National mental health report 2013

Appendix 2: Data sources and explanatory notes for part 3

Page last updated: 2013

Introduction

The following notes have been prepared to assist in the interpretation of the data measuring each of the Fourth National Mental Health Plan indicators presented in Part 3 Monitoring progress and outcomes under the Fourth National Mental Health Plan.

Table A2-1 provides summary information about the data sources used, and which indicators are based on each source. Table A2-2 provides further explanatory detail regarding the derivation of each indicator. The table does not include information about indicators that cannot yet be reported.

Data sources and explanatory notes

Table A2-1: Overview of data sources, in alphabetical order

Data sourceDescriptionRelevant indicators, figures and tables
Australian Bureau of Statistics Causes of Death, Australia, 2011, reportThe official suicide rate in any given year is produced by the Australian Bureau of Statistics, using data from coroners' courts in all states and territories. Data covering the period 2003 to 2011 are published in the Causes of Death, Australia, 2011, report. 53 Unpublished data are also used.

Information about deaths occurring in each state and territory is provided to the Australian Bureau of Statistics (ABS) by individual state and territory Registrars of Births, Deaths and Marriages f or coding and compilation into aggregate statistics. In addition, the ABS supplements this data with information from the National Coronial lnformation System (NCIS).

Indicator 9 (Table 10, Figures 55-56)
Australian Government analyses of jurisdiction dataSee Appendix 1, Table Al-l.Indicators 13-16 (Figures 59-62)
Medicare Benefits Schedule dataSee Appendix 1, Table Al-l.Indicator 7 (Table 9)

lndicator 13 (Figure 59)

National Drug Strategy Household Surveys conducted in 2010, 2007, 2004, 2001 and 1998The National Drug and Alcohol Household Surveys are conducted by the Australian Institute of Health and Welfare every three years. 77 The surveys are designed to provide data on the level, patterns and trends of alcohol and other drug use in Australia, including licit and illicit drug use.

The most recent survey - the tenth in the series - was conducted in 2010 and involved over 26,000 participants who were recruited via a household sampling strategy (a response rate of just over 50%).

Indicator 8 (Figures 52-54)
National Health Surveys conducted in 2011-12 and 2007-08The 2011-12 National Health Survey (NHS) 34 was conducted from March 2011 to March 2012 by the Australian Bureau of Statistics. Previous surveys in this series were conducted in 1989-90, 1995, 2001, 2004-05 and 2007-08. The 2007-08 NHS 36 was conducted between August 2007 to June 2008. The surveys were designed to obtain national benchmarks on a wide range of health issues, and to enable changes in health to be monitored over time.

The 2011-12 and 2007-08 NHSs each sampled more than 20,000 people across all age groups from private dwellings in all states and territories. Information was collected via personal interview. The surveys collected information about a broad range of health issues, include mental health status, as well as demographic and socio-economic information.

Indicator la (Figures 44-45)

Indicator 2a (Figures 46-47)

National Minimum Data Set (NMOS) - Mental Health Establishments (MHE) collection 2005-06 to 2010-11See Appendix 1, Table Al-l.Indicators 21-22 (Figures 65-66)
National Outcomes and Casemix CollectionData on a range of outcomes for consumers of state and territory public sector mental health services are collected via the National Outcomes and Casemix Collection (NOCC). 78 The NOCC was endorsed by all States and Territories in 2003, and all jurisdictions have reported data since 2004-05. Analysis of this data is conducted by the Australian Mental Health Outcomes and Classification Network (AMHOCN), using data submitted annually by states and territories to the Australian Government Department of Health and Ageing.

The NOCC protocol prescribes a set of measures to be collected at particular times in the clinical process. The measures are specific to three broad mental health service settings (Inpatient, Residential and Ambulatory) and also to three target populations (i.e., Children and Adolescents, Adults and Older Persons).

It is difficult to ascertain definitively the 'coverage' of NOCC reporting, however AMHOCN has previously estimated Inpatient episode coverage
at approximately 33% for Completed Episodes of at least 3 days duration and estimated ambulatory episode coverage at approximately 20% for
'Completed Episodes' and 33% for Ongoing Episodes.

Indicator 4 (Figures 49-50)

Indicator 23 (Figure 67)

National Prisoner Health Census conducted in 2010The National Prisoner Health Census 67,68 was conducted in 2010 by the Australian Institute of Health and Welfare. The Census was conducted in October and November 2010 in 44 of the 45 public and private adult correctional facilities from all jurisdictions except New South Wales and Victoria. The survey was developed to help monitor the health of prisoners, and to inform and evaluate the planning, delivery and quality of prisoner health services.

Data were collected over a two week period. Individuals were asked a number of questions, including several about their mental health. Data were collected for 610 new prison entrants.

Indicator 20a (Figure 64)
National Survey of Mental Health and Wellbeing, surveys of adult population, conducted in 2007 and 1997The 2007 National Survey of Mental Health and Wellbeing (NSMHWB) 65, survey of adult population, was conducted between August and December 2007 by the Australian Bureau of Statistics (ABS). The 2007 survey, and its precursor in 1997 4, were designed to provide reliable information about the prevalence of common mental disorders among Australian adults, and the impairment, severity, health care service use and unmet treatment needs associated with these disorders.

In both surveys, participants were recruited by a household sampling strategy and interviewed in their homes. The 1997 survey involved 10,641 participants aged 16-85 years and the 2007 survey involved 8,841 participants aged 18-99 years. The response rate was 60% for the 2007 survey and 78% for the 1997 survey.

Indicator 12 (Figure 58)

Indicator 13 (Table 11)

National Survey of Mental Health and Wellbeing, survey of children and adolescents, conducted in 1998The Child and Adolescent Component of the National Survey of Mental Health and Wellbeing 6 was conducted in 1998. This survey recruited 4,509 children and adolescents aged 4-17 through a household sampling strategy. It elicited information from participants and their parents via interview.Indicator 12 (Figure 58)
National Surveys of Mental Health Literacy and Stigma conducted in 2011, 2003-04 and 1995The National Surveys of Mental Health Literacy and Stigma 46 are a series of general community surveys designed to assess aspects of the mental health literacy in the Australian population and to monitor trends over time. The surveys were conducted using computer-assisted telephone interviews.

The surveys involved the presentation of vignettes describing males or females with symptoms of a mental illness, with subsequent questions eliciting information about respondents' ability to recognise specific mental disorders, their beliefs about treatment, and stigmatising attitudes. The 1995, 2003-04 and 2011 samples consisted of 2,164, 3,998 and 6,019 adults aged 18 years or older respectively. Response rates were 85% in 1995, 34% in 2003-04 and 44% in 2011.

Indicator 3 (Figure 48)

Indicator 11 (Figure 57)

Principals Australia's National Market Research Survey conducted in 2011Australian Government funding was provided to expand the Principals Australia's National Market Research Survey 50 to collect specific information regarding the mental health literacy component of schools' curricula. The Market Research Survey was undertaken in April and May 2011 and included a range of mental health specific questions designed to gather information on the range of mental health related activities undertaken in Australian public and private schools. The survey captured data from a large sample of principals based in all states and territories of Australia, and from all school types, sectors and all locations and is believed to be representative of all schools. Analysis of data for the mental health specific questions was restricted to responses by school principals, numbering 1,285 and covering an estimated 14% of all Australian schools.Indicator 6 (Figure 51)
Private Mental Health Alliance Centralised Data Management ServiceData on the number of people seen by private hospital-based psychiatric services, and their outcomes, are analysed by the Private Mental Health Alliance's Centralised Data Management Service. 80

Virtually all private hospitals with psychiatric beds in Australia have been routinely collecting and reporting a nationally agreed suite of clinical measures and related data since 2002. The clinical measures to be collected, and the timing of their collection, are guided by a protocol.

Valid data for private hospitals in 2009-10 covered 76% of in-scope inpatient episodes.

lndicator 13 (Figure 59)

lndicator 23 (Figure 67)
Supported Accommodation Assistance Program (SAAP) National Minimum Data Set 2005-06 to 2009-10The Supported Accommodation Assistance Program (SAAP) National Minimum Data Set (NMDS) 2005-06 to 2009-10 81 includes information about all clients receiving SAAP support lasting more than one hour. The information is collected throughout the year. The SAAP NMDS is compiled by collating information provided by agencies across Australia and by State and Territory community service departments. Analysis of the SAAP NMDS is conducted by the Australian Institute of Health and Welfare. 81

The SAAP NMDS includes information from three collections: the client collection, the demand collection and the administrative collection. The client collection captures information on all clients receiving ongoing or substantial support under SAAP. It includes basic socio-demographic information and the services required by and provided to each client. Details about accompanying children are also obtained. Additionally, information is collected about the client circumstances before and after receiving SAAP support.

lndicator 19 (Figure 63)
Top of page

Table A2-2: Explanatory notes to figures and tables presented in Part 3.

Priority area 1: Social inclusion and recovery

lndicator 1a: Participation rates by people with mental illness of working age in employment: General population
(a) This indicator estimates the proportion of the Australian population aged 16-64 years with a mental illness who are employed. Data for 2011-12 is derived from the 2011-12 National Health Survey. 34 Data for 2007-08 is derived from the 2007-08 National Health Survey.36

The 2007-08 and 2011-12 National Health Surveys included questions about the respondent's mental health status and participation in employment. Mental illness was defined as self-reported mental or behavioural problems lasting six months or more, or which the respondent expects to last for six months or more. Persons were classified as employed according to the ABS quarterly Labour Force Survey definition, that is, if they reported in the preceding week that they had worked in a job, business or farm, or if they had a job but were absent during that week. The data collected from these surveys enables comparison between the employment rate for people with and without a mental illness. The data have been age-standardised to enable comparison between 2007-08 and 2011-12.

(b) Given the relationship between employment and labour force participation and severity of mental illness, methodological aspects of the 2007-08 and 2011-12 National Health Surveys may influence the employment and labour force participation rates reported for people with mental illness. The six month duration criterion used to determine the presence of mental illness is likely to exclude people with milder forms of mental illness that resolve within this period. In addition, as with other household surveys, 2007-08 and 2011-12 National Health Survey samples may underrepresent people with more severe mental illnesses. 36 Top of page

lndicator 2a: Participation rates by young people aged 16-30 with mental illness in education and employment: General population
(c) This indicator estimates the proportion of the Australian population aged 16-30 years with a mental illness who are employed and/or are enrolled for study towards a formal secondary or tertiary qualification. Data for 2011-12 is derived from the 2011-12 National Health Survey. 34 Data for 2007-08 is derived from the 2007-08 National Health Survey. 36

The 2007-08 and 2011-12 National Health Surveys included questions about the respondent's mental health status and participation in employment and education. Mental illness was defined as self-reported mental or behavioural problems lasting six months or more, or which the respondent expects to last for six months or more. Respondents were classified as employed if they had a job or business, or undertook work without pay in a family business for a minimum of one hour per week, or if they were absent from a job/business. Respondents were classified as participating in education if they were currently enrolled, whether full-time or part-time, in secondar y school, university/ other higher education, TAFE/technical college, business college, industry skills centre, or other relevant educational institution. Enrolment in adult education courses, hobby and recreation courses were excluded. The data collected from these surveys enables comparison between the employment and education rates for people with and without a mental illness. The data have been age-standardised to enable comparison between 2007-08 and 2011-12.
(d) As per note (b).
Indicator 3: Rates of stigmatising attitudes within the community
(e) This indicator reports average scores on a measure of social distance. Social distance is the degree of closeness people are comfortable with in relation to particular groups, such as individuals with mental disorders. The desire for social distance is recognised as one component of the stigmatising attitudes and beliefs directed towards people with mental disorders. 82

Social distance has been measured in the National Surveys of Mental Health Literacy and Stigma conducted in 2003-04 and 2011. These surveys assessed rates of stigmatising attitudes in Australia using measures of social distance, which are indicators of the willingness of Australians to interact with people suffering from a range of mental disorders, in a variety of situations.

In these surveys, respondents were read one of four vignettes describing a male ('John') or female ('Jenny') with depression, depression with suicidal thoughts, early schizophrenia and chronic schizophrenia. In 2011, social phobia and post-traumatic stress disorder were also included. Respondents were asked to rate their willingness to : (1) live next door to John/Jenny; (2) spend the evening socialising with John/Jenny; (3) make friends with John/Jenny; (4) work closely with John/Jenny; and (5) have John/Jenny marry into their family. Each of these five items was rated on a scale ranging 1 ('definitely willing') to 4 ('definitely unwilling'). A 'social distance' score was calculated by summing the ratings for each of the
5 items (maximum score 20).45,46,83 36 Top of page

Indicator 4: Percentage of mental health consumers living in stable housing
(f) Data on a range of outcomes for consumers of state and territory public sector mental health services are collected via the National Outcomes and Casemix Collection (NOCC). 78
The majority of the instruments in the NOCC suite assess clinic al outcomes like severity of symptoms and level of functioning, but a new measure of social inclusion is currently under development. Known as the Living in the Community Questionnaire (LCQ), this measure
will be completed by consumers and will assess par ticipation in various life domains. It will include an emphasis on stability of housing, which will ultimately inform this indicator.

For now, proxy data on this indicator are taken from the Health of the Nation Outcome Scales (HoNOS) for adults (aged 15-64) and the HoNOS65+ for older adults (aged 65+). The HoNOS and HoNOS65+ are core clinician-rated instruments in the NOCC suite of measures. These measures are administered routinely at selected points during episodes of care in state and territory public sector mental health services. Item 11 on these instruments is concerned with problems with living conditions and is scored from 0 (no problem) to 4 (severe to very severe problem). The percentage of consumers scoring 0 on admission to episodes of inpatient, ambulatory and residential care is taken as a proxy for the percentage of consumers living in stable housing.

These data provide an indicator of the housing status of consumers but should be interpreted with caution for several reasons. Item 11 on the HoNOS and HoNOS65+ relies on the clinician knowing the living circumstances of the consumer and is not optimally completed. 36 Top of page

Priority area 2: Prevention and early intervention

Indicator 6: Proportion of primary and secondary schools with mental health literacy component included in curriculum
(g) It was originally intended that data from Kidsmatter and MindMatters routinely collected by the Australian Government Department of Health and Ageing (DoHA) could be used to assess progress against this indicator. However, practical and conceptual issues prevented this. Firstly, only relatively basic data is captured on Kidsmatter and MindMatters. Secondly, these programs offer organising frameworks for mental health literacy rather than providing specific curriculum content, making it difficult for routinely collected data regarding these programs to gauge the extent and nature of curriculum developments. More importantly, while MindMatters and Kidsmatter are funded by the Australian Government, there are other mental health frameworks used by schools that would not be captured through DoHA's reporting arrangements.

For this reason, Australian Government funding was provided to expand the Principals Australia's National Market Research Survey in 2011 50 to collect specific data to inform this indicator, at least as an interim measure. The mental health questions in the survey included the following filter question which forms the basis of this indicator:

"Does your school currently:

  • Have mental health frameworks implemented and in use (f or example, Kidsmatter, MindMatters etc.)- followed with a question on specific details
  • Provide mental health programs for staff, students or parents - followed with a question on specific details
  • Have mental health literacy resources that can be accessed by teachers and students (for example,specific printed material, web resources to online services, use computer programs etc.)".
lndicator 7: Rates of contact with primary mental health care by children and young people
(h) Data on the number of children and young people receiving relevant Medicare-funded services are provided by the Australian Government Department of Health and Ageing, based on Medicare Benefits Schedule data. 74

Relevant services relate to Medicare item numbers covering: consultations with private psychiatrists, consultations with GPs for mental health specific services (i.e., GP-related Better Access item numbers and a small number of other relevant item numbers, but not item numbers related to general consultations), and consultations with allied health professionals (i.e., Better Access and Enhanced Primary Care Strategy item numbers
covering services provided by psychologists, social workers and occupational therapists). Data are based on the year in which the Medicare claim was processed, not the year in which the service was rendered. Top of page

Indicator 8: Rates of use of licit and illicit drugs that contribute to mental illness in young people
(i) Data for this indicator come from the National Drug Strategy Household Surveys 77 These surveys provide insights into whether patterns of drug and alcohol misuse by young people have changed over time.

The survey has undergone some methodological changes over time with, for example, a computer-assisted telephone interview being dropped in 2010 in favour of self-completion booklets. Data on alcohol use are presented here from all surveys from 2001 onwards, and data on cannabis and amphetamine use are presented from all surveys from 1998 onwards.

Indicator 9: Rates of suicide in the community
(j) The data for Figure 55 were sourced from the Australian Bureau of Statistics Causes of Death, Australia, 2011, report. Figure 56 is based on recent unpublished data provided by the Australian Bureau of Statistics. The 2007-11 figures vary slightly from those presented in Figure 55 due to a different upper age group being used in the calculation of each rate. 53

Until recently, the cause of death data for a given year were finalised by the ABS at a particular point in time, and cases that were still under investigation by the coroner in the relevant year were not reflected in the statistics for that year, even if they were subsequently judged by the coroner to be suicides. Recently, this anomaly has been rectified and now when cause-of-death determinations for a given year are forwarded from coroners, the ABS updates data from previous years. However, this improved method will only be applied to deaths registered after 1 January 2006, which means that data in very recent years and data from pre-2007 is likely to represent something of an undercount. 53

The causes of death data reported for 2006, 2007 and 2008 have undergone revisions and are now considered final. Causes of death data for 2009 and 2010 have been revised and are subject to further revisions. Causes of death data for 2011 are preliminary and subject to a revision process. 53 Top of page

lndicator 11: Rates of understanding of mental health problems and mental illness in the community
(k) This indicator reports the percentage of adults who accurately recognise a range of mental disorders. Accurate recognition of individual mental disorders is one indicator of mental health literacy. 57

Data for this indicator come from the National Surveys of Mental Health Literacy and Stigma, conducted in 1995, 2003-04 and 2011. 44 These surveys have used a vignette-based approach to investigate the ability of the Australian population to accurately identify a variety of mental disorders. Respondents were read one of several vignettes describing a male ('John') or female ('Jenny') with depression and early schizophrenia (assessed in all years), and depression with suicidal thoughts and chronic schizophrenia (assessed in 2003-04 and 2011), and social phobia and post-traumatic stress disorder (assessed in 2011). After being presented with the vignette, respondents were asked what, if anything, they thought was wrong with John/Jenny.45,46,83

Indicator 12: Prevalence of mental illness
(I) Information on the prevalence of common mental disorders among adults comes from the
National Surveys of Mental Health and Wellbeing, conducted in 20078,9 and 1997.4

There were several methodological differences between the two surveys which should be taken into account when comparing their findings:

  • The 1997 survey recruited people aged 18-99, whereas the 2007 survey recruited people aged 16-85.
  • The 1997 survey had a substantially higher response rate than its 2007 counterpart (78% versus 60%).
  • The 1997 survey focused on providing prevalence estimates over a 12 month timeframe, whereas the 2007 survey was designed to provide lifetime prevalence estimates and 12 month estimates were derived.
  • The two surveys used different algorithms to derive diagnoses.
(m) Information on the prevalence of clinically significant mental health problems among children and adolescents comes from the Child and Adolescent Component of the National Survey of Mental Health and Wellbeing, conducted in 1998. 6 This survey recruited 4,509 children and adolescents aged 4-17 through a household sampling strategy. It elicited information from participants and their parents via interview. A second child and adolescent survey has been commissioned and will collect data from May to December 2013. Top of page

Priority area 3: Service access, coordination and continuity of care

Indicator 13: Percentage of population receiving mental health care
(n) Data on the number of unique individuals seen by state and territory community mental health services are based on Department of Health and Ageing analyses of data submitted by jurisdictions. These data are provided by states and territories as person counts. Person counts are confined to those receiving one or more contacts with a community mental health service. This approach picks up most people seen in inpatient services too, since the majority of these would also be seen by a community team. The submitted service contacts are counted, including those delivered 'on behalf' of the consumer (i.e., where the consumer does not directly participate). This approach ensures that the role of state and territory mental health services in providing back-up specialist services to other health care providers is captured. It should be noted that states and territories differ in their capacity to provide accurate estimates of individuals receiving community mental health services because some (South Australia and Tasmania) do not have comprehensive unique identifier or data matching systems. In addition, jurisdictions differ in their approaches to counting individuals in receipt of services. Most record all individuals seen, but some - most notably Victoria - only count the individual once a clinical decision has been made to accept the person for treatment.

(o) Data on the number of unique individuals receiving relevant Medicare-funded services are based on Department of Health and Ageing analyses of Medicare Benefits Schedule data.74 Data are based on the year in which the Medicare claim was processed, not the year in which the service was rendered.

(p) Data on the number of unique individuals seen by state and territory community mental health services and data on the number of unique individuals receiving relevant Medicare-funded services are converted to percentages using population denominator data taken from the 2006 Census.

(q) Data on the number of people seen by private hospital-based psychiatric services were provided by the Private Mental Health Alliance Centralised Data Management Service.

(r) Work is underway by the Australian Institute of Health and Welfare to use data linkage to more accurately identify the extent of duplication in consumer counts between state and territory services and MBS-subsidised mental health care. This work is progressing with the assistance of jurisdictions and in compliance with ethical requirements. Top of page

lndicator 14: Readmission to hospital within 28 days of discharge
(s) Data on 'in scope' separations from state and territory acute psychiatric inpatient units in each financial year are based on Department of Health and Ageing analyses of data submitted by jurisdictions. 'In scope' separations are defined as those for which it is meaningful to examine readmission rates, and exclude, for example, same day separations and overnight separations that occur through discharge/ transfer to another hospital.

Readmissions are defined as admissions to any public acute psychiatric unit within the given jurisdiction that occur within 28 days of the original discharge. In order to determine whether the same individual was discharged from one unit and readmitted to a different unit, it is necessary for a system of unique identifiers to be in place that allows individuals to be 'tracked' across sites. Such systems have been available in all states and territories for the full period (2005-06 to 2010-11), with the exception of Tasmania (which introduced such a system in 2007-08) and South Australia (which has not yet introduced such a system). The absence of such a system will lead to an undercount of the true readmission rate.

Available data do not yet allow a distinction to be made between planned and unplanned readmissions.

lndicator 15: Rates of pre-admission community care
(t) Estimates for this indicator are based on Department of Health and Ageing analyses of data submitted by jurisdictions. Each jurisdiction provides data on 'in scope' separations from their acute psychiatric inpatient units in each financial year. 'In scope' separations are defined as those for which it is meaningful to examine rates of pre-admission community care, and exclude, for example, same day separations and overnight separations that occur through discharge/ transfer to another hospital.

Community mental health contacts are defined as contacts with any public community mental health team within the given jurisdiction that occur within the week before the inpatient admission. Except in the Northern Territory, these contacts are restricted to those in which the consumer participates directly. These may be face-to-face or indirect (for example, by telephone), but do not include those delivered 'on behalf of the consumer'.

In order to determine whether the same individual was admitted to an acute inpatient unit and received pre-admission community care, it is necessary for a system of unique identifiers to be in place that allows individuals to be 'tracked' across service settings. Such systems were available in all states and territories for the full period (2005-06 to 2010-11), with the exception of Tasmania and South Australia. The absence of such a system may underestimate the true rate of pre-admission care.

Only contacts with state and territory community mental health services are included here. Contacts with other community-based providers (for example, GPs and private psychiatrists) are excluded. Top of page

lndicator 16: Rates of post-discharge community care
(u) Estimates for this indicator are based on Department of Health and Ageing analyses of data submitted by jurisdictions. Each jurisdiction provides data on 'in scope' separations from their acute psychiatric inpatient units in each financial year. 'In scope' separations are defined as those for which it is meaningful to examine rates of post-discharge community care, and exclude, for example, same day separations and overnight separations that occur through discharge/ transfer to another hospital.

Community mental health contacts are defined as contacts with any public community mental health team within the given jurisdiction that occur within the week after discharge from the inpatient unit. Except in the Northern Territory, these contacts are restricted to those in which the consumer participates directly. These may be face-to-face or indirect (for example, by telephone), but do not include those delivered 'on behalf of the consumer'.

In order to determine whether the same individual was admitted to an acute inpatient unit and received post-discharge community care, it is necessary for a system of unique identifiers to be in place that allows individuals to be 'tracked' across service settings. Such systems were available in all states/territories for the full period (2005-06 to 2010-11), with the exception of Tasmania and South Australia. The absence of such a
system may underestimate the true rate of post-discharge care.

Only contacts with state and territory community mental health services are included here. Contacts with other community-based providers (for example, GPs and private psychiatrists) are excluded. Top of page

Indicator 19: Prevalence of mental illness among homeless populations
(v) Data for this indicator is based on analysis of the Supported Accommodation Assistance Program (SAAP) National Minimum Data Set 2005-06 to 2009-10.81

For the purpose of this indicator, SAAP clients are categorised into four mutually exclusive groups, based on their reasons for seeking assistance:

  • Those with mental health problems: This includes clients who were: referred from a psychiatric unit; reported psychiatric illness and/or mental health issues as reasons for seeking assistance; were in a psychiatric facility before or after receiving assistance; and/or needed, were provided with or were referred on for support in the form of psychological or psychiatric services.
  • Those with substance use problems: This includes clients who: reported problematic drug, alcohol and/or substance use as a reason for seeking assistance; and/or needed, were provided with or were referred on for support in the form of drug and/or alcohol support or intervention.
  • Those with comorbid mental health and substance use problems: This includes clients who reported at least one of the mental health criteria and at least one of the substance use criteria listed above in the same support period.
  • Other: This includes clients who reported none of the criteria listed above.
A client may have more than one support period within a year and their circumstances might vary between support periods.

Routinely collected SAAP data are likely to underestimate the true prevalence of mental illness among homeless populations because they focus on clients whose referral to SAAP was associated with these problems. They do not take into account clients who may have underlying conditions that are not directly responsible for the referral. SAAP data have now been replaced with the Specialist Homelessness Services Collection (SHSC). The SHSC is designed to provide more comprehensive data on clients of specialist homelessness services. Options for using the SHSC to assess the achievement of this indicator in future National Mental Health Reports are currently being explored. Top of page

Indicator 20a: Prevalence of mental illness among people who are remanded or newly sentenced to adult correctional facilities
(w) The data for this indicator come from the 2010 National Prisoner Health Census67,68 which was conducted by the Australian Institute of Health and Welfare. The Census was conducted over a two week period in 44 adult correctional facilities from all jurisdictions except New South Wales and Victoria. Individuals who entered 44 adult correctional facilities from all jurisdictions except New South Wales and Victoria over a two week census period were asked a number of questions, including several about their mental health. Self-reported information on prison entrants' mental health status was sought across three domains:
  • Mental health history: This was assessed by a single question - 'Have you ever been told by a doctor, psychiatrist, psychologist or nurse that you have a mental health disorder (including drug and alcohol abuse)?'
  • Current mental health medication: This was also assessed by a single question - 'Are you currently on medication for a mental health disorder?'
  • Current psychological distress: This was assessed by the Kessler-10 (K-10), which measures non-specific psychological distress. 69 The K-10 comprises 10 items relating to symptoms of depression and anxiety in the past four weeks. Each item is rated from 1 (None of the time) to 5 (All of the time), resulting in a total score that ranges from 10 to 50. Standard cut-off scores for levels of psychological distress are as follows: 10-15 (Low); 16-21 (Moderate); 22-29 (High); >=30 (Very high). Top of page

Priority area 4: Quality improvement and innovation

Indicator 21: Proportion of total mental health workforce accounted for by consumer and carer workers
(x) This indicator measures the proportion of the state and territory mental health workforce who are consumer and carer workers. The data for this indicator are available through the National Minimum Data Set (NMDS) - Mental Health Establishments (MHE).75

The NMDS-MHE captures information about the size and composition of the mental health workforce, including direct care staff. Direct care staff comprises Consultant psychiatrists and psychiatrists, Psychiatry registrars and trainees, Other medical officers, Registered nurses, Enrolled nurses, Psychologists, Social workers, Occupations therapists, Diagnostic and health professionals, Other personal care, Consumer workers, and Carer workers. FTE counts for consumer and carer workers are only available from 2002-03 onwards. The definition of these categories was modified from 'consultants' to 'mental health workers' for the 2010-11 collection, in order to capture a broader array of consumer and career roles, and this may impact on the figures reported.

It is calculated as the number of full-time equivalent consumer and carer worker positions within Australian state and territory public mental health services, over the number of full-time equivalent clinical positions within Australian state and territory public mental health services.

A revision of the current, nationally agreed definition of consumer and carer workers is currently being undertaken to improve consistency in how jurisdictions report the variety of arrangements that exist between organisations and consumer and carer workers. The current data collection does not include mental health services managed by non-government organisations. The development of a Mental Health Non-Government
Organisation National Minimum Dataset is currently underway, and is it desirable that data to inform this indicator be included in that collection. Top of page

Indicator 22: Proportion of services reaching threshold standards of accreditation under the National Mental Health Standards
(y) The data for this indicator are available through the National Minimum Data Set (NMDS) Mental Health Establishments (MHE).75

The NMDS-MHE captures information about the extent of progress made by specialised mental health service units in implementing the National Standards for Mental Health Services, summarised into categories. The indicator grades services according to four categories:

  • Level 1 - Services that have been reviewed by an external accreditation agency and judged to have met all National Standards for Mental Health Services.
  • Level 2 - Services that have been reviewed by an external accreditation agency and judged to have met some but not all National Standards.
  • Level 3 - Services that are either in the process of being reviewed by an external accreditation agency but the outcomes are not known; or are booked for review by an external accreditation agency.
  • Level 4 - Services that do not meet the criteria detailed under levels 1 to 3.
The indicator is based on the expenditure reported for each of the service units accredited at the various levels. This method takes account of the size of the service unit, and the number of service units per jurisdiction, and is therefore considered a more accurate reflection of the proportion of mental health services meeting each level.

The current coverage of this indicator excludes service units that are non-government mental health service units and private hospital service units in receipt of government funding where the National Standards for Mental Health Services do not apply. It also excludes aged care residential services subject to Australian Government residential aged care reporting and service standards requirements. Top of page

Indicator 23: Mental health outcomes for people who receive treatment from state and territory services and the private hospital system
(z) Data for this indicator come from the National Outcomes and Casemix Collection (NOCC)78
and the Private Mental Health Alliance.80

For the purposes of this indicator, assessment of clinical outcomes is based on the clinician-rated Health of the Nation Outcome Scales (HoNOS), and its equivalents for children and adolescents (HoNOSCA) and older people (HoNOS65+). All three comprise items that collectively cover the sorts of problems that may be experienced by people with a mental illness. Each item is rated from 0 (no problem) to 4 (very severe problem), resulting in individual item scores, subscale scores and a total score.

HoNOS/ HoNOSCA/ HoNOS65+ data for consumers of state and territory public sector mental health services are collected via the National Outcomes and Casemix Collection (NOCC) and analysed by the Australian Mental Health Outcomes and Classification Network (AMHOCN). Equivalent data for consumers seen in private psychiatric hospital units are collected and analysed by the Private Mental Health Alliance's Centralised Data Management Service.

Outcomes according to the HoNOS family of measures are considered for four cohorts of consumers who received episodes of care during 2010-11. Outcome scores are calculated differently for these groups, depending on the setting and the duration of the episode of care:

  • Those discharged from hospital in both the public and private sector include people who had an inpatient admission that began and ended during the 2010-11 year and lasted at least three days. Outcome scores for these groups are calculated as the difference between the total HoNOS/ HoNOSCA/ HoNOS65+ scores recorded at admission to and discharge from inpatient care.
  • Those discharged from community care in the public sector include people who received an episode of community care that began and ended in 2010-11. Outcome scores for this group are calculated as the difference between the total HoNOS/ HoNOSCA/ HoNOS65+ scores recorded at admission to and discharge from community care.
  • Those in ongoing community care in the public sector include people who were receiving community care for the whole of 2010-11 and those who commenced community care some time after 1 July 2010 and continued to receive care for the rest of the year. The defining characteristic for this group is that all were still in ongoing care when the year ended (30 June 2011). Outcome scores for this group are calculated as the difference between the total HoNOS/ HoNOSCA/ HoNOS65+ scores recorded on the first and last occasions rated during the year. Top of page
In each case, outcome scores are classified based on 'effect size'. 'Effect size' is a statistic used to measure the magnitude of a treatment effect. It is based on the ratio of the difference between pre- and post- scores to the standard deviation of the pre-score. As a rule of thumb, effect sizes of 0.2 are considered small, 0.5 are considered medium, and 0.8 are considered large. Based on this rule, a medium effect size of 0.5 is used to assign outcome scores to categories - an effect siz e of greater than or equal to +0.5 equates to 'significant improvement', an effect size of -0.5 to +0.5 equates to 'no change', and an effect size of less than or equal to -0.5 equates to 'significant deterioration'.

The denominator in the analysis for each of the four cohorts is 'valid' episodes of care. To be considered valid, the episode had to have sufficiently complete HoNOS/ HoNOSCA/ HoNOS65+ data that total scores could be calculated at its beginning and end. It has been estimated that valid 2010-11 data were available for 34% of public sector inpatient episodes, 23% of public sector community episodes, and 80% of private sector inpatient episodes. It should be noted that, except in the case of ongoing community episodes, an individual may have had more than one episode during 2010-11 so the data represent episode-counts, rather than person-counts. This means that some individuals may appear more than once within a given group.

Data coverage has been estimated at around one third of potential inpatient episodes and around one quarter of community care episodes. Coverage varies widely across jurisdictions. Changes in coverage may change the pattern of results.