Evaluation of the Medical Specialist Outreach Assistance Program and the Visiting Optometrists Scheme

Appendix F: Allocation of MSOAP services to ASGC remoteness areas

Page last updated: 28 February 2012

The national MSOAP Core database was analysed to identify all locations to which outreach services are provided. These locations were then assigned an ARIA+ score and ABS SEIFA index of disadvantage, using the methods described below. Locations were also mapped to Statistical Local Areas (SLA), and mean ARIA+ scores and weighted ABS SEIFA index of disadvantage were identified for each SLA. ARIA+ scores were then mapped to the ASGC-RAs for both the locality and the SLA within which the locality is situated.

The MSOAP locations are given in the form of location names, therefore it required mapping to known localities defined by the Australian Bureau of Statistics. The main classification considered was the Australian Standard Geographical Classification (ASGC). In particular, socioeconomic data at these levels were considered, in the order of preference:

    • Urban Centre/Locality (UC/L)
    • State Suburb (SSC)
    • Postal Code
    • Statistical Local Area (SLA).
The UC/L structure groups Census Collection Districts (CDs) together to form defined areas according to population size criteria. The resulting areas are known as Urban Centres or Localities. The boundaries do not generally coincide with SLAs, and its total area covers only part of Australia.

The SSC structure is a Census-specific area where Collection Districts are aggregated to approximate suburbs. It is applicable only to the larger urban centres.

In addition, the National Localities Index was used to map areas to SLAs. The National Localities Index provides Statistical Local Area (SLA) codes for over 32,000 Australian localities, and the coverage encompasses the whole of Australia. This index also enabled suburbs to be mapped to postal areas.

The data sources used for the analysis are:
ASGC data
    • 2905.0 – Statistical Geography: Volume 2 -- Census Geographic Areas, Australia, 2006
    • 2909.0 – Statistical Geography: Volume 3 – Australian Standard Geographical Classification (ASGC) Urban Centres/Localities, 2006
NLI data
    • 1252.0.55.001 - National Localities Index, Australia, Jul 2006
SEIFA
2033.0.55.001 - Socio-economic Indexes for Areas (SEIFA), 2006
    • SEIFA, Census Collection Districts
    • SEIFA, Statistical Local Areas
    • SEIFA, Postal Areas
    • SEIFA, State Suburb Code
ARIA
Accessibility Remoteness Index of Australia (ARIA+), 2006
    • ARIA+, Census Collection Districts
    • ARIA+, Statistical Local Areas
    • ARIA+, Localities
Methodology
The following section describes the steps involved in the analysis in more detail.

Step 1: Urban Centre/Locality level data

2909.0 – Statistical Geography: Volume 3 - Australian Standard Geographical Classification (ASGC) Urban Centres/Localities, 2006 is a mapping from Census Collection District to Urban Centre/Locality.

This was mapped with the SEIFA and ARIA data at Census Collection District level, which was then summarised to the Urban Centre/Locality level by using a weighted average of the scores based on the estimated populations in each Census Collection District within the Urban Centre/Locality.

The name of the Urban Centre/Locality was cleaned up so that it is in a similar format to the locality names in the MSOAP Locations database.

Step 2: State Suburb level data

The SEIFA data at state suburb level and ARIA data at locality level was combined as an alternative source of socioeconomic information where the UC/L level data was not available.

For analysis at this level, suburbs with the same name within the same state were removed to avoid confusion.

The name of the state suburb was cleaned up so that it is in a similar format to the locality names in the MSOAP Locations database.

Step 3: Suburb to SLA lookup dataTop of page

The NLI for July 2006 was used to derive SLAs and postcodes for each location. The NLI offers a more comprehensive list of locality names than UC/L and state suburb level data. The SLA information was used later to identify the proportion of each SLA that are covered by MSOAP.

Step 4: Consolidated data

The MSOAP locations database was merged with the UC/L, state suburb and NLI data using the names of the localities in each of the data sources. Furthermore, the postcode and SLA level SEIFA and ARIA data was also merged using the data derived in the NLI data.

Step 5: SLA consolidation

Some localities are located across more than one SLA. In order to allow the calculation of the proportion of each SLA that is covered by MSOAP, several SLAs were combined as follows:
    • If two SLAs were listed in the NLI database for an MSOAP location, these SLAs were combined.
    • If the population size of the SLA were smaller than the UC/L, an attempt was made to consolidate neighbouring SLAs that constitute the UC/L. These SLAs included:
        • Lismore (C) - Pt A, Lismore (C) - Pt B, Richmond Valley (A) - Casino and Richmond Valley (A) Bal
        • Tweed (A) - Tweed-Heads, Tweed (A) - Tweed Coast and Tweed (A) - Pt B
        • Cairns (C) - Barron, Cairns (C) - Central Suburbs, Cairns (C) - City, Cairns (C) - Mt Whitfield, Cairns (C) - Northern Suburbs, Cairns (C) - Trinity, Cairns (C) - Western Suburbs and Cairns (C) - Pt B
        • Toowoomba (C) - Central, Toowoomba (C) - North-East, Toowoomba (C) - North-West, Toowoomba (C) - South-East and Toowoomba (C) – West
        • Bunbury (C), Capel (S) - Pt A, Dardanup (S) - Pt A and Harvey (S) - Pt A
        • Launceston (C) - Inner, Launceston (C) - Pt B and Launceston (C) - Pt C
        • Alice Springs (T) - Charles, Alice Springs (T) - Heavitree, Alice Springs (T) - Larapinta, Alice Springs (T) - Ross and Alice Springs (T) – Stuart
        • Alawa, Anula, Bayview-Woolner, Brinkin, City - Inner, Coconut Grove, Fannie Bay, Jingili, Karama, Larrakeyah, Leanyer, Lee Point-Leanyer Swamp, Ludmilla, Malak, Marrara, Millner, Moil, Nakara, Narrows, Nightcliff, Parap, Rapid Creek, Stuart Park, The Gardens, Tiwi, Wagaman, Wanguri, Winnellie, Wulagi and City – Remainder (Darwin)
    • After these consolidations, a few UC/Ls still had a slightly larger population size compared to the SLA. However as the population differential was not large, these localities were treated as if they encompassed the entire SLA and no other. These localities were:
        • Rockhampton
        • Toowoomba (combined, see above)
        • Launceston
        • Daly River.
The ARIA and SEIFA scores were recalculated as the weighted average of the scores in the constituent SLAS, based on the relative population sizes.

Step 6: Calculation of MSOAP component and ‘balance’

For each combined SLA, the coverage rate of the constituent MSOAP locations was estimated. The MSOAP coverage was calculated by adding up the population size of the urban centre/locality for each locations within the combined SLA. If the urban centre/locality was unavailable for the locality, the state suburb population was used. If this was also unavailable, then the postcode population for the locality was used, provided it was smaller than the size of the combined SLA (minus the other UC/L or SSCs). The resulting population size was compared with the size of the combined SLA to arrive at the coverage rate for the combined SLA.

The socioeconomic information for both the MSOAP covered area and the ‘balance’ were derived based on the weighted average of the scores in the constituent populations.Top of page