The evaluation of the Better Access initiative required drawing upon several data sources. Comparing disparate data sources which were not developed for the purpose of an evaluation of the mental health workforce was difficult because there were many areas in which direct concordance was not possible. To make the most of the data available from these sources, we sought to develop ways of interpreting the data so that reasonable comparisons could be made. We discuss the methods used in developing measures that could be compared across data sets, occupational categories, public-private sectors, and regions in this section. Despite enhancing the comparability of the data, throughout the report we urge caution in drawing generalisable conclusions where data are particularly problematic.

1.4.1 MBS and Medicare provider data
1.4.2 ABS 2006 Census of Population and Housing
1.4.3 Mental Health Establishments: National Minimum Data Set
1.4.4 Registration boards and administrative data
1.4.5 Consultations
Supplementary material

1.4.1 MBS and Medicare provider data

The Department of Health and Ageing provided unpublished data from the Medicare Benefits Schedule for Items associated with Better Access, and associated data from the Medicare Providers database (2004-2008).

MBS data for services provided between November 2006 and December 2008, inclusive, provided information about the actual Better Access workforce. Information from the Medicare Providers database informed calculations of estimates for the potential Better Access workforce, although these estimates were supplemented with information from other data sources (see section 1.4.2 below for details).

Medicare data were used in two sections of the report. In chapter 2, the data were split into calendar years (January – December) and analysed for each occupation in relation to the required characteristics for 2006, 2007 and 2008. The collation of data into calendar, rather than financial, years meant that there were two full years of Better Access to analyse,5 with a further 2 months (9 weeks) of data for 2006. In chapter 3, Medicare data were used to examine the supply and distribution of health professionals in each of the allied mental health provider categories.
Top of page
The following information relates to our use of the MBS data:
  • The analysis in chapter 2 and chapter 3 primarily used numbers of people (headcount) in each category.

  • The headcount in each provider category using Better Access Item numbers were identified using the service providers' 'provider number'6. These figures, although the most reliable available, only give an approximate number of service providers as it is possible for individual providers to have more than one provider number (particularly if an individual provides services in multiple locations or across more than one state/territory). The numbers of service providers registered with Medicare Australia were also provided, for each of the relevant provider categories.

  • Service provider categories were derived from use of occupationally specific MBS Items. For example, all service providers using MBS Items 2710, 2712 and 2713 were defined as General Practitioners. See appendix A for the Item numbers associated with each occupation.

    Given that a criterion for registration with Medicare is registration with a professional body, the use of occupationally specific MBS Item numbers provides reliable data for the standard defined occupational categories – general practitioner, psychiatrist, psychologist, social worker, OT.

    However, the use of MBS Items to differentiate between clinical psychologist and registered psychologist, within the broader psychologist occupational category, distinguishes between types of services rather than occupations. Hence, clinical psychologists can provide services for MBS Items allocated to registered psychologists. For example, 297 psychologists provided both Psychological Therapy Services and Focussed Psychological Strategies in 2006. This number rose to 1096 in 2007 and 1181 in 2008 (MBS data 2006-2008). When based on MBS Items, the occupational demarcation between clinical and registered psychologists is therefore less clear. For the purposes of this report, clinical and registered psychologists are called provider categories rather than occupational categories. Due to these data issues, clinical and registered psychologists were grouped under the category of 'psychologist' when data was compared across data sets, and separately when the MBS data was used in isolation.

  • Information about educational attainment and country of birth were provided by the Department of Health and Ageing but did not comply with standard (ABS) classifications and were not used. It was assumed that all providers met Medicare eligibility standards for their occupation, including qualification.

  • Deemed full time equivalent (DFTE) numbers have been calculated for each of the provider categories using MBS data. This information is reported in chapter 3. The DFTE was calculated by converting the hours billed against the MBS for Better Access Items to an estimated fulltime equivalent. This involved several steps:

    • Calculating the number of billed hours: for each Better Access MBS Item number, the recommended consultation time for the Items was allocated to produce a conservative estimate of the number of hours spent by providers on Better Access services. As the standard by which time was recommended was not uniform across the various MBS Items, we have based our calculations on the following:

      • Using the minimum recommended time where this is the only time provided;

      • Using the midpoint recommended time if a time range was specified;

      • Where no recommended time was specified (e.g. for MBS Items 2710 and 2712), BEACH data7 was used to estimate an average consultation time of 25 minutes.

    • The average number of billed hours per occupation per year was then divided by the number of working weeks in the year. Allowing for annual and sick leave, it was estimated that there were 46 working weeks per calendar year in the private sector, equating to 8 weeks in 2006 (Nov-Dec only). This figure provided an indication of the average number of hours per week that each allied mental health occupation worked on Better Access services.

    • To estimate how many fulltime health professionals it would take to deliver the number of Better Access hours billed, a conversion factor was used. This conversion factor took account of the difference between hours worked and hours billed (or claimed via MBS). Using information from various sources including the Australian Institute for Health and Welfare (AIHW), the Mental Health Establishments: National Minimum Data Set and consultations with the Australian Psychological Society, it was estimated that approximately 85% of time worked was billed. From the information available we therefore estimated that the allied mental health workforce would claim 30 hours per week, while a full time medical mental health professional would claim 33 hours per week.

    • The average number of hours per week spent on MBS Items for each occupational group was then divided by the conversion factor to produce the deemed fulltime equivalent (DFTE) for each group.

  • The MBS database provided Rural, Remote and Metropolitan Areas (RRMA) categories of geographical classification, based on population density, to describe the geographical distribution of service providers.8 In section 3.2, these regions have been grouped into urban and rural based on the categories in table 1.1:
Top of page

Table 1.1 Comparison of the RRMA and non-standard geographical categories

Table 1.1 is presented as a list in this HTML version for accessibility reasons. It is presented as a table in the PDF version.

RRMA comparison of the following non-standard geographics categories:
  • Urban areas: capital cities and large metro areas; and other metro areas.
  • Rural areas: large rural areas; small rural centre; other rural centre; remote centre; and other remote centre.

1.4.2 ABS 2006 Census of Population and Housing

For the purpose of this evaluation, the ABS 2006 Census of Population and Housing (ABS Census) provided information on both the estimated numbers and the demographic, geographic and labour market characteristics of the potential Better Access workforce, for each occupational category. The ABS Census is the largest survey in Australia. Involving a complete enumeration of the population, rather than of a population sample, it does not suffer the sampling problems of other surveys and allows the analysis of small or highly specific groups to be targeted. Furthermore, the collection of information for the ABS Census occurred in August 2006, only a few months before the Better Access initiative was implemented.

The ABS has policies to reduce collection and processing errors occurring in the Census. In 2006, the ABS Census Post-enumeration Survey (PES) indicated that there was a net undercount of 549,4869 Australians (approx. 2.7% of the estimated population). While this may slightly under-estimate the numbers for each occupational category, the size of the ABS Census and the sampling technique used ensures that the underlying characteristic distributions are a true reflection of the population. With a 97.3% response rate, the ABS Census was therefore considered to be the most accurate and comprehensive data source for the discussion of the characteristics of occupations in the Better Access workforce. It was the only data source that provided consistent data about workforce characteristics across all occupational groups. The required data was purchased from the ABS specifically for this project and is primarily used in chapter 2.

The ABS and MBS databases differ across some categories and, in using the analysis of these databases, the following should be noted:
  • In the ABS Census, occupations are defined according to the ABS Australian and New Zealand Standard Classification of Occupations (ANZSCO). The relationship between the ABS and MBS occupations are shown in table 1.2. The occupation information collected in the ABS Census is self-enumerated, applicable only to those employed and aged 15 years and over. The ABS Census asks individuals to provide the 'full title of' and the 'main tasks undertaken in' the occupation of their 'main job' held in the 'last week'.10 This information is then categorised by the ABS. This method of self-reporting (and subsequent coding) of occupations may result in slightly under/over-estimate of the numbers for each occupation category. In determining numbers of the potential Better Access workforce, the ABS Census is therefore used in conjunction with other data sources.

  • The ABS uses the Australian Standard Geographical Classification (ASGC) based on either population size or remoteness from services, to describe the geographical distribution of health professionals. The lack of concordance between RRMA and ASGC meant that direct comparison of the geographic distribution of the actual and potential Better Access workforce was limited. Instead, comparisons of distribution as discussed in chapter 2 have been estimated based on broad delineation of metropolitan and rural (includes rural, remote areas) as per table 1.3.

Table 1.2 Comparison of occupational categories, ABS and MBS

ANZSCO codeABS occupational categoryMBS provider category
272300Psychologists*Clinical Psychologists**
Registered Psychologists
272511Social WorkersSocial Workers
252411Occupational TherapistsOccupational Therapists (OTs)
253111Generalist Medical PractitionersGeneral Practitioners (GPs)
253411PsychiatristsPsychiatrists
253321PaediatriciansPaediatricians

* Note: The ANZSCO Major Group (6-digit) occupation categories include: Clinical Psychologist (272311), Educational Psychologist (272312), Organisational Psychologist (272313), Psychotherapist (272314), and Psychologists nec (272399).
** Note: In consultation with the Department of Health and Ageing, the more detailed ANZSCO Major Group (6-digit) 'Clinical Psychologist' (272311) occupation was not selected as a comparison to clinical psychologists in the MBS data. The ANZSCO definition does not appropriately reflect the level of qualification or experience required by the Australian Psychological Society (APS) or Medicare Australia.

Top of page

Table 1.3 Comparison of ASGC, RRMA and non-standard geographical categories

ASGC (ABS)RRMA (MBS)Non-standard
Major urbanCapital cities and large metro areas

Metropolitan
Other urban

Bounded locality

Rural balance

Migratory
Other metro areas

Large rural areas

Small rural centre

Other rural centre

Remote centre

Other remote centre

Rural

1.4.3 Mental Health Establishments: National Minimum Data Set

A collection of Mental Health Establishments data is conducted on behalf of the Department of Health and Ageing annually as at June 30.11 It collects information from specialised mental health services that are managed or funded by the state and territory governments. The services are publicly funded and are therefore outside of the MBS system (although there are 5-6 private hospitals with publicly funded services). Health professionals in the public mental health workforce may also work in the private sector or may transition between the two. In effect, the public mental health workforce is part of the broader Better Access mental health workforce. The Mental Health Establishments: National Minimum Data Set (MHENMDS) was used to examine whether Better Access had impacted on the distribution of health professionals across the Medicare and public health sectors. This is discussed in chapter 3.

The lack of concordance between the MHE: NMDS and MBS data in collection points, the unit of analysis and geographic distribution, meant that there were difficulties in making comparisons between occupational groups in the two sectors. In particular, each of the datasets has different collection points: the MBS is collected monthly and was reported on per calendar year; and the MHE: NMDS is collected annually on a financial year basis. There was no concordance in the timeframes across the data sources. We did, however, develop measures to make the MHE: NMDS and MBS data more comparable. This involved:
  • Reporting on estimates of full-time equivalent numbers in each of the data sources. The MHE: NMDS reports on the number of fulltime equivalent (FTE) places in each occupation. The MBS data was converted to provide numbers of deemed fulltime equivalent (DFTE) places in each occupation (explained above). While this has allowed a better comparison across the two sectors, the different methodologies in achieving these figures mean that caution is required when interpreting the results. In effect, there is still a lack of concordance between the datasets. Despite this lack of concordance, it was important to gain an understanding of the trends in the allied mental health workforce following the implementation of Better Access. The use of FTE and DFTE provided a better method to quantify the workforce than simple head counts, because the headcount method does not control for differences in parttime/ full-time status.

  • Converting non-concordant geographical categories to the more simplified capital – non-capital categories of analysis. In consultation with the Department of Health and Ageing, the MHE: NMDS (2004-2008) was manually coded according to a nonstandard geographic category based on whether an establishment was located in a 'capital city' or 'outside a capital city'. This provided a simple mechanism for examining whether Better Access has influenced the supply of the allied mental health workforce outside of capital cities. To make this comparable to the MBS data, which uses RRMA, the 'capital cities and large metro areas' was equalised to the 'capital city' non-standard category, with all other RRMA categories equalised to the 'not capital city' non-standard category.
Top of page

1.4.4 Registration boards and administrative data

Registration board and administrative data from professional associations were used as auxiliary information in estimating the numbers of health professionals in the potential Better Access allied mental health workforce. Part of the issue for acquiring data relating to the allied mental health workforce was that there were no consistent national registration requirements or administrative bodies that collect the data required.12 Table 1.4 summarises the jurisdictions for which data were publicly available.

To estimate the potential number of social workers, membership data was provided by the Australian Association of Social Workers (AASW).13 For psychologists and OTs, publicly available state/territory registration board administrative data were used to estimate potential numbers. The registration board data included members licensed/registered to practice, where required by state/territory legislation.14 In contrast to the ABS Census data, registration board data includes those (i) not employed (i.e. retired, on leave, unemployed or not in the labour force), (ii) employed, but not as a psychologist or occupational therapist, (iv) interstate or overseas. Due to the limitations of the information publicly available it is not known how many of the psychologists and OTs registered with their boards, were active practitioners. However, using South Australian data, from 2006 to 2008, it was estimated that approximately 7-8% of psychologists and 16-17% of OTs were not employed, 4-6% of psychologists and 2-3% of OTs resided interstate, and 2-3% of psychologists and 4-3% of OTs resided overseas.

For both the psychologists' and OTs' registration board data, the definitions and standards in licensing varied between the state/territory jurisdictions. In consultation with the department of Department of Health and Ageing, the licenses that reflect fully-trained psychologists (clinical and other) and OTs have been selected for analysis. Table 1.5 summarises the selected licenses for psychologists and OTs by state/territory used in this report.

For social workers, the AASW membership administrative data includes both unaccredited and accredited mental health social workers. Unlike the psychology and OT registration boards, the AASW is self-regulating and there are no state/territory legislative requirements for social workers accreditation. However, social workers must be a member of the AASW and have attained accreditation to register with Medicare and be able to utilise MBS Better Access Items. Specific data was acquired from AASW providing the numbers of accredited mental health social workers. Similar to the registration board data, the AASW membership data includes social workers (i) not employed (i.e. retired, on leave, unemployed or not in the labour force), (ii) employed, but not as a social worker (unaccredited and accredited), (iv) interstate or overseas. It is not known how many of the social workers (unaccredited and accredited), registered with the AASW, were active practitioners.

Table 1.4 Availability of registration board and administrative data for psychologists, occupational therapists and social workers, 2006-2008

State/Territory Psychologists
2006
Psychologists
2007
Psychologists
2008
Occupational Therapists
2006
Occupational Therapists
2007
Occupational Therapists
2008
Social Workers
2006
Social Workers
2007
Social Workers
2008
NSW
Available
Available
Available
n.a
n.a
n.a
Available
Available
Available
VIC
Available
Available
Available
n.a
n.a
n.a
Available
Available
Available
QLD
Available
Available
Available
Available
Available
Available
Available
Available
Available
SA
Available
Available
Available
Available
Available
Available
Available
Available
Available
WA
Available
Available
Available
Available
Available
Available
Available
Available
Available
TAS
Available
Available
Available
n.a
n.a
n.a
Available
Available
Available
ACT
Available
Available
Available
n.a
n.a
n.a
Available
Available
Available
NT*
n.a
n.a
n.a
n.a
n.a
n.a
Available
Available
Available

Note: for psychologists, occupational therapists and social workers, the collection of data varied from the end of the financial year and the end of the calendar year.
* Note: Administrative data from the Northern Territory Psychologists Registration Board and the NorthernTerritory Occupational Therapists Registration Board were not publicly available.

Top of page

Table 1.5 Selected categories of license/registration for psychologists and occupational therapists, by state/territory.

PsychologistsOccupational Therapists
NSWFully registered
-
Fully registered (temporary)
-
VICGeneral registration
-
Specific registration
-
QLDGeneral registrantsGeneral
Provisional general registrantsProvisional general
Deemed general registrantsDeemed
Deemed provisional general registrants
-
SAFull registrationFull registration
Limited registrationLimited registration
Deemed registrationDeemed registration
WAFully registeredGeneral
Specialist title
-
Mutual recognition
-
TASRegistered
-
ACTRegistered
-
NT*
-
-

* Note: Administrative data from the Northern Territory Psychologists Registration Board and the Northern Territory Occupational Therapists Registration Board were not publicly available.

1.4.5 Consultations

In examining the anticipated implications of the Better Access initiative for future workforce trends, consultations were held with representatives from key organisations associated with the Better Access workforce. The consultations were conducted in June, prior to the 2009-10 Budget release (Department of Health and Ageing 2009a), and therefore responses do not consider changes introduced at this time. The findings from the consultations are discussed in chapter 4.

The purpose of the consultations was to identify issues that may not yet have been apparent from the quantitative data due to the relatively short time frame (just over two years) between the implementation of the Better Access initiative and the evaluation. It would be difficult, for example, to assess the impact of Better Access on the capacity of training systems through the collection of quantitative data given that training systems typically have a lag time of more than two years depending on the occupation.

Speaking directly to key stakeholders provided an opportunity to canvass issues raised by the Department of Health and Ageing in relation to the impact of Better Access on training, and the effectiveness of referral pathways. The consultations therefore investigated the perceptions of key stakeholders about the impact of Better Access on different occupational groups.

As negotiated with the Department of Health and Ageing, stakeholders representing the Better Access occupations were invited to participate in the consultations. These representatives were primarily sourced from the peak medical/allied health associations and colleges and training providers, with two individuals participating. Table 1.6 identifies participants in the consultation component of this study.

A qualitative interview schedule (see appendix C) was designed and tailored to each organisation participating in the consultation. Ethics approval for this component of the evaluation was acquired from Monash University Standing Committee on Ethical Research in Humans. Nineteen interviews and a group discussion with the Board of the General Practice Mental Health Standards Collaboration were conducted. Except for the group discussion, each consultation consisted of a semi-structured interview of approximately 35 minutes duration. Each interview was digitally recorded and transcribed. The transcriptions were entered into the NVivo qualitative research analysis programme and analysed by coding responses thematically.

As with each of the data sources, there were limitations in the information gained through the consultations:
  • The sample of organisations from which the consultations were drawn was small and targeted. The aim was to get as broad a sample as possible with the available resources, covering each of the key stakeholder groups. This meant that the analysis covers a range of issues, but could not verify the extent to which they were widespread. For example, the discussion of the psychology workforce is based on consultations with four psychology organisations. Quantifying the responses to assess veracity was therefore not appropriate.

  • The consultations were primarily with representatives of groups. Sometimes these groups were large (e.g. Australian Psychology Association with over 20,000 members), while others were relatively small (e.g. Universities Australia with less than 50 members). In two instances, consultations were undertaken with individuals. No effort has been made to weight the responses according to the size of the organisation represented.

  • The participants did not have the benefit of the quantitative data upon which to make comments. Discussions therefore focused on their perceptions of what was happening for members of their particular organisation. Although this method enables the early identification of issues and could be viewed as a strength of the approach, it also means that the information provided was based on experience and observations which has limited generalisability.
Top of page

Table 1.6 Participants in the consultations

Table 1.6 is presented as a list in this HTML version for accessibility reasons. It is presented as a table in the PDF version.

Organisations consulted (including the acronyms used in the report):
  • Australian College of Rural and Remote Medicine (ACRRM)
  • Australian Association of Social Workers (AASW)
  • General Practice Registrars Australia (GPRA)
  • Overseas Trained Doctors Association (OTDA)
  • Australian Indigenous Psychologists Association (AIPA)
  • Australian College of Clinical Psychologists (ACCP)
  • Universities Australia
  • Australian Psychological Society (APS)
  • Royal Australian and New Zealand College of Psychiatrists (RANZCP)
  • General Practice Mental Health Standards Collaboration (GPMHSC)
  • Chair of GPMHSC
  • Australian General Practice Training (AGPT)
  • Australian Private Hospital Association (APHA)
  • Occupational Therapy Australia (OT Australia)
  • Mental Health Professionals (MHPN)
  • Royal Australian College of General Practitioners (RACGP)
  • Australian Psychology Accreditation Council (APAC)
  • Royal Australian College of Physicians (Paediatrics) (RACP (Paediatrics))
Individuals consulted:
  • Overseas trained doctor
  • Consumer representative

Supplementary material

The information collated through the consultations was supplemented by:
  • Material provided by some of the organisations that took part in the consultations, including:

  • Personal correspondence from a representative from the School of Behavioural Science at the University of Melbourne and the Director of the University of New South Wales Psychology Clinic.

  • KPMG survey questions: in collaboration with KPMG, who conducted Component D (consultations with stakeholders) of the broader Evaluation of the Better Access to Psychiatrists, Psychologists and GPs through the Medicare Benefits Schedule initiative, questions concerning access to clinical training were included in an internet based survey of mental health stakeholders (see appendix D). Please note that KPMG also interviewed many of the organisations listed above.

Footnotes

5 Although collation into financial years would have enabled concordance of collection points with the NMHS- MHE data, it would have only provided one full year of comparison (July 2007 – June 2008).
6 Provider numbers were randomised by Medicare Australia to ensure privacy.
7 Data provided by the Department of Health and Ageing (9/11/2009) from the BEACH (Bettering the Evaluation and Care in Health) dataset, The Family Medicine Research Centre, University of Sydney.
8 Although available in other categories of geographical classification, RRMA data was provided by Medicare for the purpose of this evaluation.
9 ABS, Census of Population and Housing – Details of Undercount, August 2006, Cat. no. 2940.0
10 ABS, Census Dictionary, 2006 (Reissue), Cat. no. 2901.0.
11 This data set includes both the Mental Health Establishments: National Minimum Data Set (2005-08) and the National Survey of Mental Health Services (1995-2006). In this report the combined data is referred to as the Mental Health Establishments: National Minimum Data Set (MHE: NMDS).
12 From 1 July 2012, psychologists and occupational therapists will be joining the Australian National Registration and Accreditation Scheme, which will provide consistency in the data. However social workers will continue to be self-regulating.
13 Social Workers do not have a registration board, although accredited social workers do need to be members of the Australian Association of Social Workers.
14 OTs are not required to register in all states/territories