A Resource Utilisation and Classification Study for the Residential Aged Care Sector

Stakeholder CommunicationPaper:

A resource utilisation and classification study for the residential aged care sector

October2017

Suggestion citation:

McNamee J, Gordon R, Samsa Pand Eagar K. (2017)Stakeholder CommunicationPaper - A resource utilisation and classification study to inform reform of residential aged care funding. Australian Health Services Research Institute, University of Wollongong

Table of Contents

1Purpose of this document

2The Resource Utilisation and Classification Study (RUCS)

3Background and context

3.1Fixed and variable payment models

4Design of the RUCS

4.1Study One – The service utilisation and classification development study

4.2Study Two – The fixed costs analysis study

4.3Study Three – The casemix profiling study

5Expected outcomes and benefits of the project

6The RUCS project team

7Engaging with the sector

8Participating in the RUCS

9Further information

List of figures

Figure 1 An illustration of a fixed and variable payment model

Figure 2 A branching casemix classification structure (illustrative only)

Figure 3 Conceptual approach to the three RUCS studies

Figure 4 Model of costing study design (illustrative only)

Figure 5 RUCS Project team

1Purpose of this document

The purpose of this paper is to introduce the Resource Utilisation and Classification Study (RUCS) that is being undertaken by the Australian Health Services Research Institute (AHSRI), University of Wollongong. It provides an overview of developments that have led to the study being commissioned, summarises the key issues the study is aiming to address, and provides an overview of the design of the study.

A range of strategies will be established to ensure that the sector is kept informed as the RUCS progresses, and that the project team benefits from the expertise of key stakeholders. In this context, this paper outlines the approach to stakeholder engagement that will be adopted during the study.

2The Resource Utilisation and Classification Study (RUCS)

AHSRI has been commissioned by the Commonwealth Department of Health (the Department) to undertake a study of current resource use in residential aged care facilities and to develop a system that classifies residents based on their use of these resources. The study will be undertaken between August 2017 and December 2018.

The RUCS aims to:

• Identify the drivers of cost that are related to the need for care (variable costs) and develop acasemix classification system for aged care residents.
• Distinguish between fixed costs that are associated with care of all residents (such as supervision by night staff)and costs that vary based on the care needs of individual residents.
• Develop and test the feasibility of implementing a blended funding model withfixedand variable components for residential aged care in Australia.

This project will also seek to understand whether residents use significantly moreresources for a discrete period of time when they first enter residential care. It is expected that any increased costs during this time would be related to the additional care time associated with the transition from their home to the facility and adjustment to a new environment.

3Background and context

In 2016 the Government announced it would examine long-term reforms to residential aged care funding arrangements to deliver more stable funding arrangements to both the sector and the Government.

As a first step in this process, the Department engaged AHSRI todevelop options and recommendations to help inform the design of future residential aged care funding models. AHSRI delivered its report entitled Alternative Aged Care Assessment, Classification System and Funding Modelsin February 2017.

This report found the ACFI was no longer fit for purpose because it does not adequately discriminate between residents based on their care needs. In addition, the additive nature of the ACFI is problematic as it assumes that each assessed item stands alone and that care needs are met item by item rather than in combination.

The report identified five possible funding options but ultimately recommended afixed and variable model, with the variable component determined by a branching casemix classification system.The Department has subsequently commissioned the RUCS to identify and measure the drivers of resource utilisation in residential aged care and develop and test an aged care classification system and payment model. While the RUCS is designed around the preferred option identified in the AHSRI report, it will also provide an evidence base for any future funding reform options

3.1Fixed and variable payment models

The recommended model has two main features;the separate allocation of fixedand variable payments, and the use of a branching casemix classification (for the variable component).

The fixed and variable payment structure allows the different drivers of cost to be recognised and funded in a transparent way:

• The fixed payment addresses those costs that are associated with the provision of non-individualised or ‘shared’ care, such as that provided as part of clinical supervision (e.g. quality managers and staff in clinical training or education roles) or night supervisors, whose activities concern all residents equally. These costs may differ depending on facility type, size and location.
• The variable payment covers the costs of providing individualised care tailored to the residents’ assessed needs. The ‘branching’ casemix classification will inform this funding component.
• The total subsidy that is allocated for each resident is the sum of the fixed and variable payment amounts.

An illustrative model of the fixed and variable payment structure is presented in Figure 1.

Figure 1An illustration of a fixed and variable paymentmodel

In this model, the fixed payment is the same for every resident. Although the daily fixed payment amount may differ for different types of facilities (i.e. small, large, remote, specialised), it will always be the same for each resident within a facility.

The variable payment, however, is based on the individual care needs reflected in each resident’s class in the classification described further below. In Figure 1, Resident 1 and Resident 3 have been assigned to the same class and have therefore attracted the same variable payment.

An illustrative model of a branching casemixclassificationis provided in Figure 2 using the current domains of the ACFI as the starting points for the purpose of illustration.

The key features of this type of classification are:

• the first branch in the classification is based on the resident characteristic that is the principal driver of the need for care in a residential setting, (ADLs in this case)
• the subsequent branches are added based on additional characteristics that explain significant differences in costs between members of a group (cost drivers).
• the final classes in the classification represent groups of residents with similar care needs and costs on a daily basis.A ‘cost weight’ will be determined for each of the final classes. The cost weight represents the relative costliness of one class relative to all others (i.e. a class with a weight of 2 is twice as costly as a class with a weight of 1 and half as costly as a class with a weight of 4).

Figure 2A branching casemixclassification structure (illustrative only)

In Figure 2, the residents are divided into those with high needs for assistance with activities of daily living (ADLs) and those with low or no need for ADL assistance. Each of these main branches is then further split based on whether residents also experience significant problems within other domains. Each branching split is only created where there is (resource utilisation) evidence that additional care resources are required and a classification does not have to be symmetrical in design as is the case in the Figure 2 illustration. For example, if the care resources required for a resident with high care needs in ADLs and CHCs are not affected by the presence of behavioural (BEH) issues then the final ‘BEH’ split on the left side of the tree in Figure 2 would not be necessary.

4Design of the RUCS

The RUCS will comprise three separate but closely related sub-studies. Some elements of each study will be undertaken concurrently. However each study has a distinct purpose and will produce different outcomes. Each of the studies is outlined below. The conceptual approach to the three studies is summarised in Figure 3.

Figure 3 Conceptual approach to the three RUCS studies

4.1Study One – The service utilisation and classification development study

The objectives of Study One are to determine the aged care resident factors that drive costs of individualised care and to develop a branching classification based on those findings. The classification will include the key features described in Section 3. Study One includes the core activities of data collection, financial and statistical analysis and clinical review.

A sample of 30 residential aged care facilities will participate in Study One. A cluster sampling approach will be used to select Study One sites from three geographical regions: North Queensland, the Hunter region of New South Wales and metropolitan Melbourne. To support the data collections within the Study One facilities comprehensive training and support will be provided. The facility selection process will ensure the inclusion of Aboriginal and Torres Strait Islander, Culturally and Linguistically Diverse (CALD) residents, and a mix of residents based on length of time in care.

The cluster sampling approach will provide sufficient numbers of aged care residents (approximately 2,200) with a range of care needs suitable for classification development whilst keeping the data collection activities within limited geographic areas. This approach enables support to be provided during the data collection period by staff allocated to each cluster.The three east coast regions that were identified include a mix of facility sizes and types across remote, regional and major city locations whilst also being relatively accessible for the project team to provide training, co-ordination and support.

Study One includes a number of activities that will occur in two phases as outlined below.

4.1.1Study data collection

This phase of Study One and includes three separate data collections that will occur in participating facilities. They comprise resident assessments, service utilisation and financial data collections.

The service utilisation data collection will be the most resource-intensive component of this study. It will occur over a four week period in each participating facility during early 2018. During this time, staff involved in delivering care to residents will be asked to record the amount of time spent undertaking different types of activities during each shift. Purpose-designed bar-coding technology will be provided to facilitate this data collection.

The aim will be to ensure accurate service utilisation data are collected while minimising the burden on participating facilities. Business rules are currently being developed to underpin this data collection and an extensive training program will be provided to staff in all facilities prior to the commencement of the data collection. Ongoing local and on-line support will also be provided at all times during the data collection period and feedback will be provided to facilities as required to ensure data quality.

The financial data collection will involve extracting expenditure data corresponding to the service utilisation period from the facility’s finance system. The study team will liaise with relevant finance staff at each participating site to facilitate the provision of this information. An electronic extract template will be provided that includes definitions andinstructions for the supply of this data.Additional support will be also provided to sites as required to ensure the accuracy and correct formatting of the extract.

Finally, resident assessment data will be collected at each Study One facility by qualified assessors with expertise in aged care. A suite of assessment tools that is currently being finalised will be used in this data collection. Clinical panels with expertise in key aspects of aged care such as palliative care and dementia will provide advice in the selection of assessment tools.

The selection of tools will be focussed on the drivers of the costs of care and resource consumption. Tools that used currently in the ACFI are being considered as part of this process; however, they will only be selected if they assess drivers of the cost of care. One major criticism of the ACFI is that many assessment items are captured that do not reflect the level of care and resources required and are therefore not meaningful in a funding context.

The AHSRI study team will liaise with each facility to arrange a suitable time for these assessors to attend the facility and undertake clinical assessments of all residents. This will be scheduled to occur during, or as close to the service utilisation data collection period as possible. Provision will be made for the assessment of new residents arriving in the facility during the study period and for re-assessments to be undertaken as required.

4.1.2Cost analysis and classification development

Following data collection the AHSRI team will synthesise the assessment, service utilisation and financial data collections and undertake a range of statistical analyses.

The total expenses within each facility will be separated into those that are not care related and therefore out-of-scope for this project; those related to the care of all residents (such as general supervision and care quality activities); and those related to care provided to individual residents based on their assessed care needs. This separation of expenses is illustrated in Figure 4where the darker shaded boxes represent the in-scope expenses.

The total in-scope expenses relating to individual and shared care related time are then allocated to the residents who were in care during the data collection period. The service utilisation data collected for each resident will be used as the basis for allocating costs relating to individualised care and more global allocation mechanisms will be used to allocate thecosts of shared care activities. This costing process will produce an overall cost and a cost per day for each resident.

The service utilisation and cost data will then be statistically analysed with the assessment data that was collected for each resident and relevant demographic data. This analysis will identify the assessment variables and other resident characteristics such as age, indigenous status and time in care, that are associated with different amounts of care received and resources consumed. Any increased costs associated with new residents will be identified as part of this analysis.

The variables that will be identified during this analysis as being the strongest ‘predictors’ of the cost of care provided will be included in the residential aged care classification system. The results of the data analysis will be prepared for review by a Clinical Reference Group to ensure the clinical validity of the findings. The aim of the classification development process will be to create a manageable number of classes, with the residents assigned to each class being similar in their assessed care needs and the cost of care provided to them. Cost relativities will then be developed for each of these aged care resident classes.

Figure 4 Model of costing study design (illustrative only)

Study One will have three key outputs:

1. A resident-level data set that will be used to develop a residential aged care casemix classification.
2. Service utilisation and cost data suitable for calculating the percentage of costs that are fixedand variable by facility type and relative value units for each class in the classification.
3. A cost of care profile during the initial adjustment period for each new resident in the study.

The data collection for Study One will occur between March and May 2018

4.2Study Two – The fixed costs analysis study

The purpose of Study Two is to assess the findings of Study One in relation to the distribution of fixed and variable costs. To do this, financial expenditure data covering a 12 month period will be collected from a representative sample of 110 (including approximately 30 remote and very remote)residential aged care facilities.

A stratified sampling approach will be used to select facilities Study Two.Facilities selected will be formally invited to participate by AHSRI. The data collection process and support for Study Two facilities will then be provided by StewartBrown, an accounting company that will be sub-contracted specifically to complete this task. StewartBrownhave extensive experience collecting financial data in the aged care sector and will draw on their expertise for this study. StewartBrown will work closely with AHSRI in the development of a purpose-designed template that outlines the financial data requirements.

StewartBrown will liaise with each participating facility and provide support in the preparation and submission of the data. The financial data collected will then be analysed by AHSRI to assess differences in fixed costs that are associated with characteristics of the facility rather than the individual resident. The analysis is expected to focus on identifying any differences associated with:

• Small and large facilities
• Regional, metropolitan and remote locations
• Facilities with specialist services
• Seasonal effects

The data collection for Study Two will occur between March and July 2018.

4.3Study Three – The casemix profiling study

The purpose of Study Three is to address the specific project requirement to model the financial implications of transitioning to a new funding model.Study Three has three related objectives:

1. To identify any differences in casemix by location, ownership status and size;
2. To develop a national casemix profile of residents in residential aged care facilities in Australia;
3. To apply the cost data results from Study One and Study Two to identify the financial implications of moving to a new funding model.

A stratified sampling approach will be used to select facilities for inclusion in Study Two. Again, facilities selected will be formally invited to participate and relevant information and support will be provided by AHSRI.