescriptions may be helpful to understand the extent to which this occurs.

Where can I find more information?

For more information refer to:

• National Health Data Hub
• Mental health online report
• Glossary terms

Methodology and technical notes