Variation in direct healthcare costs to the health system by residents living in long-term care facilities: a Registry of Senior Australians study

Introduction

Globally, most countries are experiencing significant growth in the proportion of older people in their populations.¹ This demographic shift is resulting in significantly higher government spending on aged (also referred to as social care) and healthcare services required to meet the growing number of older people and their care needs.² As in many Organisation for Economic Cooperation and Development countries,³ the Australian aged and healthcare sectors are multi-billion dollar industries.⁴ In Australia, both sectors are largely funded by the federal government through taxation revenue.⁴ In 2021–22, the Australian government expenditure on aged care was $25 billion and healthcare was $98 billion,⁴ with these costs projected to increase significantly over the next 10 years (projected average annual growth of 3.33%, all costs in $A).⁵

Aging has a direct effect on healthcare expenditure. Reorienting aged care systems to effectively and efficiently deliver high-quality aged care, may improve health and quality of life outcomes for older people and subsequently alleviate downstream healthcare costs.^6,7 Older people accessing long-term care facilities (LTCFs) (i.e. where both medical and personal support services are provided to people who are unable to live independently in the community) typically are frail, have complex physical and mental health issues and consequently are also major healthcare service users.^6–10 For example, older people entering LTCFs in 2015 had a median of 10 medicines dispensed, two-thirds had ≥five comorbid conditions and half had a high frailty index score.¹¹ It is also known that one in five residents have unplanned hospitalisations and emergency presentations within 90 days of entering LTCFs.¹²

The aged care and healthcare sectors are inextricably linked, both in Australia and internationally, and yet at least in Australia these sectors remain largely structurally and financially separated.^4,13 However, both the healthcare and aged care sectors are jointly responsible for the health and wellbeing of older Australians. Policy and practice in one sector directly impact the other. To our best knowledge, the average costs of healthcare incurred by individuals in LTCFs in Australia, which can inform our understanding of the downstream costs to the health system associated with the experiences older people have in care, has not been evaluated before.

Our study aimed to estimate the government-subsidised healthcare costs incurred by older people living in LTCFs over a 12-month follow-up period and assess the variation in these costs by specific resident and facility characteristics.

Methods

Study cohort and setting

People aged 65 years and older and living in LTCFs across three Australian states (New South Wales, Victoria and Queensland) on 1 January 2016 were included. Residents who identified as Aboriginal or Torres Strait Islanders, were Department of Veteran Affairs’ card holders or had an activity in private hospitals within the follow-up period were excluded from the study (Fig. 1, for cohort selection). The cohort was followed up for a year (31 December 2016) to assess their direct healthcare costs.

Fig. 1.

Flow chart showing cohort selection. *Provides how the cohort has been selected for the analysis for clarity and aids to replicability of the study.

Results

A total of 75,142 (median age 83, IQR 78–88) unique residents living in 2112 LTCFs were studied. Of these, 52,814 (70.3%) were women, 40,137 (53.4%) had a diagnosis of dementia, 48,992 (65.2%) were born in Australia, 52,627 (70%) were living in LTCFs located in major cities and 41,997 (55.9%) were living in not-for-profit LTCFs (Table 1). Of the total, 20,266 (27.0%) died during the year follow-up period (Table 2). Of the LTCFs studied, 1587 (75.1%) had at least 20 or more residents (Table 2).

The adjusted average annual healthcare cost per LTCF resident was $9233 (95% CI: $9150–$9295) in 2016. Residents without dementia incurred higher average healthcare costs ($11,097, 95% CI: $10,995–$11,200) than those with dementia ($7561, 95% CI: $7502–$7620). Residents living in for-profit LTCFs incurred higher healthcare costs ($11,324, 95% CI: $11,185–$11,463) than residents in not-for-profit ($11,017, 95% CI: $10,895–$11,139), while those in government LTCFs had the lowest healthcare costs ($9731, 95% CI: $9365–$10,099). The residents living in LTCFs outside major cities ($8314, 95% CI: $8213–$8414) incurred similar healthcare costs to those in major cities ($8233, 95% CI: $8153–$8314). Residents living in larger LTCFs (≥20 residents) incurred higher adjusted annual healthcare costs ($9324, 95% CI: $9290–$9357) in comparison to smaller facilities (<20 residents) ($8895, 95% CI: $8771–$9019) (Table 2). Supplementary Table S1 shows the GLM estimates of the total healthcare costs as a function of individual and facility variables.

Overall, hospitalisations accounted for the largest proportion of costs (47.2%) followed by MBS (27.2%). ED presentations represented the lowest proportion of healthcare costs (6.0%) (Fig. 2, Supplementary Table S2). Similar healthcare resource cost component breakdown was observed by dementia status, facility ownership, rurality and size (Fig. 2, Supplementary Table S2).

Fig. 2.

Healthcare resource use as a proportion (average) of all healthcare costs in 2016 A$, overall and by individual and facility characteristics.

Of the 1587 LTCFs with ≥20 residents, residents living in 561 (35.3%) incurred costs below the 95% confidence interval upper limit and 537 (33.8%) below the lower limit (Fig. 3). Similar variation by residents’ (Supplementary Fig. S1) and LTCFs characteristics were also observed (Supplementary Figs S2 and S3).

Fig. 3.

Funnel plot of annual adjusted healthcare (total) costs per long-term care facility resident in 2016.

Discussion

Individuals in LTCFs incurred an average of $9233 in public healthcare costs per resident with an estimated total healthcare cost of $694 million for the study cohort in 2016. Significant variation in these costs was observed among the over 2100 LTCFs studied across Australia’s three largest states, with over one-third above the average annual healthcare costs. While key individual characteristics were considered in our analysis, the variation in healthcare costs observed are likely attributed to the varying experiences and care provided to older people in LTCFs including care models (e.g. healthcare integration, types of services provided in facility, reliance on external healthcare) or quality of care provided.

Individuals living in government run LTCFs incurred the lowest average healthcare costs ($9731) compared to those in not-for-profit ($11,107) and for-profit ($11,324) LTCFs. Individuals living in government LTCFs, which in Australia are mostly managed by state health departments, may access healthcare services differently than other ownership types, and are reported to have lower hospitalisations in some cases, which may have led to lower downstream healthcare costs.²¹ Previous research has also highlighted that residents in not-for-profit LTCFs experienced higher comfort and better quality of care than those in for-profit LTCFs,^22,23 which might have positively influenced residents’ health outcomes and again downstream interaction with the health system. A prior study has also identified that in comparison to government LTCFs, residents living in for-profit or not-for-profit LTCFs are up to twice as likely to suffer serious injuries that require a hospital visit, hospitalisations for falls or hospitalisations with reported pressure injury, all incidents that would accrue significant costs.²¹ When compared by facility size, our study found that residents living in smaller facilities incurred lower healthcare costs than those in larger facilities. This aligns with prior evidence that suggests smaller facilities tend to deliver higher quality care, resulting in better quality of life²⁴ and residents experiencing fewer hospitalisations when compared to larger facilities, which are the events influencing healthcare costs of individuals.²⁵

As previously reported,^26,27 the healthcare cost for LTCF residents with dementia were approximately $3500 less a year than those without dementia. In Australia, LTCF residents with dementia are often placed in a dedicated managed care environment, which may contribute to successfully lowering the hospitalisation rates, and therefore costs, experienced by these individuals.^26,28 However, while it is possible that residents with dementia have been successfully kept away from hospitals,²⁹ it is also possible that these residents have better physical health compared to others in LTCFs and may not have the same type of intensive clinical needs that others may have required. The possibility that these individuals are not adequately accessing the healthcare services that they need is also likely. There is evidence that suggests residents with dementia are less likely to complain about certain ailments (e.g. pain)³⁰ and access certain healthcare services only.^31,32

Innovative care models (e.g. smaller scale LTCFs and clustered domestic models)²⁵ and quality improvement strategies in LTCFs have shown to be effective in reducing use of health services, including through reducing potentially preventable hospitalisations.^25,33,34 With major differences in average costs for people aged ≥65 years of public hospital stay per day (~$1700 in 2019–20)³⁵ and LTCF resident costs per day ($243) in Australia,³⁶ government investment in aged care models that support residents’ high quality of care and better integration between the aged and healthcare sectors will likely affect the outcomes of individuals and reduce downstream healthcare costs.²⁵ Furthermore, studies have reported that older people, mostly still in the community or on a national waiting list to access aged care at home in Australia, have higher annual average healthcare costs than those living in the LTCFs we are reporting.^5,37 Given that almost half of the healthcare costs are attributed to hospitalisations in our study, LTCFs are likely offering some benefits to keeping older people out of hospital leading to lower overall health costs than when they are in the community. However, the variation in healthcare costs across the LTCFs reported in this study also suggests that practices and outcomes regarding health care are different among LTCFs and opportunities for improvement are still possible.

The study has several limitations. The study relied on linked administrative and assessment data that were collected for purposes other than research and its incompleteness and the potential for incorrect linkage (<1% of records had no linkage records), miscoding or misreporting of events is possible. However, we attempted to minimise these issues by conducting several between datasets logic checks to ensure that anything missing was non-differential between the groups studied. Albeit our results were adjusted for important individual and facility level characteristics, there are likely residual confounding that can affect our estimates, some of which may be driven by individual and facility characteristics not captured by the data used in this study. Our estimates were also not adjusted for important domains such as care quality and delivery of person-centred care, which are not captured in our linked datasets and have also been identified as drivers of downstream healthcare costs.^33,34 Furthermore, our study only analysed public healthcare costs and not aged care costs, which are currently between $73,166 and $82,316 per resident per year, and the out-of-pocket costs for the residents themselves.²⁶

In conclusion, our study demonstrated a wide variation in annual healthcare costs accrued by LTCF residents, with hospitalisation contributing to almost half of the costs. Understanding sectoral variation and learning from LTCFs where residents are experiencing lower hospitalisations, is a potential strategy to reduce healthcare costs accrued by LTCF residents. This requires cross-sectoral collaboration and efficiency maximisation strategies.

Supplementary material

Supplementary material is available online.