Limitations
The findings in this research are subject to certain limitations due to the data sources and methodology used. Some of these limitations include:
- Estimates of health service use do not include Defence or Open Arms funded health or non-Government funded health services (such as by private insurance, through workers compensation arrangements, or the individual). As such, use of health services is likely an underestimate in this report.
- Emergency department care and public hospital admitted patient care were analysed separately. Therefore, people who presented to ED and transferred to admitted care were counted as having used both services. This should be considered when interpreting the results.
- The services that Medicare subsidises, and how services are coded has changed over time, particularly for mental health services provided by GPs and allied health professionals which can impact analysis of trends over time.
- The PBS/RPBS data does not include medicines supplied to public hospital in-patients, over the counter medicines or private prescriptions.
- There are differences between MBS services and DVA-funded MBS equivalent services. For example, DVA clients may be eligible to higher or ‘uncapped’ services than through MBS. DVA has also increased access to mental health services through various policy changes since 2001, most significantly in 2016 when eligibility was expanded to include all current and former ADF members with at least one day of continuous full-time service.
- DVA-funded MBS or MBS equivalent services were not included in the latent class analysis. Their exclusion may lead to underestimation of health service use among DVA clients.
- There are multiple limitations that apply which relate to mental health analysis:
- Community mental health care services were not included in this report. These services often treat mental health conditions in specialised community and hospital-based outpatient psychiatric services provided by state and territory governments.
- Mental health items could be miscoded or reported, for example, many GP mental health services are billed under general GP consultations (Medicare mental health services - Mental health).
- Mental health-related ED presentations refer to presentations that have a principal diagnosis that falls within the Mental and behavioural disorders chapter (Chapter 5) of ICD‑10‑AM (codes F00–F99). It should be noted that this definition does not encompass all mental health‑related presentations to ED (see Emergency departments - Mental health).
- Mental health admitted care is defined by a principal diagnosis in the Mental and behavioural disorders chapter. ‘Any mental health admitted care’ is defined by a principal or secondary diagnosis in the Mental and behavioural disorders chapter. It should be noted that this definition does not include all mental health-related admitted care (see Characteristics of ex-serving Australian Defence Force members hospitalised for suicidality and intentional self-harm).
- Analysis presented by separation reason from the ADF only contains ex-serving ADF members who separated from 1 January 2003 onwards because of changes to the way the reason for separating the ADF was recorded in 2002.
- The odds ratios describing characteristics associated with health service access are crude estimates, reflecting the effect of each characteristic individually and not adjusted for other demographic, health, or ADF service-related factors.
Aflaki K, Vigod S and Ray JG (2022). Part I: A friendly introduction to latent class analysis. J Clin Epidemiol;147:168-170. doi: 10.1016/j.jclinepi.2022.05.008.
Chitty KM, Sperandei S, Carter GL, Ali Z, Raubenheimer JE, Schaffer AL, Page A and Buckley NA (2023) Five healthcare trajectories in the year before suicide and what they tell us about opportunities for prevention: a population-level case series study. EClinicalMedicine, 63:102165. doi: 10.1016/j.eclinm.2023.102165
Den Teuling NGP, Paws SC and van den Heuvel ER (2021) A comparison of methods for clustering longitudinal data with slowly changing trends. Communications in Statistics - Simulation and Computation, 52(3): 621–648. doi:10.1080/03610918.2020.1861464
Dreiher J, Comaneshter DS, Rosenbluth Y, Battat E, Bitterman H and Cohen AD (2012) The association between continuity of care in the community and health outcomes: a population-based study. Isr J Health Policy Res, 1(1):21. doi: 10.1186/2045-4015-1-21.
Gisev N, Pearson SA, Dobbins T, Buizen L, Murphy T, Wilson A, Blyth F, Dunlop A, Larney S, Currow DC, Mattick RP and Degenhardt L (2023) Cohort profile: POPPY II - a population-based cohort examining the patterns and outcomes of prescription opioid use in New South Wales, Australia. BMJ Open, 13(5):e068310. doi: 10.1136/bmjopen-2022-068310.
Holman CD, Preen DB, Baynham NJ, Finn JC and Semmens JB (2005) A multipurpose comorbidity scoring system performed better than the Charlson index. J Clin Epidemiol, 58(10):1006-14. doi: 10.1016/j.jclinepi.2005.01.020.
Lennon H, Kelly S, Sperrin M, Buchan I, Cross AJ, Leitzmann M, Cook MB and Renehan AG (2018) Framework to construct and interpret latent class trajectory modelling. BMJ Open, 7;8(7):e020683. doi: 10.1136/bmjopen-2017-020683.
Lu Z (2024) Clustering Longitudinal Data: A Review of Methods and Software Packages, International Statistical Review. doi: 10.1111/insr.12588.
Magidson J and Vermunt JK (2002) Latent class models for clustering: A comparison with K-means. Canadian Journal of Marketing Research, 20:37–44.
Naldi L and Cazzaniga S (2020). Research Techniques Made Simple: Latent Class Analysis. J Invest Dermatol, 140(9):1676-1680.e1. doi: 10.1016/j.jid.2020.05.079.
Nguena Nguefack HL, Pagé MG, Katz J, Choinière M, Vanasse A, Dorais M, Samb OM and Lacasse A (2020) Trajectory Modelling Techniques Useful to Epidemiological Research: A Comparative Narrative Review of Approaches. Clin Epidemiol,12:1205-1222. doi: 10.2147/CLEP.S265287.
Pratt NL, Kerr M, Barratt JD, Kemp-Casey A, Kalisch Ellett LM, Ramsay E and Roughead EE (2018) The validity of the Rx-Risk Comorbidity Index using medicines mapped to the Anatomical Therapeutic Chemical (ATC) Classification System. BMJ Open, 13;8(4):e021122. doi: 10.1136/bmjopen-2017-021122.
Sinha P, Calfee CS and Delucchi KL (2021). Practitioner's Guide to Latent Class Analysis: Methodological Considerations and Common Pitfalls. Crit Care Med, 49(1):e63-e79. doi: 10.1097/CCM.0000000000004710.
Toson B, Harvey LA and Close JC (2016). New ICD-10 version of the Multipurpose Australian Comorbidity Scoring System outperformed Charlson and Elixhauser comorbidities in an older population. J Clin Epidemiol, 79:62-69. doi: 10.1016/j.jclinepi.2016.04.004.
Welsh J, Soga K, Yazidjoglou A, Douglas A, Butler D, Lancsar E, Korda R and Banks E (2023) Measuring and monitoring health systems performance using routinely collected linked data. The Australian National University.