How should we measure the use of social care and informal care?

KEYWORDS:

• Costs
• economic evaluation
• informal care
• resource-use measurement
• social care

1. Introduction

Meaningful economic evaluation relies on the accurate assessment of not only patient-relevant outcomes but also resource use affected by a choice of intervention. Economic evaluations conducted from the societal perspective, which is standard in some jurisdictions (e.g. The Netherlands [1] and the US [2]) require all relevant societal costs and benefits to be assessed, including social care and informal care, alongside healthcare resources. Formally defining social care is challenging [3] but, broadly, it describes care that is intended to help people to live independently, and can include help with bathing or eating, for example. The definition of informal care is also complex [4], but was described as ‘care provided by a relative, friend or the community in an unpaid role’ by a public advisory group convened by the authors.

Typically, healthcare resource use is well characterized in economic evaluation and relatively easy to measure, with a range of measurement methods at the analyst’s disposal. These include patient self-report and the use of routinely collected data in mature data collection systems such as the Hospital Episode Statistics dataset in England or the National Patient Register in Sweden. When it comes to measurement of social care, however, the picture is far less clear. In the UK, routinely collected social care datasets exist, but in fragmented independent sources rather than a coherent national format, which makes them challenging to use in economic evaluation. Furthermore, little work has been conducted to ascertain their fitness for research purposes, considering that the data are collected in order to run the services effectively, and are not designed to be used for research.

Informal care is indelibly linked to social care; where social care services are lacking, informal care may be substituted. The inclusion of informal care can also change the cost-effectiveness outcomes of a study [5]. For informal care, there has been much written about the nature of the caregiving burden [6]. However, less attention has been paid to measurement of the informal care costs for economic evaluation, with a systematic review by Grosse et al finding a lack of consensus arising from both the difficulty in distinguishing caregiving activities from everyday activities and the lack of an agreed definition leading to variability in what is collected in economic evaluations [7]. Furthermore, there is no possibility of using routine datasets that capture the full extent of caring activities and the economic impact of informal care, and participant self-report will therefore remain necessary.

2. Challenges with patient self-report

Assuming that health economists will need to rely on self-reported data for the foreseeable future, the relevant questions to decide how we should measure informal care and social care are (1) who we should ask, (2) the format in which we should ask them and (3) what we should ask them about.

In terms of who we should ask, where possible that is likely to be the patient or service user themselves for social care services. However, where the participant lacks capacity, or is unable to supply information themselves, it may be that a carer is asked to supply information as a proxy for the participant. For informal care, there are multiple measurement challenges and we do not currently know whether carers or care recipients are able to supply better data [8]. To facilitate completion by the study participant themselves wherever possible, instruments should be accessible in a range of formats (such as pictorial or screen reader-friendly versions).

For all resource-use measurement using participant-reported methods in economic evaluations, it is necessary for the analyst to weigh up the balance between requesting more data to refine the estimates against the need to avoid overwhelming participants. Too substantial a burden of completion could potentially result in poorer completion rates. Diaries or logs completed regularly over the period of time in question may help participants to record their resource use [9]. The items to request information on should be determined by a systematic and transparent approach. For informal care, appropriate items will depend on the planned valuation method (e.g. opportunity cost or proxy good method [10]).

Social care and informal care are always going to present measurement challenges for health economists. Social care systems are complex; in the UK, for example, a combination of state funding, personal payments and third sector contributions may all play a part, and the landscape changes rapidly, with services renamed and/or reconfigured. Informal carers do not always identify themselves as being carers, and the people they care for do not necessarily see it as a caring relationship, particularly in the case of spouses.

3. Participant self-report options

Earlier self-report initiatives included the Client Service Receipt Inventory (CSRI) [11], which was developed in the 1980s for interview completion in the context of mental health conditions, and included questions about living situation, social security benefits and informal care (amongst others). A cursory search in Google Scholar indicates that the instrument is still well used (145 results returned for 2023 alone), but typically the instrument used is described as a ‘modified CSRI;’ these modifications have not been systematically tracked and the validation status is mostly unknown. Other patient-reported resource-use measures in the past have suffered from a lack of attention; old instruments are recycled regardless of their performance, and have rarely been subjected to any validation [12]. However, recent initiatives have aimed to overcome some of the issues previously seen, with transparent descriptions of the development process, and version control.

A standardized 13-item questionnaire that could be used to measure both informal care costs and carer productivity losses was developed by Landfeldt et al [10], as a response to the lack of inclusion of informal care in economic evaluations. The Caregiver Indirect and Informal Care Cost Assessment Questionnaire (CIIQ) is designed to be completed by the carer (who may be a study participant themselves) rather than the person for whom they care.

The PECUNIA project (ProgrammE in Costing, resource use measurement and outcome valuation for Use in multi-sectoral National and International health economic evaluAtions) aimed to develop a standardized, harmonized and validated resource-use measurement tool for use in economic evaluations across Europe [13]. Alongside modules covering health care and other sectors, comprehensive social care and informal care modules were developed. Cost drivers were identified through a literature review, and consolidated by expert surveys, before being developed into questionnaire items, and preliminary validation was undertaken [14]. The PECUNIA RUM is now available for use in economic evaluations [13].

A different approach has been taken in the development of ModRUM (Modular Resource-Use Measure), which is designed to be concise rather than comprehensive, while still being flexible enough for use in real studies in the UK. Alongside an existing healthcare module [15], social care and informal care modules are currently under development and will shortly be available for use. Items were identified by a literature review and prioritized in focus groups involving both health economists and members of the public with experience of social care (either as carers or recipients of care). Testing in the form of in-depth interviews with health economists and cognitive interviews with people with experience of social care has been undertaken, and the instrument has been piloted with members of the public.

4. Expert opinion

So, going forward, how should we measure informal care and social care? To a certain extent, the answer lies in pragmatism; for informal care, self-report will always be necessary, and it is likely that self-report will remain dominant for social care utilization data for some time to come, too. Instruments that have undergone thorough testing now exist to make the task an easier one. Analysts can select from a range of approaches including a fully-featured instrument, such as the PECUNIA RUM, or a concise minimized instrument such as ModRUM, or an instrument for carers to complete such as CIIQ. Using such instruments should increase data quality and comparability across studies.

It will also be vital to start assessing the validity and usability of routinely collected social care datasets, and to share information and good practice. The PRIMORANT study has produced guidance on the decision-making process for the use of healthcare datasets, which may prove useful for assessing social care datasets as well [16]. The measurement methods that we now have available will need to be assessed in real-world studies, and development will need to be ongoing and iterative to ensure that we capture everything of importance in a changing landscape. Other instruments that are used should be tested for validity, and any development processes should be rigorously described.

While much about the measurement of social care and informal care remains uncertain, one thing is clear. With health and social care systems increasingly seeking to take an integrated approach [17], this needs to be reflected in economic evaluation, and it is increasingly important to do more than make a token attempt to include social care and informal care when resource allocation decisions should be based on data that are equally well collected for all relevant sectors. Ultimately, it is time that we gave social and informal care equal weighting with healthcare when it comes to inclusion in economic evaluations conducted from the societal perspective.

Declaration of interests

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.