Appendix A: Data quality information

This appendix includes data quality summaries and additional detailed information relevant to interpretation of the information in this report.

States and territories are primarily responsible for the quality of the data they provide. However, the AIHW undertakes extensive validations on receipt of data, checking for valid values, logical consistency and historical consistency. Where possible, data in individual data sets are checked with data from other sets. Potential errors are queried with jurisdictions, and corrections and resubmissions may be made in response to these queries. Except as noted the AIHW does not adjust data to account for possible data errors or missing or incorrect values.

Complete data quality statements for the National Non-admitted Patient Care (aggregate) Database (NNAPC(agg)D) and the National Non-admitted Patient (episode-level) Database (NNAP(el)D) are available online in the AIHWs online metadata registry METeOR.

National Non-admitted Patient Care (aggregate) Database

The NNAPC(agg)D is based on data provided for the:

  • Non-admitted patient care National Minimum Data Set (NAPC NMDS, METeOR identifier 649281)
  • Non-admitted patient care local hospital network National Best Endeavours Data Set (NAPCLHN NBEDS, METeOR identifier 649576).

It holds aggregated clinic-level data on the type of outpatient clinic, counts of individual and group service events, the funding source for the service events, and whether the service involved care from multiple health care providers.

The reference period for this data set is 2017–18. The data set includes records for non-admitted patient service events provided between 1 July 2017 and 30 June 2018.

Non-admitted patient care NMDS

For 2017–18, the scope of the NAPC NMDS was non-admitted patient service events provided by public hospitals. It also included public hospital services that were funded through the jurisdictional health authority or Local Hospital Network.

Non-admitted patient care Local Hospital Network aggregate NBEDS

For 2017–18, the scope of the NAPCLHN NBEDS (METeOR identifier 649576) is non-admitted patient service events provided by:

  • local hospital networks (LHNs)
  • other public hospital services that are managed by a state or territory health authority and are included in the General list of in-scope public hospital services, developed under the National Health Reform Agreement (COAG 2011).

Summary of key issues

  • For 2017–18, the NNAPC(agg)D included data for most public hospitals that provided non-admitted patient care in Australia. In addition, non-admitted patient care information based on data provided for the NAPCLHN NBEDS was reported for:
    • the state health authority, 15 LHNs and 2 other services in New South Wales
    • the state health authorities in Victoria and Queensland
    • the state health authority and 5 LHNs in Western Australia
    • the state health authority and 4 LHNs in South Australia.
  • The NNAPC(agg)D data have limitations. For example, there is variation in admission practices (AIHW 2017) between states and territories and there is variation in the types of services provided for non-admitted patients in a hospital setting.
  • Due to changes between 2013–14 and 2017–18 in the scope, definitions and counting rules, time series information has not been included in this report and these data are not comparable with data presented in earlier reports.
  • Victoria did not collect information for Diagnostic services that could not be linked to services provided at another clinic. All diagnostic services for non-admitted patients that could be linked to a service provided at another clinic were reported as part of the service event provided in that clinic (medical consultation, procedural or allied health and/or clinical nurse specialist intervention clinic). Therefore, the national counts of service events for Diagnostic services are likely to be underestimated.
  • Queensland did not report any non-admitted patient service events with a funding source of Medicare Benefits Scheme. MBS-funded non-admitted patient activity for doctors practising right of private practice were not included in the data provided to the NNAPC(agg)D, as these patients are not considered by Queensland to be patients of the hospital.
  • Western Australia did not report specialist mental health service events for the 2017–18 NAPC NMDS and the NAPCLHN NBEDS for the clinics: 20.45 Psychiatry, 20.50 Psychogeriatric, 40.34 Specialist mental health and 40.37 Psychogeriatric. Western Australia was also not able to provide data for the clinics: 30.02 Magnetic resonance imaging30.03 Computerised tomography (CT), 30.06 Positron emission tomography and 30.07 Mammography screening. Therefore, national counts of service events for these clinics are likely to be underestimated.

Differences in scope between 2013–14 and 2017–18

Between 2013–14 and 2014–15, the scope of the NAPC NMDS changed—from a focus on activity-based funded hospitals to all public hospitals, resulting in increases in the number of hospitals and other services reporting for the NNAPC(agg)D between 2013–14 and 2014–15.  Table A1, available to download in the Data section illustrates the changes is coverage for the data provided for the NNAP(agg)D between 2013–14 and 2017–18.

For 2014–15 to 2017–18, information was also provided for non-admitted patient service events at the LHN-level, at state/territory health authority-level, for other public hospital services and by some private hospitals providing public patient non-admitted patient services under contract.

National Non-admitted Patient (episode-level) Database

The NNAP(el)D is based on the Non-admitted patient National Best Endeavours Data Set (NAP NBEDS, METeOR identifier 650086).

It holds episode-level data including selected patient characteristics; the type of outpatient clinic; whether the episode was an individual or a group service event; the source of the request for service; the service delivery setting; the service delivery mode, the type of care provided, whether the service involved care from multiple health-care providers and the funding source for the service event.

The reference period for this data set is 2017–18. The data set includes records for non‑admitted patient service events provided between 1 July 2017 and 30 June 2018.

In 2017–18, the scope of the NAP NBEDS was defined as non-admitted patient service events in:

  • public hospitals
  • Local Hospital Networks
  • other public hospital services that are managed by a state or territory health authority and are included in the General list of in-scope public hospital services, which have been developed under the National Health Reform Agreement (COAG 2011).

For the NNAP(el)D, a record is included for each service event, not for each patient, so patients who receive more than one non-admitted patient service event in the year have more than one record in the NNAP(el)D.

Summary of key issues

  • For 2017–18, episode-level non-admitted patient service events data reported to the NNAP(el)D was estimated as 72% of the non-admitted patient service events provided for the NNAPC(agg)D. The proportion varied among jurisdictions, ranging from 62% for New South Wales to 100% for the Australian Capital Territory
  • For Victoria, episode-level non-admitted patient service events are underreported due to data collection issues, and are not comparable over time.
  • The NNAP(el)D data have limitations, For example, there is variation among states and territories in admission practices (AIHW 2017) and also in the types of services provided for non-admitted patients in a hospital setting.
  • The sex of the patient was not reported for 144,000 service events (0.5% of records).
  • The patient’s date of birth was not reported for 8,000 service events (0.03% of records) and therefore the age of the patient could not be determined.
  • The Indigenous status of the patient was not reported for 1.4 million service events (5% of records). In addition, the quality of the data reported for Indigenous status in non-admitted patient settings has not been formally assessed, so caution should be used when interpreting these data.
  • The patient’s area of usual residence was not reported for 4% of service events, and therefore, the patient’s remoteness area and socioeconomic status group could not be determined for those records. 
  • Service request source was not reported for 15.7 million service events.

Fluctuating coverage between 2013–14 and 2017–18

Before 2015–16, the scope of the NAP NBEDS was defined as non-admitted patient service events in activity-based funded hospitals only. Between 2014–15 and 2015–16, the scope of the NAP NBEDS changed to include public hospitals and other services that were not activity-based funded. Therefore, changes in the numbers of service events reported between 2013–14 and 2017–18 should be treated with caution.

Table A2, available to download in the Data section, illustrates the changes is coverage for the data provided for the NNAP(el)D between 2013–14 and 2017–18.

Other factors affecting the interpretation of non-admitted patient care data

Changes to the collection of non-admitted patient care over time mean that these data are not comparable over time.

Before 2013–14, information on non-admitted patient care was reported using different clinic categories and counting units.

Time series information between 2013–14 and 2017–18 is not presented due to changes that affect the interpretation of these data, including:

  • the scope of the NAPC NMDS
  • the scope of the NAP NBEDS 
  • the coverage the NAPCLHN NBEDS and the NAP NBEDS
  • the definitions and counting rules for non-admitted patient care.

Differences in definitions of non-admitted patient care

For the NAPC NMDS, the NAPCLHN NBEDS and the NAP NBEDS, a non-admitted patient service event that involves multiple health professionals (and related diagnostic services) within the same clinic is counted as one service event. If a patient attends more than one clinic on the same day, then each attendance is counted as a separate service event.

In AIHW reports for the 2012–13 financial year and earlier, non-admitted patient occasions of service were counted as the number of services provided to a patient in each functional unit of a health service establishment. Each diagnostic test or simultaneous set of related diagnostic tests for a patient were counted as a separate occasion of service.

Therefore, the data presented for non-admitted patient service events in this report are not comparable with data reported for non-admitted patient occasions of service in reports for the 2012–13 reference year and earlier periods.

In addition, the activity of the clinic 40.01 Aboriginal and Torres Strait Islander people’s health clinic (which was in-scope for 2013–14 and 2014–15) was out-of-scope for 2015–16 to 2017–18.

Differences in counting rules for non-admitted patient care

In 2013–14 and 2014–15, for the NAPC NMDS, the NAPCLHN NBEDS and the NAP DSS/NBEDS, each session of renal dialysis, total parenteral and enteral nutrition, and ventilation performed by the patient in their own home was counted as a non-admitted patient service event.

For 2015–16 to 2017–18, the counting rules for some home-delivered non-admitted patient services changed to ‘temporal care bundling’. Temporal care bundling means that all non-admitted patient sessions performed per month are ‘bundled’ and counted as one non-admitted patient service event per patient per calendar month regardless of the number of sessions (IHPA 2016).

This resulted in a marked decrease in reporting of non-admitted patient services events in total, and for Procedural clinics, and for the following Tier 2 clinics:

  • 10.15 Renal dialysis–haemodialysis–home delivered
  • 10.16 Renal dialysis–peritoneal dialysis–home delivered
  • 10.17 Total parenteral nutrition–home delivered
  • 10.18 Enteral nutrition–home delivered
  • 10.19 Ventilation–home delivered. 

Information no longer collected

Between 1993–94 and 2013–14, the AIHW reported aggregated non-admitted patient occasions of service data from the National Public Hospital Establishments Database (NPHED), which covered a wider range of non-admitted patient care than is collected for the NNAPC(agg)D and NNAP(el)D. From 2014–15 onwards, information has not been available for:

  • Emergency occasions of service provided by hospitals that do not have a designated emergency department.
  • Information on emergency presentations provided by hospitals that have a designated emergency department were reported in Emergency department care 2017–18: Australian hospital statistics
  • Pharmacy occasions of service
  • most Pathology and Radiology and organ imaging services occasions of service—as these are considered ‘related diagnostic services’ connected with other service events and are not reported separately for the NNAPC(agg)D and NNAP(el)D
  • most occasions of service for Community health services—although some community health services are in scope for the NNAPC(agg)D and NNAP(el)D.

References

AIHW 2017. Variation in hospital admission policies and practices: Australian hospital statistics. Health services series no. 79. Cat. no. HSE 193. Canberra: AIHW.

COAG (Council of Australian Governments) 2011. National Health Reform Agreement. Viewed 5 June 2017. Available to download from the COAG website.

IHPA (Independent Hospital Pricing Authority) 2016. Tier 2 Non-admitted services definitions manual 2016–17, October 2015 V4.1. Sydney: IHPA. Viewed 16 April 2019. Available to view  on the IHPA website.