In December 2024, Statistics Netherlands (CBS) published the revised figures for health and social care expenditure (hereafter: Health and Social Care Accounts) for the years 2021 to 2023, including two new tables containing a breakdown of expenditure by care type. In principle, the revision of the Health and Social Care Accounts takes place every five years, at the same time as the revision of the National Accounts. This publication describes the main changes arising from this revision and provides a guide for interpreting the figures.

The Health and Social Care Accounts present national health and social care expenditure broken down by groups of healthcare providers and financing schemes. The Health and Social Care Accounts also provide the basis for the figures on expenditure on healthcare in accordance with the internationally agreed definition, with a breakdown by healthcare provider, financing scheme and function based on international standards (System of Health Accounts). 

The most important innovation within this revision is the addition of the breakdown by care type. Previously, the figures were only broken down by healthcare provider and financing scheme. However, many healthcare providers supply more than one type of care, a fact which remains entirely obscured in the intersection between healthcare providers and financing schemes. For example: hospitals’ revenues under the Healthcare Insurance Act (‘Zorgverzekeringswet’) are not only spent on medical specialist care, but also on patient transport, medicaments, dental care, perinatal care, etc. The creation of wide-ranging care conglomerates has contributed to the diversity of provision. Breaking down the figures by care type brings that diversity into focus.

The revision has also prompted an update to the presentation on StatLine. A clearer distinction has been made between companies within the Health and Social Care section and those outside it, such as pharmacists and secondary providers of healthcare. The terminology has also been amended; the term ‘care expenditure in the broad sense’ has been replaced by the term ‘health and social care expenditure’ in order to better express the fact that it includes more than merely healthcare in the narrow sense. 

Finally, the revision has also led to a more complete coverage of care expenditure, a better demarcation of care versus non-care and a better breakdown of care expenditure by financing scheme.

All of the changes have been enabled by the use of new data sources and improved use of existing sources. The most important new source takes the form of claims submitted under the Healthcare Insurance Act (Zvw) and the Long-term care Act (’Wet langdurige zorg’; Wlz) per healthcare provider, supplied by Vektis. Without this source, the breakdown by healthcare type and the improvement of the breakdown by financing would not have been possible. 

An important impulse for the revision was the fact that since the 2021 reporting year, CBS has been required to provide detailed statistics on revenues and costs for all components of the Human Health and Social Work Activities section (SBI divisions 86, 87, 88) as part of the Structural Business Statistics (SBS). Among other things, this has resulted in a much more intensive use of fiscal data and public annual reports of healthcare providers. As a result, data are now available for components of the Human Health and Social Work Activities section for which there were previously little or no data available.

Other data sources used for the first time include the government’s open data on government subsidies, data from the Central Bureau for Fundraising (CBF) on charities, data from the National Health Care Institute on personal contributions under the Healthcare Insurance Act and the Vektis data management (AGB) register. 

In principle, the CBS statistics on health and social care expenditure (hereafter: Health and Social Care Accounts) are revised every five years, at the same time as the National Accounts. The base year for this revision is 2021. When a revision is carried out, any errors in the figures are corrected and efforts are made to improve quality through the use of new data sources and/or the application of new insights. In this revision, the output of the statistics, in the form of StatLine tables, has also been expanded. This publication not only describes the approach and results of the revision, but seeks to help users in interpreting and using the figures. The definitions and demarcations used, the way in which missing data is handled, the correlation between the tables and the strengths and weaknesses of the statistics are explained. The level of detail within the underlying calculation system used to create the StatLine tables is also explained, for the benefit of researchers who are authorised to access the CBS microdata. 

First of all (section 2), the shortcomings of the statistics before revision are discussed. Then, in section 3, the new sources used are presented, along with a description of the improved utilisation of existing sources. In section 4, the new classification of healthcare providers is explained, while in section 5, the new breakdown by care type, the basis of two new StatLine tables, is presented. After discussing the approach to preventing double counting in section 6, section 7 presents the effects of the revision on the 2021 figures. Section 8 looks in more detail at how the figures for the international demarcation of healthcare expenditure (System of Health Accounts) are compiled and describes the improvements generated by the revision. Finally, the development of care expenditure over 2022-2023 is discussed, followed by a brief preview.

The statistics which CBS publishes annually on health and social care expenditure are stand-alone statistics. The Health and Social Care Accounts do not correspond exactly with the part of the National Accounts that relate to care. However, the Health and Social Care Accounts and National Accounts figures have been aligned as closely as possible. The figures on care expenditure published in the Health and Social Care Accounts are far more detailed than those in the National Accounts. Another important difference is that the National Accounts cover only that portion of total care that falls within SBI divisions 86-88 (Human Health and Social Work Activities section), whereas the Health and Social Care Accounts describe all care consumption in the Netherlands, i.e. including care produced outside that section. The bulk of the latter is made up of medicaments and medical goods, but it also includes patient transportation by taxi firms, home help supplied by companies whose main activity is not care, care consumed abroad, care delivered within the Ministry of Defence and the Ministry of Justice and Security, etc. The linking of claims registers to the CBS business register has revealed that care is supplied from unsuspected sectors of the economy. 

The Health and Social Care Accounts are structured in accordance with the methodology of the System of Health Accounts (SHA). In the SHA, care expenditure is broken down by three axes: provider, financing and function. The standard output consists of the three two-dimensional tables generated by intersecting the three axes (i.e. provider*function, provider*financing  and function*financing ). CBS also publishes these three tables for the internationally defined demarcation of ‘Health Expenditure’. The classifications used for providers, financing and function have been agreed internationally (OECD-Eurostat-WHO) and are therefore a compromise aimed at achieving international comparability. 

The Health and Social Care Accounts differ from the SHA on a few points:

1. The Health and Social Care Accounts feature a much broader demarcation of ‘expenditure on care’: the SHA relates only to healthcare, whereas the Health and Social Care Accounts also include social care (including children's day care, social services). Furthermore, the Health and Social Care Accounts comprise the totality of long-term care, whereas in the SHA, social care, such as home care financed under the Social Support Act (‘Wet maatschappelijke ondersteuning’, Wmo) is only included as a health-related memorandum item. In the international demarcation, that care is not counted as ‘health care’ but as ‘social care’. R&D expenditure and training are also not counted as expenditure according to the SHA, but they are included in the Health and Social Care Accounts. 
2. The classification of provider categories differs on several points, see annex 2.
3. For that part of the care that falls outside the SHA definitions, no breakdown by care type was provided in the statistics before the revision. 

Traditionally, the ‘Providers*Financing ’ table has been central to the Health and Social Care Accounts. The main shortcoming of this table is that it does not reflect the fact that the provider categories in the table are based on a classification by principal activity, whereas providers often supply multiple forms of care. However, the names of these categories, such as ‘Providers of ... care’ create the impression that they limit themselves to one type of care. For instance, ‘General medical practices’ provide not only GP care but also – albeit to a limited extent – medicaments. Similarly, alongside medical specialist care, ‘Providers of medical specialist care’ also provide other services, such as mental health care, ambulance care, nursing home care and medicaments. A particularly striking example is the fact that while the providers of nursing and care primarily provide elderly care, they also provide maternity care. This is because maternity care is often provided by home care institutions which principally provide elderly care. In order to address this point, two new tables have been developed: ‘Providers*Care types ’ and ‘Financing*Care types ’. This effectively completes the alignment with the methodology of the System of Health Accounts by adding a third axis (care type) to the two axes already published (provider category and financing scheme). 

Another limitation of the “Providers*Financing” table was that prior to the revision, the classification of the providers in categories was not seamlessly aligned with the Standard Business Classification (SBI), which is used for virtually all economic statistics within CBS. This made proper comparison of the Health and Social Care Accounts with other economic statistics impossible. This problem arises primarily because a significant proportion of the care is provided by companies that do not belong to the Human Health and Social Work Activities section. Examples include cleaning firms that offer home help, fitness centres that provide physiotherapy, government agencies such as the Ministry of Defence and the Ministry of Justice and Security that provide their own care, beauty salons that provide cosmetic care, educational consultancies that offer youth care, etc. In the figures before revision, this care was often grouped under the type of provider the company in question most closely resembled. For instance, care provided by cleaning firms under the Social Support Act (Wmo) was assigned to the ‘Providers of nursing and care’ category. With this revision, we have moved to a classification based on providers from the Human Health and Social Work Activities section on the one hand and providers from the ‘rest of the economy’ on the other. Moreover, within the latter category, we draw a distinction between providers who supply care as their principal activity (pharmacies, suppliers of medical goods) and providers who supply care as a secondary activity.

A third limitation of the Health and Social Care Accounts before revision was that the Human Health and Social Work Activities section (SBI divisions 86, 87, 88) was not fully covered. For instance, the ‘Psychologists’ and ‘Other paramedical’ subclasses and the ‘Other providers’ categories in divisions 86 and 88 were not fully covered, due to the lack of reliable figures. This issue is particularly relevant to components of the Human Health and Social Work Activities section where a relatively large amount of uninsured care is provided.

Finally, developments in recent years have revealed the need for a stronger focus on the prevention of double counting which can occur when goods/services are supplied by one healthcare provider to another, or through the use of self-employed persons by healthcare providers and as a result of financing constructions such as integrated care and medical specialist companies. A phenomenon which has gained momentum in recent years and has attracted a lot of attention as a result is the growth in the number of self-employed persons working in the Human Health and Social Work Activities section, which increased from 88,000 in 2016 to 134,000 in 2022, a rise of slightly over 50%.  And the ‘Cost of agency staff and other temporary contract workers’ of the major care institutions (hospitals, care for the disabled, mental health services, youth care, shelter care and nursing and care) rose from €2.0 billion in 2016 to €4.8 billion in 2022. Here, too, there is a risk of double counting: the revenues of these self-employed persons are also included in the revenues of the companies that hire them and submit claims for the care provided to the insurance companies. 

With the revision, an improved methodology has been introduced in order to identify and correct the different forms of double counting as far as possible.

The revision described in this publication has been made possible by the availability of new data sources and improved utilisation of existing sources. A summary of the most important sources and their impact is given below:

1. Linking Zvw and Wlz claims to the business register

At the request of CBS, Vektis has aggregated the claims available to them (at the individual and provider level) to amounts per delivery code (‘prestatiecode’) and per provider. By making use of the Vektis AGB register, these amounts have been linked to the CBS business register. This makes it possible to determine the total amount of each care type claimed per group of healthcare providers, classified in accordance with the SBI. It also allows a calculation of the volume of insured care provided by companies that provide care as a secondary activity (outside the Human Health and Social Work Activities section). The care type is determined by grouping delivery codes (for both Zvw and Wlz), and by also making use of indications (data on care needs) from the Care Needs Assessment Centre (CIZ) for Wlz care.

2. Linking CAK figures on personal contributions to the healthcare providers that supplied the care

The Central Administration Office (CAK) data provide a detailed insight into the personal contributions that individuals pay for both Wlz care and Wmo residential care. For Wlz care, these contributions are linked to the care delivered per person and then assigned to the care provider that submits a claim for that care. For Wmo residential care, CBS does not have data for care claims, but it does have figures for total personal contributions per person and AGB code. By linking the AGB register to the business register, this personal contribution for Wmo residential care can be assigned to healthcare providers. These detailed data create a more accurate picture of the financing flows.

3. Linking information from different sources on personal budgets (cash benefits)

It was already possible to link personal budgets to healthcare providers with the information on personal budgets held by the Social Insurance Bank (SVB). That existing linkage has been improved. By combining that data with the indications from the CIZ, it is now also possible to break down the personal budgets by care type up to a certain point.

4. More intensive use of fiscal data and public annual reports

Since the 2021 reporting year, prompted by the European requirement relating to Structural Business Statistics (SBS), CBS has begun compiling economic statistics for virtually the entire Human Health and Social Work Activities section (SBI divisions 86, 87, 88). Previously, there were no comprehensive data available on revenues and costs for all components within the Human Health and Social Work Activities section. In order to produce economic statistics for those components, CBS is now making intensive use of fiscal data, supplemented by information from public annual reports (especially for foundations). This has meant that data have become available for components of the Human Health and Social Work Activities which were lacking before the revision, in particular: a large part of the ‘Other paramedical practitioners’ subclass (86919), part of the ‘Practices of psychologists’ subclass (86913), part of the ‘Preventative health care’ subclass (86923), the ‘Umbrella organisations in the field of health care and other support activities for health care’ subclass (86929), the ‘Counselling and welfare for disabled’ subclass (88103) and the ‘Community centres, other consultancy and cooperative bodies in the field of social care’ subclass (88999).

5. Use of open data from the national government

Following the revision, CBS is making use of open databases of the national government, which contain data on subsidies, contributions to other public authorities, income deductions and financing to agencies. Because many of these files contain the Chamber of Commerce registration numbers of the recipients, it is possible to link these amounts to the CBS business register and so allocate them to the correct provider category. This information is then used to improve the breakdown of care expenditure by financing source.

6. Wider use of DigiMV annual reports 

Annual reports filed by care institutions that fall under the Care Institutions (Accreditation) Act (WTZi), the Child and Youth Act (‘Jeugdwet’) and, since 2022, under the Care Providers (Accreditation) Act (WTZa) were already used before the revision. Before the revision, the focus was on units from the CBS Care Institutions statistics,  i.e.: University Medical Centres (UMCs), general hospitals, specialist hospitals, residential mental health care, care for the disabled, residential youth care, shelter care, ambulatory youth care, and residential nursing care and home care. After the revision, these data sources have also been used for healthcare providers outside those sections as far as possible, primarily in order to obtain more information about financing flows under the Wmo and Child and Youth Act. This is because the supplementary CBS survey which is part of the DigiMV reporting requirement contains the revenues per Act of Parliament (Zvw, Wlz, Wmo and Child and Youth Act).

7. Identification of subcontractors by means of extensive source linking

Linking various sources to the CBS business register, such as Zvw and Wlz claims, personal budgets (cash benefits), the BIG register (healthcare professions), the SKP register (Quality Register of Paramedics), the AGB register and the population of Respondents for Youth Policy Information (‘Berichtgevers voor Beleidsinformatie Jeugd’), has produced improved insight into which companies (potentially) function as subcontractors. This extensive utilisation of sources helps to more accurately catalogue the complexity of the care chain and improves the extent to which corrections can be made for (potential) double counting in the Health and Social Care Accounts.

The most important elements of the revision are explained in more detail below:

• the new classification of Health and Social Care providers
• the new breakdown by care type
• the approach to preventing double counting

After this, the other changes resulting from the revision are described and the figures for 2021, 2022 and 2023 are presented.

Thanks to the revision, there is now a clearer distinction between companies within the Human Health and Social Work Activities section and companies outside it, such as pharmacists and secondary healthcare providers. In addition, Human Health and Social Work Activities section (SBI divisions 86-88) is now described in full in the Health and Social Care Accounts. 

The Health and Social Care Accounts system used to produce the figures works with actors (categories of healthcare providers). After the revision, this system comprises 73 actors (annex 1), which are grouped in the 22 categories presented in the StatLine tables. The new classification is shown in the table below.

Some of the changes warrant further explanation:

1. Before the revision, independent medical specialists and dental specialists working in general hospitals were not included in the expenditure on general hospitals but in the expenditure on other suppliers of medical specialist care. This had come about for historic reasons, going back to a time when those independent medical specialists submitted claims to the insurance companies themselves. Now only the hospital submits claims and the independent specialists are, as it were, ‘subcontractors’. By including them with general hospitals, the approach in the Health and Social Care Accounts has now been brought into line with the approach in the Care Institutions statistics. Following the revision, independent medical specialists who work outside hospitals are still included with other providers of medical specialist care; independent dental specialists who work outside hospitals have been merged with dental practices.
2. The new classification contains a new Households category. This is for families that provide care via a personal budget (cash benefit; PGB). Before the revision, those households were classified under the healthcare providers providing the relevant type of care. For example: households that offer care for the disabled were grouped with companies offering care for the disabled. With the revision, which brings the Health and Social Care Accounts more closely into line with the SBI classification of providers, households are now classified separately, because they are not companies. For the time being, one exception has been made to this: households that provide children's day care continue to be subsumed under companies that provide children's day care, because it is difficult to properly establish the scale of the provision by households. The intention is to resolve this issue by the time of the next revision.
3. The pre-revision category ‘Other providers of healthcare’ has been scrapped, with the relevant actors being moved to categories that better reflect the nature of the care they provide.
4. It has been found that the ‘Umbrella organisations in the field of health care and other support activities for health care’ and ‘Community centres, other consultancy and cooperative bodies in the field of social care’ actors (subclasses 86929 and 88999, respectively), which before the revision had been included in the ‘Policy and management’ group, include many companies that provide care. The former belief that these two categories provide very little, if any, care themselves, instead of concerning themselves primarily with the organisation and financing of care, does not correspond with reality. ‘Umbrella organisations in the field of health care’ have therefore been moved to the new ‘Providers of other healthcare services’ category, while ‘Other welfare and consultancy’ has been moved to the ‘Providers of social work and other welfare’ category.
5. A separate category has been introduced for foreign health and social care providers. Although this category is limited in terms of size and the figures are subject to some uncertainty, it has been decided to make it separately visible because CBS is required to supply this category to Eurostat, which also publishes it.
6. There is an important difference in our approach to healthcare providers within Human Health and Social Work Activities section and providers outside it. For healthcare providers within the Human Health and Social Work Activities section, in principle all revenues are counted as care expenditure, unless it is known that they also provide non-care services and the value of those services cannot be determined. For providers of care outside the Human Health and Social Work Activities section, only the care expenditure revealed by claims (Zvw, Wlz, supplementary insurance schemes) or other sources, such as annual reports, is counted. This approach entails the risk of underestimating care supplied by providers outside the Human Health and Social Work Activities section. For example: a company that provides Wmo care but does not explicitly state this in its annual report – companies that only provide Wmo care are not subject to compulsory corporate social responsibility reporting under the WTZa – may remain unrecorded.
   The same is true of companies that provide uninsured care, including a significant number that provide cosmetic care. Analyses by CBS reveal that a proportion of those firms fall outside the Human Health and Social Work Activities section and are classified within the Beauty Care sector. This makes it difficult to fully reflect this care in the Health and Social Care Accounts, particularly if the companies in question do not explicitly report on their care activities.
7. The names of a few categories have been changed, not because the content has changed but because the new names better describe them:
   a. ‘Providers of nursing, care and maternity care’ instead of ‘Suppliers of nursing and care’;
   b. ‘Providers of prevention and public healthcare’ instead of ‘Providers of preventive care’;
   c. ‘Providers of social work and other welfare’ instead of ‘Providers of social work’.
   d. ‘Providers of medical appliances’ instead of ‘Suppliers of therapeutic devices’
8. The pre-revision category ‘Suppliers of medicaments’ has been divided up due to the new approach: a separate ‘Pharmacies’ category has been created, and the sale of medicinal products by supermarkets and drugstores now falls under the ‘Providers of care and welfare as a secondary activity’ category.
   The latter change may appear to be a backward step, given the fact that part of the expenditure on medicaments is now ‘hidden’ within a much broader category, but the opposite is true: because, since the revision, CBS is also publishing the provider*care type intersection, with medicaments being one of the care types, the information on expenditure on medicaments is now actually published in more detailed form than it was before the revision. The breakdown by care type is the subject of the next section.

Following the revision, CBS is for the first time publishing a detailed breakdown of expenditure by care type. Care types are combinations of activities which are homogenous in terms of content and/or target group, such as home care, shelter care, medical specialist care, care for the mentally disabled and mental health care. This new care type classification establishes a third dimension, alongside the existing dimensions of healthcare providers and financing schemes. This gives users a better insight into which types of care are supplied by which providers and how they are financed. 

Zvw and Wlz claims are an essential source for the new care type classification in healthcare and long-term care. Vektis has also provided CBS with claims amounts per provider for care financed under supplementary insurance schemes. However, there is an important limitation: dental care and physiotherapy claims in particular tend to be submitted by factoring companies instead of by the healthcare providers themselves. This means that those claims cannot be linked to the CBS business register. Claims submitted by factoring companies are grouped proportionally with claims which have the same delivery codes and are linked to the business register. The same method is used for other claims which cannot be directly linked to the business register, for example due to an incorrect AGB code. 

CBS does not (yet) have access to iWmo and iJeugd claims, but hopes to be able to make use of these in the future. As a result, opportunities for breaking down the figures for care financed under the Child and Youth Act and the Wmo by care type are currently limited. The same is even more true of social welfare activities, for which our sources (fiscal data, CBS surveys) contain little or no information about care types. Sometimes, the care type is necessarily assigned on the basis of the classification of the providers: providers of children's day care (subclass 8891) are assumed to only provide children's day care; providers of occupational health & safety support and reintegration (subclass 86923) are assumed to focus exclusively on those activities. This means that if those providers do also provide other forms of care, it will be incorrectly allocated to the ‘principal care type’. Such assumptions are only made for provider categories which, as far as we are aware, provide few if any care types other than those corresponding to their classification.

5.1 Identifying care types

The classification by care type was developed in consultation with various experts and relevant organisations from the section. The care type classification is hybrid: in some cases, the classification is based on target groups (such as care for the mentally disabled), whereas in other cases it is determined by the nature of that care (for example, in the case of palliative care). 

The availability of data from different financing sources largely determines the level of detail to which the care types can be broken down. For example, it is currently not possible to further subdivide mental health care into more specific care types such as addiction care because the Care Performance Model (introduced in 2022) no longer explicitly differentiates within the data. This represents a potential risk for the continuity of the new tables and published care types.

Moreover, medicaments which are part of a broader treatment are not recorded separately as expenditure for the ‘medicaments’ care type. Similarly, preventive care which forms part of a treatment is not explicitly recognised and recorded as such.

In the system used to generate the figures, 62 care subtypes are currently used. The care subtypes are grouped according to the 26 care types listed in the StatLine tables, which are shown in the table below. Annex 2 contains a list of all 62 care subtypes and their allocation to care types.

In defining these 26 care types, CBS has sought to strike a good balance between their value in monetary terms, the wishes of the consulted users and the quality of the data. It is not desirable to have very small care types, given that they will be intersected with providers and financing. Aspects of the quality of the data include the extent to which all care of the relevant type is covered and whether the amounts can be accurately allocated to the correct provider categories. Based on the latter requirement, the figures for Wmo home care and youth care, for example, have not been broken down. Although it is generally known how much money goes to specific groups of providers under the Wmo or Child and Youth Act, insight into the care types within these categories is lacking. This means that, although a good estimate is available for national expenditure on, for example, home care and support, it cannot be allocated to specific provider categories. Both forms of care have therefore been allocated to the ‘Nursing and care at home’ care type. 

For a number of care types, somewhat different criteria have been used, as explained below. 

5.1.1 Care type(s): Medical specialist care

CBS and the experts it has approached have discussed at length how to break down the very large figure for expenditure on medical specialist care (after separating off expensive medicinal products, see previous point). A two-stage approach has been chosen:
Step 1: medical specialist care is a very broad field that partially overlaps with some more specific types of care. As far as possible, those overlaps have been removed by isolating the relevant parts of medical specialist care and allocating them to the more specific care types. Specifically:

1. Population screening (i.e. diagnostics following a positive screening test, for example internal intestinal examination after a positive stool test for colorectal cancer screening);
2. First line diagnostics and other laboratory activities;
3. Perinatal care;
4. Geriatric rehabilitation;
5. Other rehabilitation;
6. Palliative care;
7. Oral and maxillofacial surgery;
8. Psychiatry;
9. Radiotherapy (as a care subtype).

Step 2: By isolating these care types, around €3.8 billion of the total medical specialist care costs can be separated off for 2021. The remaining share of medical specialist care is still very large and heterogenous. That part is therefore differentiated further on the basis of the ‘highest’ setting of the care provided (inpatient, day treatment, outpatient). The setting is inferred from the care activities (such as nursing days) delivered within a Diagnosis Treatment Combination (DTC), in some cases from the treating specialism (for some specialisms, such as audiology, it is assumed that only outpatient care takes place) and, for DTCs for which no setting can be inferred from care activities and specialism, an imputation. DTC subprogrammes that remain without a setting are allocated to the outpatient setting. This classification by setting is also used in the individual statistics on medical specialist care.

For ‘Other Care Products’, no setting is inferred. That care is therefore allocated to the ‘Other Medical Specialist Care’ care type, along with radiotherapy. The same applies to the portion of the expenditure on medical specialist care that is not reimbursed under basic or supplementary insurance.

5.1.2 Care type: Mental health care

The care type ‘Mental health care’ covers treatment and assistance care for people with psychiatric disorders and psychological problems, ranging from short-term help with mild symptoms to long-term, intensive treatments (and inpatient care) for serious conditions. It also includes addiction treatment and care delivered by healthcare assistants attached to GP practices for mental health care.

The following sources are used for the classification of ‘Mental health care’:

• Claims under the Healthcare Insurance Act, Long-term care Act (Wlz) and supplementary insurance, based on delivery codes:
• Delivery code list for ‘GP care’: POH-GGZ
• Delivery code list for ‘Curative mental health care’
• Delivery code list for ‘Medical specialist care’ (psychiatry specialism)
• Wlz deliveries which belong to the GGZ-B care profile

• Expenditure on Wmo care at mental health care institutions
• Expenditure on psychological assistance care (uninsured)
• Expenditure on forensic care

5.1.3 Care type: Preventive care

The ‘Preventive care’ care type covers care and activities focused on preventing disease and health issues. This includes vaccinations, screening programmes, health information provision and lifestyle interventions. It also includes specific preventive services such as dental check-ups, oral hygiene, dietary advice, youth healthcare and occupational health & safety services. Preventive care which is part of a broader treatment is not included in this category, because it is not explicitly recorded as such. This implies that the figure recorded for expenditure on prevention is far from complete. It has nevertheless been decided to publish this figure, because it is informative, even if it only relates to earmarked preventive care.

Various sources have been used for the classification of the care type ‘Preventive care’, for example:

• Claims under the Healthcare Insurance Act and supplementary insurance, based on delivery codes:
• Delivery code list for ‘GP care’: vaccinations, smear test, stopping smoking, fall risk assessment
• Delivery code list for ‘Paramedical’: fall prevention and fall risk assessment
• Delivery code list for ‘Dental care’: consultation and diagnosis, making and evaluating images and preventive care
• Delivery code list for ‘Medical specialist care’: follow-up on population screenings.
• Expenditure on care by dieticians, oral hygienists and occupational health & safety services.
• Government spending, such as the national vaccination programme and municipal health services (‘GGD’).
• Population screening financed by the national government.

5.1.4 Care type: Palliative care

The ‘Palliative care’ care type comprises care aimed at improving the quality of life for patients with incurable diseases. This includes services such as pain management, symptom management, psychological support and guidance with end-of-life decisions. As with ‘Prevention’, palliative care that is part of a broader treatment is not recorded separately.

The classification of ‘Palliative care’ is based on claims under the Healthcare Insurance Act and the Long-term care Act (Wlz):

• Delivery code list for ‘GP care’: euthanasia, euthanasia consultation, palliative consultations.
• Delivery code list for ‘Primary residential care’: palliative terminal care and residential care in the final phase of life.
• Delivery code list for ‘District nurse services’: palliative terminal care.
• Delivery code list for ‘Medical specialist care’: palliative care and support and symptom relief.
• Wlz services which fall under the VV10 care profile.

5.1.5 Care type: Sheltered accommodation

The care type ‘Sheltered accommodation’ comprises care combined with accommodation for individuals with psychological or psychosocial problems who cannot live independently (temporarily or otherwise), with support for their daily activities, personal development and recovery. This is not the same as assisted living.

The ‘Sheltered accommodation’ classification is made up of the following components:

• Sheltered accommodation under the Wmo: this is based on the total expenditure on Wmo tailored provisions for ‘Residential care and shelter’, as reported in the Municipal Social Domain Monitor (GMSD). The distribution between ‘Sheltered accommodation’ and ‘Other shelter care’ is estimated using information from the municipal fund’s May circular. The Wmo expenditure on ‘Sheltered accommodation’ is proportionally allocated to the provider categories based on the out-of-pocket payments they receive for Wmo residential care, for which the AGB code is known.
• Sheltered accommodation under the Long-term care Act (Wlz): this is based on claims for residential mental health care under the Wlz.

5.1.6 Care type: Medicaments

The ‘Medicaments’ care type comprises the provision of medicinal products and consumables (such as bandages) for the prevention, treatment or management of diseases and symptoms. This includes both medicinal products available on prescription and over-the-counter medicaments (available without a prescription). The so-called expensive medicinal products (care subtype) provided as part of medical specialist care also fall under this care type. What does not fall under this care type are the ‘ordinary’ medicinal products provided as part of a hospital treatment or residential care in a nursing home. The reason for this is that their cost cannot be distilled from our data. Also not included under this care type and in fact completely absent from the Health and Social Care Accounts are medicinal products sold without a prescription on the internet.

5.1.7 Care types: Integrated care

This care type is an odd one out because the demarcation is based not on the content of care but on the method of financing. Integrated care is multidisciplinary coordinated care for chronic conditions (diabetes type 2, COPD, cardiovascular risk management), provided under the Healthcare Insurance Act, with different care providers and organisations working together to integrate the care for a patient. 

It is made up of GP care, physiotherapy, medical specialist care, etc. Ideally, integrated care would be broken down further, but it is not possible to distribute the amount spent on integrated care between underlying care types. For this reason, it has been decided to include integrated care as a separate care type. In the table showing the intersection of care types with provider categories, this care type is listed under the provider that submits the claim rather than the providers that deliver the care.

5.1.8 Care type: Shelter care

The ‘Shelter care’ care type comprises care and support for persons temporarily without accommodation or in a crisis situation – e.g. homeless shelters, women's shelters, asylum seekers shelters and other provisions.

5.1.9 Care types: Training and R&D

The ‘Training’ care type comprises only expenditure for further medical training under the Zvw and Wlz. Because this training takes place ‘on the job’ at care institutions, the costs are included in the Health and Social Care Accounts. The ‘R&D’ care type comprises only R&D carried out by UMCs.

When calculating health and social care expenditure, CBS takes the revenues of care providers from supplying care as its starting point. A drawback of this approach is that instances of double counting can arise when a care provider outsources part of the care to another (possibly independent) care provider. In such cases, the revenues of both the principal provider and the subcontractor may be added together, which can result in the same expenditure being recorded twice, and therefore to an overestimate of total care expenditure. 

Further complicating factors are the fact that the principal provider and the subcontractor do not always fall within the same subclass of section Q, and that a care provider can be a principal provider and a subcontractor at the same time. Generally speaking, it is not possible to infer relationships between the principal provider and subcontractors from the available sources, and hard figures about the scale of subcontracting are lacking for much of the care sector. For this reason, it is often necessary to use indirect evidence of subcontracting. In order to better estimate the scale of possible double counting, specific methodologies have been applied. The most significant examples of double counting and methods of correction are explained below.

6.1 Examples of double counting 

1. Consolidation within group structures

In their annual corporate social responsibility reports (DigiMV), some care providers issue both a consolidated statement at group level and a statement per underlying unit. CBS maintains a Care Units Register, in which a record is kept of the chamber of commerce numbers (and CBS business units) associated with a conglomerate, in order that we do not count the income from the separate financial statements twice. This has allowed us to effectively prevent this form of potential double counting. (This was already the case before this revision.)

2. Procurement of healthcare and services from other healthcare providers and enterprises

Many healthcare providers purchase supplementary services from healthcare and non-healthcare providers alike, such as laboratory testing or cleaning services. In the case of procurement of care from independent medical laboratories (subclass 86924), which are monitored by means of a separate CBS survey, double counting would occur if no correction were made for these services between providers in the case of laboratories or hospitals (or other procuring healthcare providers). This is because the amount the hospital receives for providing care already includes the fees it charges for laboratory testing. For the procurement of non-care services, such as cleaning or financial administration, no correction is necessary because the income from these providers is not monitored and therefore there is no double counting. Figures for subcontracting and outsourcing often make no distinction between procurement of care services and non-care services.

3. Financing schemes in which the care provider does not submit claims itself.

Within the care sector, it is increasingly common to find examples of care financing in which the party that claims back the costs of care is not the party that provides the care. For instance, multidisciplinary care (also known as integrated care) is claimed by one party, which is then responsible for payment to all the care providers that provided part of the care. The claiming party usually falls within the SBI subclass ‘Umbrella organisations in the field of health care and other support activities for health care’ (86929), while the care may be provided by independent GPs and physiotherapists who fall within their ‘own’ subclasses. The claims are counted as care expenditure and a correction needs to be made on the side of those GPs and physiotherapists in order to prevent double counting. Another example are care farms. Many care farms work together in partnerships. For example, a partnership may act as the contracting partner for local authorities for the delivery of Wmo care on behalf of all the participating care farms. These partnerships are often classified in SBI subclass 86929, as referred to above, whereas the care farms are primarily classified in SBI subclass ‘Residential and day care for mental retardation’ (8720). Other examples of a risk of double counting associated with the method of financing are integrated perinatal care and patient transport.

4. Use of self-employed persons

As stated in the introduction, the number of self-employed persons in the care sector has risen sharply in recent years. Self-employed persons often fall into a different SBI category than the care institution that hires them in. The revenue of the self-employed person is included in the care claimed by the care institution. In order to prevent double counting, we would need to know whether each individual self-employed person is working on behalf of a care institution or not, plus the monetary value of the care provided. After all, self-employed persons can also deliver and invoice care directly to clients.

6.2 Methodology for correcting double counting

The manner in which CBS corrects for double counting depends on the available data sources and the financing structure. Cataloguing all the principal contractor-subcontractor relationships within the care sector would be impossible; the use of one or more generic methods is essential. 

1. In SBI subclasses for which an Annual Social Responsibility Document (DigiMV) is compulsory: use of specific variables about subcontracting

For several years now, hospitals and institutions for residential mental healthcare, care for the disabled, shelter care, residential youth care, outpatient youth care and nursing, caregiving and home care have been required to state in their annual accountability statements  how much revenue they receive from subcontracting and how much they pay out to subcontractors. The quality of the figures is not yet of a standard that would allow the tracking of year-on-year developments. However, they are used to make a good estimate of the revenues from subcontracting and are deducted from total operating returns in order to prevent double counting. That does not yet solve the problem of self-employed persons hired in by care providers. That potential double counting is addressed on the side of the self-employed persons – see below under point 4.

2. Providers of GP care and dental care: preventing double counting by determining the level of care expenditure from the available financing sources 

These SBI subclasses contain thousands of small providers of care, some of which are part of a chain and some of which are not. There are locums, partnerships, chains that submit claims for the participating healthcare providers, chains that only provide services to participating healthcare providers, financing via financial controllers (integrated care, GP services structures). For these categories, the problem of double counting is avoided by not calculating the care expenditure as the sum of the income of the providers but by adding up how much money is spent from the different financing sources. That is to say: an estimate of the amounts patients themselves pay is added to the claims per provider (Zvw and supplementary insurance). While making such an estimate is difficult, it yields a great benefit in terms of preventing the double counting problem. This approach is also used for health centres (subclass 86921) and medical laboratories (subclass 86924).

3. Health care providers surveyed by CBS: Cost of agency staff and other temporary contract workers

For healthcare providers surveyed by CBS, such as children's day care centres, local and social work and occupational health and safety services, the variable ‘Cost of agency staff and other temporary contract workers’ is used. This item is broken down into the cost of agency staff on the one hand and the cost of other temporary contract workers on the other, on the basis of figures from the Care and Welfare Labour Market (AZW) project on numbers of agency staff and self-employed persons hired in. The estimated figure for other temporary contract workers is then used to correct for instances of double counting. This therefore means the correction takes place on the side of the hirer of care staff. This is not a particularly robust method, due to the assumptions which are needed to break down the costs of agency staff and other temporary contract workers, and because it is not known whether those temporary contract workers are actually engaged in care activities and, if they are, which SBI subclass the self-employed persons in question belong in. The implicit assumption is that they fall within the same SBI subclass as the care provider hiring them. 

4. Identifying subcontractors (including self-employed persons) who provide insured health care 

Whether companies and self-employed persons that provide care which is typically insured are classified as ‘double counters’ or not is based on the following considerations: if those companies do not submit claims themselves, have not submitted an annual corporate social responsibility report, do not provide care financed from personal budgets (cash benefits) and are not listed in the AGB register, they are regarded as double counters. For self-employed persons in a medical profession, such as nurses, an even stricter criterion is used: if they do not submit claims themselves, this is taken to mean that they are hired in on a long-term basis and therefore should be regarded as double counters. 

In the domain covered by the Child and Youth Act and Wmo, AGB registration is a virtual requirement in order to be able to submit claims to municipalities, so in this case the assumption is that if a self-employed person or (small) company is not listed in the AGB register, it is a double counter. However, if a self-employed person or company is listed in the AGB register, the revenue in question is counted. This approach may not be strict enough: there will be providers of Wmo care who work for the party which has a contract with the municipality but who are nevertheless AGB-registered. Although we use a great many sources to support our conclusion, it remains a crude method. A refinement of the method has been introduced for the aforementioned care farms: the care farms often submit claims related to personal budgets (cash benefits) themselves and in many cases are also AGB-registered, but they earn the remainder of their income as subcontractors of a party that concludes contracts with municipalities. In such cases, only the personal budget-related revenue of the care farms themselves is counted.

This method does not work for SBI subclasses that provide a lot of uninsured health care or social work. For this reason, there is a lack of instruments to identify double counters within the ‘Other paramedical practitioners’ subclass (excluding the BIG occupations) and the ‘Psychologists’ subclass (excluding those who are BIG-registered), and two subclasses comprising ‘Other providers of care’ (86929 and 88999). These subclasses consist of many thousands of sole proprietorships for which only a name is known that reveals little. As a result, it is not possible to determine whether they have actually delivered care (see the text box on the ‘Human Health and Social Work Activities’ section <link to text box in chapter 2> ). For this reason, there is no correction for double counting in these subclasses. Although this can lead to an overestimate of care expenditure, there are few indications that subcontracting takes place here to any great extent. 

This section discusses the effects of the revision for the reporting year 2021 in more detail. Following the revision, the figure for health and social care expenditure in 2021 is up by €4.3 billion, to a total of €129.1 billion. 

The fact that the figure for total care expenditure is higher after the revision is due to a number of factors, in particular the expanded coverage of the statistics and the improved approach to double counting (section 6). In addition, there have been major shifts between provider categories, as described in section 4. The figures per provider category in table 7.1 are the result of all of these factors. The different factors influence one another, so that no definitive figure can be assigned to each factor. 

Broadly speaking, the expanded coverage of the statistics, i.e. the addition of parts of the Human Health and Social Work Activities section that had not been included before revision, comprises the following (shown to the nearest €50 million):

• Non-insured care in the ‘Psychologists’ SBI subclass (86913): +€800 million
• Paramedical practices: +€1000 million. This increase is primarily a consequence of the improved coverage of the SBI subclass ‘Other paramedical practitioners’ (86919). Before the revision, only a few specific occupations from this subclass were included – speech therapists, podiatrists, remedial therapists, oral hygienists, occupational therapists and dieticians – and often not completely because data was only available for the insured part of the care provided. Thanks to an improved typology of companies in this subclass and the use of fiscal data, a more complete picture has now been obtained of the occupations listed above. This subclass also contains many other occupations such as dental prosthetists, opticians, audiologists, orthoptists, reflexologists, GP assistants and healthcare assistants, nurses, childbirth coaches and skin care therapists. Before the revision, these occupations were absent from the figures; following the revision, the entire subclass is included. Alternative healers are also included in this subclass, but in the Health and Social Care Accounts they are included under Providers of other healthcare services. With the help of sources including claims details, the BIG and SKP registers and a manual classification of companies based on their trading or legal names, it has been attempted to classify as many companies as possible, in order to achieve the most accurate possible breakdown by care type. For approximately 5000 companies, with combined revenues of more than €200 million, it has not been possible to achieve a classification. It is to be expected that this group will partly consist of providers of alternative therapies. In addition, some of these 5000 unclassified companies, self-employed persons and small firms may be self-employed persons and small firms hired in by other care providers, for example in the case of GP assistants and healthcare assistants. As a result, it is possible that the care expenditure in these categories is somewhat overestimated.
• Providers of other healthcare services: +€650 million. This is the sum of two changes. In the first place, the estimated value of alternative healers has nearly doubled, to slightly over €500 million. Before the revision, the expenditure on alternative healers was estimated on the basis of usage data (number of clients, consultations and rates). Following the revision, the expenditure is estimated on the basis of the classification of companies (in the SBI subclass ‘Other paramedical practitioners’, 86919). This classification is obtained by scanning company names for the presence of terms that indicate alternative therapies. In view of the large number of unclassified providers in the ‘Other paramedical practitioners’ subclass (see previous point), the current figure may still be an underestimate. In the second place, following the revision, the entire SBI subclass ‘Umbrella organisations in the field of health care and other support activities for health care’ (86929) is included, whereas before the revision only that part that was engaged in the organisation and financing of health care was estimated. This has resulted in an increase of approximately €400 million. In this subclass, too, an attempt has been made to classify as many units as possible manually, but this has not proved possible for approximately 80% of the companies. Although the units in question are on the smaller side, they represent a combined revenue of approximately €300 million. This unknown expenditure has therefore been assigned to the ‘Other health care’ type.
• Providers of care for the disabled: +€500 million. The increase is the result of adding the ‘Counselling and welfare for disabled’ SBI subclass (88103), which was erroneously not included in the Health and Social Care Accounts before the revision. This relates primarily to the so-called MEE organisations. However, in the overall figures, the ‘Providers of care for the disabled’ category has remained virtually unchanged. This is because personal care budget provided by family (cash benefit) has now been moved to the ‘Households’ category.
• Providers of social work and other social care: +€250 million. The estimate for the ‘Community centres, other consultancy and cooperative bodies in the field of social care’ subclass (88999) has been revised upwards by approximately €250 million. Before the revision, these estimates were based on limited source data. They have now been improved with the help of fiscal data and publicly available annual reports. Nevertheless, there remains a group of more than 4000, mostly small, firms within this subclass with combined revenues of approximately €500 million for which no care type is known. 
• Policy and management organisations: +€150 million. Before the revision, the care-related costs of the SVB and the overhead costs of the Ministry of Justice and Security for the procurement of forensic care were missing; they have now been added. 

Set against this improved coverage of the ‘Human Health and Social Work Activities’ section, as part of this revision, the boundary between care activities and non-care activities has been critically re-examined. The intention is that only revenue received by healthcare providers for delivering care should be included in the Health and Social Care Accounts, with other income being excluded. A clear distinction is not always drawn between care and non-care revenue in the revenue data for healthcare providers. Sometimes, care revenue is booked under the item ‘operating returns not otherwise specified’ in the annual report. As a result, before the revision, those operating returns were often still counted as care revenue. Following the revision, the principle applied is that operating returns of healthcare providers which cannot be attributed directly to care, such as income booked under ‘operating returns not otherwise specified’, are only included if there is evidence that they are care-related. This has resulted in a downward correction of care expenditure by approximately €350 million.

Before the revision, the entire costs of the Netherlands Food and Consumer Product Safety Authority (NVWA) were included in the Health and Social Care Accounts, based on the view that much of what the NVWA does directly or indirectly impacts public health. However, in other fields the cost of policy that indirectly impacts public health is not included, such as the cost of drinking water provision or improving air quality. For this reason, following the revision it is been decided to only include the Ministry of Health, Welfare and Sport’s share of the costs of the NVWA. A similar correction relates to the costs of the National Institute for Public Health and the Environment (RIVM); following the revision, environment-related costs are no longer included. These changes have resulted in a downward correction of over €300 million.

7.1 The effects of the revision on financing 

In some cases, the improved coverage of the Q section has resulted in higher out-of-pocket payments and higher expenditure by (local) government. The higher out-of-pocket payments are primarily the result of adding the non-insured care part of SBI subclass ‘Practices of psychologist’ (86913), the full inclusion of the SBI subclass ‘Other paramedical practitioners’ (86919) and a higher estimate for non-insured care within the ‘Specialist medical practices’ subclass (86221). The increased expenditure by municipal authorities in particular is primarily the result of adding the counselling and social care for disabled (SBI subclass 88103) and of a high estimate for the ‘Providers of social work and other social care’. With regard to the total amounts for out-of-pocket payments and government financing, the caveat must be added that the two financing schemes are often used as a residual item in order to make the financing of expenditure on a particular provider category ‘add up’. This means that any errors in the estimate of total expenditure, including possible errors in the estimates of double counting discussed previously, ‘end up’ in the estimates for out-of-pocket payments or government expenditure. This means that the figures are less accurate than the total amounts for Zvw, Wlz and supplementary insurance schemes. 

The increase for the Wlz, which is the result of the use of claims to facilitate improved estimates, means there is now much better alignment with the National Health Care Institute figures. The total extent of supplementary insured care has not changed as a result of the revision; however, the distribution across the different provider categories has improved significantly, thanks to the use of claims. The total figure for Zvw expenditure differs very little; however, there are major shifts between provider categories, caused by different factors:

• An amount of approximately €2.3 billion has moved from ‘Other providers of medical specialist care’ to ‘General hospitals’. This is because the revenues of independent medical specialists working in general hospitals are now included under the hospital submitting the claim.
• An amount of over €700 million has moved from ‘GP practices’ to ‘Providers of other healthcare services’. These shifts are a result of, among other things, using claims details and linking them to the CBS business register, with the associated SBI classification: a large part of integrated care (multidisciplinary care) and of GP care delivered in the evenings and at weekends by GP practices is claimed by partnerships of GP practices, which fall under the SBI subclass ‘Umbrella organisations in the field of health care and other support activities for health care’ (86929). In addition, expenditure on training GPs has moved from ‘GP practices’ to ‘Providers of other healthcare services’.

There have also been some small shifts, for example because the SBI subclass ‘Dental specialists’ has now been merged with ‘Dental practices’, and because a limited share of ‘Medical goods and medicaments’ claims originate from companies that provide care as a secondary activity. In general, the use of claims reveals that expenditure on a particular care type is often distributed across many different provider categories.

The same phenomenon is true of expenditure under the Wmo and Child and Youth Act. Although claims were not (yet) available at the time of writing, the wider use of Corporate Social Responsibility annual reports, as described in section 3, has provided better insight into which provider categories provide care under the Wmo and Child and Youth Act.

A fact which was not considered in the Health and Social Care Accounts before this revision is that charity money also goes to the ‘Human Health and Social Work Activities’ section (section Q). An attempt has now been made to estimate its size. In this connection, it is noteworthy that social care-related charitable funds fall under section Q, namely within SBI subclass ‘Community centres, other consultancy and cooperative bodies in the field of social care’ (88999). Aside from a few exceptions, healthcare-related charitable funds fall into another subclass, outside section Q. With the help of data from the Central Bureau for Fundraising (CBF), an estimate has been made of how much money charitable funds spend on their causes. However, there is no data on how much of that money ends up with care institutions. Charitable funds also undertake all kinds of social care activities. Based on information from annual reports and on websites of the various charitable funds, a cautious estimate has been made of how much money is involved, with the figure being put at €380 million. This is money that goes to UMCs for research (before the revision, this was still booked as out-of-pocket payments) and money that is spent on social care work (before the revision that was ‘concealed’ within other financing schemes). As stated above, this figure is a cautious estimate and, moreover, an important financing flow is still missing: there are many care institutions that have ANBI status and, as such, undoubtedly also receive charity money directly. We have not (yet) succeeded in making a good estimate of this financing flow. 

In terms of the shares of the different financing schemes within the overall financing of health and social care expenditure, the effects of the revision are as follows:

Alongside total health and social care expenditure, CBS also publishes a figure for expenditure on healthcare. This is the internationally comparable figure in accordance with the System of Health Accounts and comprises medical care, rehabilitation care, long-term healthcare, support services, medicines and medical goods, preventive care and policy and management of the healthcare system. 

The following components are not included in the international figure but are included in the national figure: long-term social care (primarily home care and support), social work activities, children's day care, training and R&D expenditure. 

The international figures are broken down along three axes: providers (HP: health providers), financing (HF: health financing), care function (HC: health functions). With new sources such as claims under the Healthcare Insurance Act and Long-term care Act (Wlz) claims becoming available, CBS has improved both the allocation of financing to providers and the allocation of expenditure to functions of health and social care in the figures after revision. In addition, the change of provider categories following the revision also has an impact on the allocation of HP categories. 

The international functions of health and social care have been agreed internationally (OECD-Eurostat-WHO) and are therefore a compromise, aimed at international comparability. In the functions of health and social care, there is a strong focus on the setting in which the care is provided: residential, day care, ambulatory or at home. With regard to care types, setting has a less prominent role. When operationalising the figures by function, CBS takes the care type as its starting point. With the help of supplementary information such as microdata on claims and indications, the care types are then broken down by setting, for example for perinatal care, care for the disabled and (geriatric) rehabilitation care. The relationship between the international functions of health and social care and the care types is explained further in annex 2.

8.1 Expenditure on healthcare 2021, before and after revision

The figure for total expenditure on healthcare, in accordance with the internationally comparable definition, is €3.3 billion higher after revision. The figures before and after revision are published in the table below, broken down by function. The most important changes are described beneath the table. 

• Major shift from ‘In-patient curative care’ (HC.1.1) to ‘Specialist outpatient curative care’ (HC.1.3.3). The distribution of medical specialist care between inpatient care, day treatment and outpatient care was already based on Zvw claims before the revision. An important change relates to the expenditure on expensive medicinal products: whereas previously they were distributed using the same general distribution key, they are now entirely allocated to ‘Specialist outpatient curative care’ (HC.1.3.3).
• Major shift from ‘Day cases of curative care’ (HC.1.2) to ‘Specialist outpatient curative care’ (HC.1.3.3). Previously it was assumed that all care provided by private clinics and independent treatment centres consisted of day treatments. However, the independent treatment centres’ claims reveal that the care provided is largely outpatient care. Similarly in curative mental health care, based on older sources, a significant proportion of expenditure was attributed to day treatments. However, consultation with experts from sector bodies reveals that day treatments barely occur in curative mental health care. In addition, since the introduction of the Care Performance Model (‘Zorgprestatiemodel’), the only differentiation that can be made in curative mental health care is between consultations and residential care, which means that curative mental health care falls exclusively under ‘In-patient curative care’ (HC.1.1) and ‘Specialist outpatient curative care’ (both HC.1.3.3 and HC.1.3.9). Those consultations which coincide with residential care are assigned to ‘In-patient curative care’ (HC.1.1).
• The further rise in ‘Specialist outpatient curative care’ (HC.1.3.3) is largely the result of improved coverage of the ‘Specialist medical practices and outpatients’ clinics’ subclass (SBI 86221), which has yielded a better picture of expenditure on insured and uninsured outpatient care.
• Sharp rise in ‘Other outpatient curative care’ (HC1.3.9). The increase is primarily the result of improved coverage of subclasses such as ‘Other paramedical practitioners’ (SBI 86919) and ‘Practices of psychologist’ (SBI 86913). Not all activities of psychologist practices are deemed to be healthcare; psychological assistance care is classified under social work activities, and as such falls outside the System of Health Accounts.
• Major shift from ‘Long-term social care cash benefits’ (HCR.1.2) to ‘Long-term nursing care [health]’ (HC.3.4). The personal budgets (cash benefits) under the Wlz which accrue to ‘Households’ (HP.8) were previously assigned to ‘Long-term social care cash benefits’ (HCR.1.2), with the assumption being that the bulk is long-term social care. Thanks to an improved linkage, we observe that a large proportion of these personal budgets in fact do not accrue to households but to other providers, including ‘Secondary providers of healthcare’ (HP8.2) and ‘Rest of the world’ (HP.9). For this reason, this expenditure is now classified as ‘Long-term nursing care [health]’ (HC.3.4).
• Sharp fall in ‘Support services’ (HC.4). Analyses of the annual accounts of independent laboratories reveal that they primarily supply services to other healthcare providers (see 6.1 Examples of double counting). This has led to a correction in order to prevent double counting, which has resulted in a fall in expenditure on ‘Laboratory services’ (HC.4.1). In addition, it has been determined that the expenditure previously assigned to ‘Imaging diagnosis’ (HC.4.2) largely relates to R&D activities, which do not belong under health expenditure.
• Sharp rise in Preventive care (HC.6). The increase is primarily explained by a broader allocation of preventive dental care. Whereas previously only dental care for under 18s (reimbursed under the Zvw) was allocated to ‘Preventive care’ (HC.6), now preventive activities for adults, such as x-rays, check-ups, diagnostics and oral hygiene, are also classified under ‘Preventive care’ (HC.6). In addition, dietary advice has been more accurately catalogued thanks to a refined classification of dieticians’ practices and improved coverage of the ‘Paramedical’ subclass (SBI 86919). Improved coverage of the SBI subclass ‘Preventative health care (no health and safety at work)’ (86923) has also contributed significantly to the increase in expenditure in this category. Finally, a shift is taking place from ‘Health promotion, multi-sector’, (HCR.2) to ‘Preventive care’ (HC.6).
• Sharp increase in ‘Healthcare n.e.c.’ (HC.9). The significant increase in this category is due to the fact that more detailed distribution keys have been used (based on microdata) in order to more accurately determine what expenditure cannot be specifically assigned to a care type. Previously, this expenditure was distributed across different functions of health and social care but it is now explicitly classified as ‘Healthcare n.e.c.’ (HC.9). This relates primarily to items for which the applicable function is not known, such as ‘Other care revenue’ and ‘Subsidies’.
• There is no longer any expenditure reported under ‘Health promotion, multi-sector’ (HCR.2). Because only the Ministry of Health, Welfare and Sport-funded part of the NVWA’s costs is now included following the revision, this portion has been assigned to ‘Preventive care’ (HC.6). 

New classification of actors based on HP classification

With the introduction of a new classification of actors within the Health and Social Care Accounts (see The new classification of care and social care providers), adjustments have been made to align them with the international classification of providers (HP classification). The main changes are:

1. Independent psychiatrists have been included in the more comprehensive ‘Ambulatory mental health care and psychiatrists’ actor, moving from ‘Medical practices’ (HP.3.1) to ‘Ambulatory health care centres’ (HP.3.4).
2. Umbrella organisations in the field of health care have been moved from ‘Other administration agencies’ (HP.7.9) to ‘Ambulatory health care centres’ (HP.3.4).
3. Several existing and new actors have been assigned to ‘Secondary providers of health care’ (HP.8.2). They include retailers of medicaments, welfare institutions for the elderly, children's day care centres, local welfare work and other secondary healthcare providers because they all provide health care as a secondary activity.

The table below shows the figures before and after the revision, with the new classification of actors based on the HP classification being applied in both cases. The differences are largely explained by the factors elucidated previously regarding health and social care expenditure (see 7. Health and social care expenditure). For this reason, the causes of these differences are only touched on briefly in the text below. 

A complete allocation of the 73 actors based on the HP classification may be found in annex 1.

• Major shift from ‘Medical practices’ (HP.3.1) to ‘General and university hospitals’ (HP.1.1). Fees for the work of independent medical specialists and dental specialists in hospitals are now allocated to hospitals. The decrease in ‘Medical practices’ (HP.3.1) is largely offset by improved coverage of the ‘Specialist medical practices and outpatients’ clinics’ subclass (SBI 86221), which has yielded a more complete picture of expenditure on insured and uninsured outpatient care.
• Decrease in ‘Long-term nursing care facilities’ (HP.2.1). Due to the use of microdata, it is now possible to draw a clearer distinction between ‘Long-term social care’ (HCR.1) and ‘Services of long-term nursing care’ (HC.3). This means that a larger share of Wlz expenditure is assigned to ‘Long-term social care’ (HCR.1), resulting in a fall in expenditure for this provider type. This shift is slightly offset by the fact that more Wlz personal budgets (cash benefits) are allocated to these providers, leading to an increase in ‘Home care’ (HC.3.4).
• Increase in ‘Facilities for the disabled’ (HP22-29). For this provider type, the use of microdata to differentiate between ‘Long-term care (social)’ (HCR.1) and ‘Long-term nursing care’ (HC.3) results in a larger share being allocated to healthcare.
• Sharp fall in ‘Medical practices’ (HP.3.1). Previously, claims for GP care were functionally allocated to the ‘General medical practices’ actor and those for multidisciplinary care to ‘Partnerships’. In the new figures, the claims are allocated to the actors submitting them, which means a significant share falls outside ‘Medical practices’ (HP.3.1).
• Decrease in ‘Dental practices’ (HP.3.2). As in the case of GP care, claims for dental care were previously functionally allocated to the ‘Dental practices’ actor. Now they are allocated to actors that submit claims for care, even if they fall outside the ‘Dental practices’ subclass (86231).
• Sharp rise in ‘Other health care practitioners’ (HP.3.3). The increase is the result of improved coverage of the SBI subclasses ‘Practices of psychologist’ (86913) and ‘Other paramedical practitioners’ (SBI 86919), including alternative healers.
• Increase in ‘ Providers of home health care services’ (HP.3.5). Due to improved coverage of the subclasses for ‘Residential and day care for the disabled’ (SBI 8720, 87301) and ‘Residential care for the elderly’ (SBI 8710, 87302), combined with the use of Zvw and Wlz claims, we have now achieved more accurate insight into the revenue from micro-institutions outside the Care Institutions statistics.
• Sharp fall in ‘Laboratories, other ancillary services’ (HP.4.2-4.9): Analyses of the annual accounts of independent laboratories reveal that they primarily supply services to other health care providers, which has led to a correction for double counting and a fall in expenditure on ‘Laboratory services’ (HC.4.1).
• Sharp fall in ‘Pharmacies’ (HP.5.1) and ‘Suppliers of medical appliances’ (HP.5.2): Previously, claims for prescription medicaments and medical goods were allocated to the actors ‘Pharmacies’ and ‘Suppliers of medical appliances’ by default. Now those claims are allocated to the actors which actually submit them. An exception to this are Wlz-financed medical goods, which, in the absence of better options, remain allocated to the actor ‘Suppliers of medical appliances’.
• Increase in ‘Providers of preventive care’ (HP.6): The increase is the result of improved coverage of the SBI subclass ‘Preventative health care (no health and safety at work)’ (86923).
• Small increase in ‘Other administration agencies’ (HP.7.9). The amount assigned to this category after the revision consists of care provided by the actor ‘Umbrella organisations in the field of health care’.
• Increase in ‘Households’ (HP.8.1): The amount assigned to ‘Households’ after the revision consists of expenditure from personal budgets (cash benefits) under the Zvw. Before the revision, this was incorrectly assigned to ‘Long-term social care cash benefits’ (HCR.1.2).
• Very sharp increase in ‘Secondary providers of health care’ (HP.8.2). The use of claims (and personal budgets), linked to the business register, reveals that significantly more healthcare is supplied by secondary providers of healthcare than had previously been assumed.

Health and social care expenditure rose by 2.9 percent in 2022 compared to 2021; before the revision, the growth was 2.0 percent. The main growth in 2022 was seen in the SBI subclasses which have been added with the revision, resulting in a higher revised figure for the growth of total care expenditure. In 2023, expenditure rose by €10 billion to €142.8 billion, an increase of 7.5%. The amount spent by the government remained stable; expenditure financed under the Dutch Healthcare Insurance Act and the Long-term care Act (Wlz) rose by 7.7 and 10.6 percent, respectively, in 2023. Out-of-pocket payments, including insurance excesses and personal contributions, rose by 9.2 percent.

Expenditure on preventive healthcare fell in both 2022 and in 2023, as a result of the cessation of a large part of the coronavirus-related costs, in particular expenditure on testing and vaccination and the ‘healthcare bonus’ scheme. On the other hand, expenditure on shelter care rose substantially in both years. This was primarily due to the soaring costs of asylum seeker shelters in 2022 and 2023.

9.1 Composing preliminary figures

When composing preliminary figures for the previous year (in this case 2023), a lot of data used for the Health and Social Care Accounts are not yet available or only partially so: claims under the Healthcare Insurance Act (Zvw), the Long-term care Act (Wlz) supplementary insurance schemes, data from annual Corporate Social Responsibility reports, data from CBS surveys of healthcare providers and fiscal data. For this reason, the methods used to produce quick, preliminary figures differ significantly from the methods used to produce the more detailed provisional and final figures. 

For the Zvw and Wlz domain, expenditure per care act and care type are calculated using the percentage change over 2022-2023 based on the care figures from the Care Institute’s database, which, by contrast, are available in time. For SBI subclasses in which uninsured care plays a primary role, the development of wages and salaries (policy records of the Employee Insurance Agency (UWV) and the Tax Administration) is used to estimate total revenues. Because the care sector is highly labour intensive, it is assumed that this is a good approach for estimating the development of revenues. Out-of-pocket payments, supplementary insurance schemes and Child and Youth Act and Wmo revenue are estimated using various CBS data on care use (the health survey, youth policy information, Municipal Social Domain Monitor) and rate and price developments. 

With the publication of the revised figures for the years 2021 to 2023, the current revision is not completely at an end. The table showing the price and volume development of health and social care expenditure  will be revised in the course of 2025. In addition, we will be working to add older years to the tables currently published. In view of the far-reaching changes involved in this revision (the coverage of the entire section Q, the use of claims and the methodology to correct for double counting), it is not possible to compile the entire 1998-2020 series in the same way as has now been done for the years 2021-2023. The expectation is that 2020 and 2019 will be added in any event, with further research being needed to establish whether it is also possible to add the years 2016-2018. Because the figures before revision (for the years 1998-2022) will remain available on StatLine under ‘Archive’), this means that for multiple years there will be figures arrived at using both the old and the new approaches. 

As was stated in the introduction, a revision takes place every five years. This means that no major changes will be implemented in the interim. Because revision activities often have a long lead time, consideration is already being given to further improvements of the statistics. Based on the experiences gained during the current revision, a number of topics are being considered. Some examples:

• Acquiring greater insight into the size of groups of providers of care activities delivered under the Wmo and the Child and Youth Act and the distribution between them, and into alignment with the figures for the social domain in the CBS statistics on the finances of municipalities. The use of claims data can undoubtedly contribute to that. 
• Acquiring greater insight into the level of out-of-pocket payments, and their distribution across both provider groups and care types. This relates not to the insurance excess and personal contributions under the different care acts, but to what people themselves pay for uninsured care. The possibility will be explored of obtaining greater insight into out-of-pocket payments for care in the next round of the CBS Budget Survey, which will take place in 2026. Efforts will also be made to identify other potential sources on this subject.
• Improving and possibly expanding the current figures by care type – for example, making the figures on the ‘Prevention’ care type more complete and possibly adding a care (sub)type for youth health care. 
• Making use of the information generated as a result of the expansion of the population of healthcare providers required to file annual corporate social responsibility reports, starting with the 2024 reporting year. Since as part of that accountability requirement, income is reported per Act of Parliament, it may be possible to further improve the financing distribution in the Health and Social Care Accounts.
• Acquiring greater insight into the subclasses of section Q for which little data is currently available, particularly with regard to the breakdown by care types. Even more intensive use of annual reports and the classification of units using online information may offer potential benefits, but would be highly labour-intensive.

The table below shows the 22 groups of providers published on StatLine, with the underlying 73 actors (categories of healthcare providers) in the Health and Social Care Accounts system. The allocation of actors according to the international HP classification is shown in the right-hand column. 

The table below shows the 26 care types with the underlying 62 care subtypes and their relationship to the functions of health care from the System of Health Accounts. In many cases, the care subtypes are broken down further by setting with the help of microdata for the distribution by care type, for example for mental health care, perinatal care, care for the disabled and (geriatric) rehabilitation care.