The identification of economically relevant health and social care services for mental disorders in the PECUNIA project

Health economic evaluations are an integral part of health technology assessments and can inform policymakers and facilitate decision-making processes about the costs and the effectiveness of different health interventions. Cost-effectiveness analyses (CEAs) take changes in costs and the effectiveness of two or more alternative interventions into account, and are consequently of particular importance for decision-making within and across countries. Therefore, the need for standardization of economic evaluations to enable proper decision-making and transferability of results has long been recognized in the health economics community [1]. However, the reporting standards do still differ substantially. Health economic research is still facing significant problems regarding the standardization and international comparability of the definitions of services [2, 3], the measurement of service utilization [4], and the monetary valuation of services [5‐7]. There is a demand for comparable health service data to improve evidence-based decision-making in health policy. The health agencies of the European Union (EU) need commensurable descriptions of care to evaluate health equity and ensure mobility of health service users in the EU [2]. Currently, the comparison of services and their contents is difficult, even if they have identical names, due to terminological variability [8]. The definition of a service is not always clear and can differ substantially. Even the term ‘service’ itself can refer to a range of elements such as the provider, a combination of functions, programs, resources, the facility of service provision or an organizational unit [2]. A service may be organized differently on an international, federal, regional, or even communal level, and it can also be challenging to define the target population of a service. There is a recognized shortcoming of international agreement on classification systems. In particular, there are no uniform coding matrices for different settings, which may lead to biased results in health services research [9, 10].

Another problem for the comparability of health economic studies is the lack of standardized resource use measurement (RUM) instruments, which may lead to unnecessary fluctuations of economic evaluation data [4, 11]. This impedes the comparability, transferability, and applicability of these studies within and across countries [12, 13]. RUMs are considered an important method for measuring the resources used by health service users within economic evaluations. Nevertheless, existing RUMs are often country-specific [14] or population-specific [15]. Hence, there is a need for harmonized RUMs to improve the overall methodology of both cost-effectiveness and comparative effectiveness studies [11, 13] in terms of international comparability. According to Mayer et al. [16], the first step in a RUM development process is the identification of economically relevant items by means of a literature review.

Likewise, there is also a substantial degree of variation in the applied valuation methods in European health economic studies [17]. For example, the micro-costing method and the gross costing method not only differ in their complexity but also their resulting estimates [5]. This lack of standardization potentially complicates the comparability and transferability of economic evaluations. The general methodological credibility of monetary valuations is hence questionable, and economic evaluations may be less useful for decision-making [18]. The underlying reason is that the applied valuation method has a significant impact on the results [19].

The EU project “ProgrammE in Costing, resource use measurement and outcome valuation for Use in multi-sectoral National and International health economic evaluAtions” (PECUNIA) aims to overcome the problems described above by developing standardized multi-sectoral, multi-national and multi-person resource use measurement instruments, standardized unit costs, and outcome assessment tools to improve the methodology of economic evaluations in the European Union [13, 20‐24]. The PECUNIA Consortium coordinated by the Medical University of Vienna consists of ten institutions situated in six European countries: Austria (AT), Germany (DE), Hungary (HU), the Netherlands (NL), Spain (ES), and the United Kingdom (UK) [25]. Three mental disorders: depression, schizophrenia and post-traumatic stress disorder (PTSD) were chosen as exemplary diseases due to their high disease burden and economic relevance. The project is exploring multiple economic sectors and domains, namely “health and social care” sectors, “education and (criminal) justice” sectors, “employment and productivity” domains, and “patient, family and informal care” domains.

The aim of this study was to identify services in the health and social care sectors that are relevant for the calculation of costs in economic evaluations from a societal perspective. According to the “whole systems approach”, the inclusion of the social care sector is pivotal in providing comprehensive mental health care. In particular, community-based social care interventions might be neglected in some healthcare systems, while they can have a significant impact on the course of a mental disorders [26]. Hence, social care services were also explicitly considered in this study to achieve a cross-sectoral broad perspective.

This study included two major steps: step I) a systematic and grey literature review for data identification, data extraction and the production of a preliminary listing of terms following initial consolidation steps (“preliminary service list”); step II) final consolidation step following external expert surveys in all six PECUNIA countries for the production of a “basic service list” used in all further steps in the PECUNIA project. These steps involved two working groups (A and B) as well as internal and external expert panels. The initial identification and consolidation steps included working group A from UKE (PH, AK, CD, HHK) and working group B from MUW (JS, CF, SM) supported by the relevant country-lead PECUNIA partners for country-specific tasks (internal expert panel: AT/MUW, DE/UKE, HU/CUB, NL/UM & EUR, ES/Psicost, UK/LSE & UnivBris). The external expert panel composed of 29 health and social service researchers, health economists, and planners from public agencies and other stakeholders in every participating country (AT: 7, DE: 3, HU: 4, NL: 5, ES: 6, UK: 4).

Step I was led by working group A and involved working group B as well as the internal expert panel. Step II was led by working group B and involved working group A, the internal expert panel and the external expert panel. The whole process and the activities performed by the working groups and the expert panels are shown in Fig. 1.

The objective of this study was to identify services relevant for persons with mental disorders in the health and social care sectors for economic analyses purposes. A provisional service list was compiled through systematic literature review and a grey literature search. Further data from existing RUM instruments were added. The preliminary service list resulting from these searches and several consolidation steps included 186 items, of which 56 were included in the final basic service list used for the development of the PECUNIA costing tools such as the PECUNIA Coding System [29], the PECUNIA RUM (Janssen LMMP, Evers I, Drost SMAA, Thorn RMWA, Hollingworth J, Noble W, et al: Involving respondents in the development process of the PECUNIA RUM instrument: Results of Think Aloud interviews, submitted) [30], the PECUNIA Reference Unit Costing (RUC) Templates [31] and the PECUNIA RUC Compendium [32, 33] https://​www.​pecunia-project.​eu/​tools).

The difference between the preliminary service list and the final basic service list resulted primarily from two reasons. First, we did not apply a specific definition of “service” when compiling the first preliminary service list. Instead, we collected all resource items related to health and social services use, in order to be as comprehensive as possible. As a result, the preliminary service list was a mix of not only services but also interventions, procedures or professions that had to be excluded subsequently. Second, for the further typology of services, the PECUNIA Consortium agreed to define a service according to the international DESDE classification system as adapted for the PECUNIA project. This allowed achieving a higher consolidation level where different items could be combined into one internationally harmonized service item for costing purposes. For example, the service “outpatient healthcare service” includes all types of outpatient physicians and non-physician healthcare providers.

This study allowed the realization of the magnitude of the problem when “words” are used to describe services for health economic research, because often “names” of services were prone to Babylonian confusion not allowing the comparison like with like. Whilst, for example, it was very clear for the national project partners and contact points what a simple service use like “mental hospital stay” meant in their respective country and/or region, a detailed analysis in terms of cost measurement for economic evaluations revealed that mental hospital stays may mean very different things in different countries. Affected dimensions were for example eligible target populations (regarding age group and gender, diagnosis, disease severity or functional impairment), treatments offered, average duration, setting (with and/or without overnight stay), and degree of specialization (mental ward versus closed mental ward). This problem has several implications. First, it points to a general limitation of comparisons of cost-of-illness studies and cost-effectiveness analyses when identically named services, do in fact not represent identical but rather similar compositions of utilized resources. Second, the description of services or utilized units in cost-of-illness studies and cost-effectiveness analysis can be questioned. Overall, studies rarely presented detailed descriptions of the units measured to calculate costs. We assume that this information is often regarded as unnecessary for evaluation purposes because the definition of services seems to be clear in the national context. While this might to be acceptable from a national perspective, it becomes a substantial problem, if one tries to compare the units of analysis internationally. However, if one tries to describe a service, the inherent question arises of how best to do this. Using words can introduce new problems by creating circular references. For these reasons, it was decided in the PECUNIA Consortium to use the DESDE coding system for the description of services.

As expected, most of the identified services were located in residential and outpatient facilities. Hence, currently these types of services seem to play a pivotal role in mental healthcare provision in Europe. Conversely, services in the remaining segments of care (day care, accessibility to care, information to care, self-help and voluntary care) are very much underrepresented. This might change to some extent due to care transitions induced by the COVID-19 pandemic like increased use of video-consultations that outlast the end of the pandemic. Nevertheless, we identified a large number of services beyond “medical care” for example services providing protected living, assistance in different areas of daily living or the workplace.

We found an extensive variety of services provided to persons with mental disorders in Europe. It became apparent that the distinction between a service and a procedure is not always clear-cut. Hence, these discrepancies in the conceptualization of a service potentially impede their comparability in economic evaluations. In this study, economically relevant services for mental health and social care were identified and organized according to the different “main types of care” of the DESDE classification system. As a result, a basic list of services for mental healthcare provision in Europe emerged. The resulting service list provided the basis for a typology coding system of services for internationally comparable units of analysis, the development of a harmonized RUM instruments, and corresponding reference unit costs for international mental health economic studies within the PECUNIA project [30‐33] (https://​www.​pecunia-project.​eu/​tools).