Designing the Belgian “To Be” State: Transforming MZG/RHM and Finhosta for Patient-Based Costing, Care-Process Alignment, and DRG Reform
Introduction
In Belgium, hospital financing reform should start from the target state to be achieved rather than from the inherited institutional configuration to be incrementally adjusted. That is especially important because the Belgian health system is structurally path-dependent: it combines near-universal compulsory insurance with direct access, predominantly fee-for-service remuneration, and a division of responsibilities between the federal state and the federated entities that was deepened by the sixth State reform. In hospital financing, Belgium already uses All Patient Refined Diagnosis Related Groups (APR-DRG)/Severity of Illness (SOI) information for budget allocation, but not as a true case-based payment system with national cost weights; the current B2 logic of the Budget of Financial Means (BFM) uses national average length of stay by APR-DRG/SOI as a proxy for resource use. At the same time, Belgium has two mandatory hospital datasets with major strategic value: Minimal Hospital Data (MHD) (Minimale Ziekenhuisgegevens (MZG)/Résumé Hospitalier Minimum (RHM)), the compulsory minimum hospital dataset for non-psychiatric hospitals, and Finhosta, the compulsory collection of financial, personnel, and operational hospital data used for the Budget of Financial Means (BFM) or Budget van Financiële Middelen (BFM)/Budget des Moyens Financiers (BMF). The central reform question is therefore not whether Belgium lacks data, but how to redesign these existing datasets into a patient-level costing architecture that supports a more coherent and consistent future model of care and payment.
The most defensible Belgian “to be” state is therefore a target-driven, inter-federal, data-governed system in which care pathways, not legacy silos, become the principal unit of design; patient-level costs are measured and reconciled transparently; APR-DRG/SOI groups are used where clinically and economically appropriate; and hospital payment is aligned with quality of care, affordability, and continuity of care rather than with opaque and convoluted historical compromises. Belgian work on national health and healthcare targets already points in that direction by emphasizing stewardship, participation, monitoring, and explicit links between targets and multiannual budgeting. Internationally, WHO’s DRG guidance and mature patient-level costing systems show that case-based payment works best when it rests on standardized activity data, reliable cost accounting, reconciliation to audited financial totals, and continuous monitoring of incentives and unintended effects (gaming, cherry picking, lemon dropping, ...) . This essay is therefore a normative reform design inferred from Belgian and international evidence, not a description of a fully implemented Belgian model (i.e. deliverables and milestones).
Note: Structurally path-dependent processes are systems where initial conditions, historical events, and early decisions disproportionately shape future outcomes, creating entrenched trajectories that are difficult to alter.
The Belgian “To Be” State in Hospital Financing
A Belgian target-state reform should define the future hospital financing architecture as follows: MZG/RHM becomes the event backbone, Finhosta becomes the standardized cost-and-resource backbone, and both are linked under common governance to produce patient-level hospital costs that can then be aggregated into robust APR-DRG/SOI cost weights. In the current system, Belgium has compulsory MZG/RHM reporting for hospital stays, day care, and emergency contacts, and compulsory Finhosta reporting for accounting, analytical costs, allocation keys, investments, activity, and personnel. Belgium also already has a Technical Cell or Technische Cel/Cellule Technique (TCT) that annually links MZG/RHM to hospital billing data in order to analyse expenditure by treated condition and inform financing rules. What Belgium still lacks is (reliable/transparent) compulsory nationwide patient-level cost registration that covers total hospital resource use rather than reimbursed expenditures alone. KCE has been explicit on that point: Belgium does not yet have compulsory patient-level cost data, and if APR-DRG/SOI-based payment per case is to replace the current allocation logic, insight into underlying costs is required to set the tariff.
This implies a crucial methodological distinction. The “to be” state should not merely aim at better hospital accounting; it should aim at patient-based costing aligned with care processes. Patient-level costing means first allocating resources to activities and then matching those costed activities to the patient event in which they occurred, so that the final unit cost shows the internal structure of cost within an admission, attendance, contact, or pathway segment. Applied to Belgium, that means shifting from parallel reporting systems used mainly for regulatory and budgetary purposes to an integrated costing architecture that can answer operational questions: what does a given APR-DRG/SOI actually cost, which components drive variation, which care pathways are standardizable, and where bundled or case-based payment is justified.
Note: A care pathway is a structured, evidence-based (EBM/EBP) multidisciplinary care plan used to manage the care of a specific patient group over a set period, detailing key (timed) interventions, timeframes, and expected outcomes. It coordinates professional care across settings - from primary to specialized care - ensuring optimal, coherent, consistent, and patient-focused treatment. The “First Hour Quintet” (FHQ) - cardiac arrest, severe trauma, stroke, acute respiratory failure, and cardiac chest pain - are some examples of highly critical care pathways. Focus clinic care pathways for hip and knee arthroplasty are evidence-based, multidisciplinary, and standardized protocols designed to enhance patient recovery, optimize outcomes, and reduce hospital stays, often utilizing ERAS (Enhanced Recovery After Surgery) or fast-track protocols. These focus clinic care pathways generally include preoperative optimization, standardized intraoperative techniques, early mobilization, and structured post-discharge care.
How MZG/RHM Should Be Transformed
Minimal Hospital Data (MHD) (Minimale Ziekenhuisgegevens (MZG)/Résumé Hospitalier Minimum (RHM)) should be redesigned as the master clinical-event file for hospital costing. Official Belgian sources already show that MZG/RHM is mandatory, biannual, and rich in clinically relevant information: it includes demographics, diagnoses, comorbidities, procedures, complications, length of stay (LOS), nursing units involved, discharge status, same-hospital readmission, and APR-DRG, severity of illness (SOI), and risk of mortality (ROM) outputs from the Solventum APR-DRG grouper. It is also used for public financing, hospital feedback, and epidemiology. A target-state reform should therefore make the MZG/RHM stay, day-care attendance, or emergency contact the basic “costed event” for hospital patient-level costing.
However, MZG/RHM in its current central form is not timely enough for prospective case payment. KCE has already warned that the APR-DRG/SOI of a stay is currently available only six to twelve months after discharge at the federal level, which is incompatible with routine case-based invoicing or near-real-time management (e.g. pandemics, disasters, conflicts). For the target state, Belgium should therefore preserve the statutory central MZG/RHM submission but require, or at least standardize, hospital-level near-real-time extraction from the same source systems (EHR), with local or centrally callable grouping at or shortly after discharge. That is not a claim that Belgium already has such an operational architecture; it is the necessary design consequence of using APR-DRG/SOI for prospective costing and payment and in times of crisis.
A further implication is that MZG/RHM should not be treated only as a classification file. Because it records nursing units involved and care-period information, it can support care-process segmentation within the hospital stay. That matters because patient-based costing aligned with care processes should not stop at one single undifferentiated cost per admission. In the Belgian target state, MZG/RHM should support costing by clinically meaningful subcomponents such as emergency phase (ER/ED), ward phase, intensive care (ICU), theatre use (OR), diagnostics, and discharge phase. This would also make APR-DRG/SOI analysis more (hospital) policy-relevant, because the same group can then be examined not only for total cost but for internal cost structure and care pathway variation.
Note: Reliable and timely health data during crises (pandemics, disasters, conflicts) are crucial for tracking system performance, monitoring service disruptions, and guiding rapid responses. It relies heavily on administrative data, routine health information systems (RHIS), and, increasingly, digital tools for real-time monitoring. Effective large scale crisis data management requires internationally standardized, interoperable systems to avoid fragmented, delayed responses. Reliable and timely health data are foundational to effective crisis management, directly impacting epidemiological surveillance, the delivery of care, and overall survival rates during pandemics, natural disasters, and conflicts. In such emergencies, data unavailability or significant delays - often due to damaged infrastructure, fragmented reporting systems (data silos, vendor lock-in), or security challenges - can lead to poor diagnoses, improper patient referrals, and increased mortality.
How Finhosta Should Be Transformed
Finhosta should be redesigned as the nationally standardized resource and cost ledger for patient-level costing. Official Belgian documentation shows that Finhosta is mandatory for all Belgian hospitals and a condition for receiving the Budget van Financiële Middelen (BFM)/Budget des Moyens Financiers (BMF); it collects general accounting data, analytical accounting data, allocation keys by cost centre and budget item, equipment, activity volumes, investments, personnel data, and personnel costs by cost centre. That gives Belgium a much stronger starting point than countries that must build costing infrastructure from scratch. The difficulty is that Finhosta was designed primarily for regulatory submission and budgetary analysis, not for fully operational patient-level costing.
Belgian Health Care Knowledge Centre’s (KCE) recent work shows why the present Finhosta architecture is not yet sufficient. Costs can be registered first in temporary cost centres and then cascaded to final cost centres using allocation keys such as square metres and nursing days, but KCE notes that these keys are often criticized because they insufficiently reflect reality and because the link between the temporary centre and the final consuming centre is not always clear. KCE also found that differences in how hospitals use temporary cost centres limit comparability, and that Finhosta cannot always distinguish categories of personnel cost relevant for precise costing. For the target state, Belgium should therefore retain Finhosta as the statutory financial source of truth, but add a national patient-level costing layer that classifies costs into: directly attributable patient-specific costs; patient-facing shared costs to be allocated by activity drivers; and residual overhead to be allocated using standardized hierarchical rules.
The logic should follow established patient-level costing standards. A healthcare integrated costing model is organized around: collecting costing information, ensuring the correct cost quantum, clearly identifying costs, allocating costs to activities, matching costed activities to patient events, and reconciling outputs back to the financial source. Applied to Belgium, that means every hospital should be required to reconcile the total costs used in patient-level costing to the relevant Finhosta financial totals, while activity-level feeds should be matched to MZG/RHM events wherever possible. Directly traceable items such as pharmaceuticals, implants, blood products, laboratory tests, imaging, theatre minutes (OR), critical-care days (ICU), and specific consultations should be assigned to patient events directly; only the residual shared and overhead components should be allocated indirectly.
For nursing and accommodation-related hospital costs, Belgium should avoid defaulting to crude length-of-stay (LOS) proxies wherever more meaningful drivers are available. That follows both from KCE’s critique of simplistic allocation and from the Belgian structure of MZG/RHM itself, which includes nursing units and nursing-related information. The official MZG/RHM publication material also shows that Belgium already works with nursing-related grouping (NRG) logic. A target-state model should therefore use ward, nursing-unit, care-period, and intensity-sensitive drivers wherever feasible, rather than relying on a single undifferentiated hotel-day logic for all inpatient resource use.
Note: The financial ledger is the overall accounting record, while the cost ledger is a cost-focused record for costing and allocation. A cost ledger is the costing equivalent of a financial ledger: instead of mainly tracking revenues, assets, and liabilities, it tracks where costs arise, what type of costs they are, and to which department, activity, service, product, or patient they belong. A cost ledger is a structured accounting record that captures and classifies the costs incurred by an organization, usually by cost type, cost centre, activity, or service, so that those costs can be analyzed and allocated for management, reporting, or payment purposes. A patient-level costing file is the output produced after costs from the cost ledger are assigned to individual patients or care events.
Note: A healthcare integrated costing model is a structured approach that combines clinical patient flow data with financial information to determine the "true cost" of care, covering the entire patient journey rather than just fragmented services. Unlike traditional, siloed accounting, these models often use micro-costing (detailed, item-level) or Activity-Based Costing (ABC) to allocate costs based on actual resource consumption, allowing for more precise management, better decision-making, and improved value-based healthcare (VBHC).
From Patient-Level Costing to APR-DRG/SOI Weights
Once MZG/RHM and Finhosta are (reliable/transparent) linked through a standardized patient-level costing architecture, Belgium would be in a position to derive national APR-DRG/SOI relative case weights from observed costs rather than from average length of stay (LOS) or negotiated (proxy) prices. WHO defines a relative case weight as the coefficient that adjusts payment upward or downward according to the cost of treating a case group relative to the average cost per case. KCE’s work points in the same direction: if Belgium moves from today’s APR-DRG/SOI-based budget allocation toward payment per case, tariffs need to rest on (reliable/transparent) underlying costs.
That does not mean every hospital service should automatically be paid through a pure DRG tariff. WHO is clear that case-based payment is not a magical solution and that hospital payment systems usually require a mix of methods. KCE’s earlier work on Belgian clustering also warned that not all APR-DRG/SOI groups are suitable for a prospectively determined lump sum, that gaming and upcoding risks rise when payment stakes differ across groups, and that outlier treatment is both a technical and a policy question. The appropriate Belgian target state would therefore use patient-level cost data to identify which APR-DRG/SOI groups are sufficiently standardizable for prospectively set case rates, while preserving other payment components for teaching, research, highly specialized care, emergency preparedness, or extreme cost outliers. Relative weights should preferably be calculated from trimmed means or similarly robust estimators, with explicit outlier rules and regular recalibration.
Note: A trimmed mean is a measure of central tendency obtained by excluding a verified and validated predefined proportion of extreme low and high observations before calculating the average, thereby reducing the influence of outliers.
Note: DRG-based yardstick competition is a regulatory approach used in healthcare, particularly in prospective payment systems (PPS), where hospitals are reimbursed based on a fixed price for a "Diagnostic Related Group" (DRG) rather than their actual costs. This system, first analyzed in this context by Shleifer (1985), acts as a "yardstick" because a hospital's reimbursement rate is linked to the average performance (or cost) of other, similar hospitals.
Governance and Implementation for the Belgian Target State
In Belgium, this transformation can only work through inter-federal governance with clear stewardship. The Health Data Agency (HDA) was created precisely because Belgian health data are fragmented across institutions and systems and because secondary use needs a uniform, transparent, and secure governance framework (EHDS). The European Social Observatory (OSE) work on Belgian health targets similarly emphasizes that effective target governance must combine open participation with clear stewardship, clarify roles in monitoring and evaluation, and link targets to multiannual budgeting. Applied to MZG/RHM and Finhosta reform, that means a competent and independent inter-federal programme board should govern data standards, cost-accounting rules, pseudonymized linkage, metadata, version control, audit, publication, and yearly recalibration.
Implementation should be phased. Belgian and WHO evidence both support gradual rollout where administrative burden and coding maturity are relevant. The prudent sequence would be: first, standardize data models and costing manuals; second, pilot patient-level costing in a subset of general hospitals and high-volume APR-DRG/SOI groups; third, run the new model in parallel with current financing logic; fourth, publish comparative benchmarks and undertake targeted audits; and only then use mature cost weights for national tariff-setting. Belgium’s own earlier reform analyses explicitly recommended gradual implementation and warned against launching case-based payment for groups where predictability and coding robustness are insufficient. The problem is not de DRG-system, but the lack of reliable and timely patient- and cost-based data in relation to the care process itself (patient-based costing).
The Main Risks and Pitfalls in Belgian Reform, and How to Address Them
The first major risk is institutional fragmentation. In Belgium, hospital financing, primary care, mental health, prevention, and long-term care are not governed from a single centre of authority. A reform that begins from the status quo will therefore tend to mirror existing competencies rather than redesign them around patient pathways. The antidote is not (the usual) institutional fantasy, but explicit target-state governance: define a small set of inter-federal outcomes, assign competent lead stewardship, and force MZG/RHM-Finhosta transformation into a common programme with binding data and costing standards.
The second risk is reform by layering. Belgium has a long tradition of adding corrective instruments around inherited fee-for-service and budgetary structures (fail forward). In hospital financing, that risks producing new reporting burdens without changing the logic of how costs are understood or how pathways are financed. The way to avoid this is to separate clearly the target-state architecture from transition mechanisms: current MZG/RHM and Finhosta returns can continue during transition, but they should be subordinated to a published and maintained roadmap toward (transparent/reliable) patient-level costing, cost-based DRG weights, and pathway-sensitive monitoring.
The third risk is poor comparability of cost data across hospitals. KCE has already shown that the way hospitals use temporary cost centres and allocation keys can materially affect comparability. That is a central threat to national case weights. The response should be a national costing manual with mandatory minimum driver sets, reconciliation rules, restricted use of temporary centres for tariff-setting purposes, external validation, and annual recalibration. Hospitals can keep richer internal methods for management, but the national tariff engine should rest on a common minimum standard.
The fourth risk is gaming under DRG-type incentives. WHO warns that case-based payment can induce skimping, early discharge, cream-skimming (cherry picking), lemon dropping, unbundling, readmissions, and volume growth; KCE warns specifically about upcoding risks in the Belgian APR-DRG context. These are not reasons to reject DRGs, but reasons to design them correctly. Belgium should therefore combine APR-DRG/SOI payment with competent coding audits, monitoring of readmissions, mortality, length of stay (LOS), referrals and waiting lists, quality review against care pathways, and explicit outlier payments for exceptionally complex or costly cases. A blended payment model is safer than a DRG-only model.
The fifth risk is digital and administrative overload. Patient-level costing is data-intensive, and WHO explicitly notes that coding capacity and administrative burden must be manageable. This risk is sharper in a fragmented system already affected by staffing shortages, including hospital workforce pressure. The answer is phased implementation, (clinically relevant) automation wherever possible, protected implementation funding, and joint training and certification for coders, clinicians, finance teams, and informatics staff. Reform that assumes unlimited administrative capacity will fail even if its design is conceptually sound.
The sixth risk is equity blindness. A technically sophisticated patient-level costing system could still fail Belgium if it is used only to optimize hospital tariffs while ignoring out-of-pocket (OOP) exposure, and care pathway costs borne by patients outside the hospital. OECD data show that Belgian out-of-pocket spending remains relatively high. The Technische Cel/Cellule Technique (TCT) and related linkages already show that Belgian hospital data can be connected with reimbursement and expenditure data, and other Belgian datasets can inform co-payments and patient-borne costs (out-of-pocket expenses). The target state should therefore use costing reform not only for hospital efficiency, but also to identify where financing design shifts costs onto patients and undermines affordability.
Note: A national costing manual in healthcare is a standardized guideline, such as the Dutch Costing Manual, Independent Health and Aged Care Pricing Authority (IHACPA) in Australia, Agence Technique de l'Information sur l'Hospitalisation (ATIH) in France, NHS Approved Costing Guidance, or InEK Costing Standard in Germany, providing standardized methods and reference prices for calculating the cost of health services. It enables consistent economic evaluations, ensures transparency for policy decisions, and allows for fair comparison of healthcare costs across different providers.
Note: Coding certifications and training provide structured learning paths for software development, IT, and medical coding, with options ranging from free online courses to intensive bootcamps. Providers include professional bodies like AHIMA and AAPC, offering training in (specialized) medical coding (AHIMA CCS (Certified Coding Specialist), AAPC CPC (Certified Professional Coder)).
Note: DRG Skimping (Stinting) is reducing the quality, quantity, or intensity of services provided to a patient to keep costs below the fixed DRG reimbursement amount. Early Discharge is discharging patients before they have fully recovered or before they are clinically stable, often done to free up beds and reduce costs associated with long hospital stays. Cream-Skimming (Cherry-Picking) means actively selecting the most profitable, low-risk patients (the "cream" or "cherries") while avoiding complex, high-cost patients. Lemon Dropping is the opposite of cherry-picking; this involves discharging, transferring, or avoiding ("dropping") high-risk, unprofitable, or complex patients ("lemons"). Unbundling is the illegal practice of billing for services individually that are already included in the bundled DRG payment, aiming to receive higher reimbursement. Readmission is a patient’s return to the hospital shortly after discharge, often within 30 days, due to a complication from the initial stay. Potentially preventable readmissions (PPRs) are unplanned, clinically related return hospitalizations within a specific timeframe (often 30 days) that could be avoided through better inpatient care, discharge planning, or post-discharge follow-up. These potentially preventable readmissions highlight gaps in care quality and care coordination. Volume Growth is the strategy of increasing the overall number of admissions, procedures, or discharges to increase revenue.
Note: Equity blindness in healthcare occurs when providers or systems ignore social, economic, and racial factors affecting patient outcomes, mistakenly treating everyone the same rather than addressing specific needs. This, often unintentional, approach perpetuates disparities for underserved, disabled, and minority populations. Addressing it requires targeted, equitable care models. In healthcare, "cherry-picking" when applied to Social Determinants of Health (SDOH), specifically targets patients based on their socio-economic stability. "Lemon-dropping" in the context of Social Determinants of Health (SDOH) refers to a practice in value-based healthcare (VBHC) systems where providers or insurance plans avoid, disincentivize, or drop patients with complex social needs - such as poverty, housing instability, or lack of transportation - because these patients are associated with higher costs and lower quality scores.
Conclusion
Applied to Belgium, the argument for beginning from the “to be” state, instead of continuing patching up the "as is" state, is especially strong. The country already possesses the core data infrastructures needed for a major hospital-financing redesign, but those infrastructures were built for regulation and budget administration, not yet for a patient-level, clinical pathway-sensitive costing system. The proper reform objective is therefore not to continue tweaking MZG/RHM and Finhosta at the margins, but to transform them into a unified costing architecture in which MZG/RHM defines the patient event, Finhosta defines the reconciled cost quantum, and both together generate credible APR-DRG/SOI cost weights under competent independent inter-federal governance. That architecture would make Belgian hospital reform more transparent, more analytically defensible, and more compatible with broader goals of quality, affordability, and continuity of care. The principal dangers are fragmentation, layering, non-comparable cost accounting, gaming, overload, and equity blindnes. None of these risks is trivial, but each can be managed through competent and clear stewardship, phased implementation, standardized costing rules, audit, blended payment design, and systematic monitoring of milestones and deliverables as a learning process rather than a one-off technical exercise.
Bibliography (selection)
Bedford, D., & Sivabalan, P. (2021). Patient level costing in Australia: Uses, challenges, and future opportunities.
Belgian Health Care Knowledge Centre (KCE). (2016). Clustering pathology groups on hospital stay similarity – Short Report (KCE Report 270C).
Belgian Health Care Knowledge Centre (KCE). (2020). Health System Performance Assessment: how equitable is the Belgian health system? (KCE Report 334C).
Belgian Health Care Knowledge Centre (KCE). (2024). Indirect hospital costs as a mark-up percentage on direct costs in Belgium (KCE Report 380C).
Belgian Health Care Knowledge Centre (KCE). (2025). Protocol for NEED (needs examination, evaluation and dissemination) research projects (KCE Report 406).
Bourn, M., & Ezzamel, M. (1986). Costing and budgeting in the National Health Service. Financial Accountability & Management, 2(1), 53-71.
da Silva Etges, A. P. B., Cruz, L. N., Notti, R. K., Neyeloff, J. L., Schlatter, R. P., Astigarraga, C. C., ... & Polanczyk, C. A. (2019). An 8-step framework for implementing time-driven activity-based costing in healthcare studies: APB Silva Etges et al. The European Journal of Health Economics, 20(8), 1133-1145.
Federal Public Service Health, Food Chain Safety and Environment. (2026). Minimale Ziekenhuis Gegevens (MZG).
Federal Public Service Health, Food Chain Safety and Environment. (2026). Inleiding MZG (guidance manual).
Federal Public Service Health, Food Chain Safety and Environment. (2026). Finhosta: onmisbare inzamelingen voor Belgische ziekenhuizen.
Federal Public Service Health, Food Chain Safety and Environment. (2025). Brochure Finhosta_BUD_2024 4.8 (budget manual).
Federal Public Service Health, Food Chain Safety and Environment. (2025). Brochure Finhosta_ANN_2024 4.7 (annual manual).
Gerkens, S., & Merkur, S. (2024). Belgium: Health System Summary, 2024. European Observatory on Health Systems and Policies / WHO Regional Office for Europe.
Goldfield, N. I., McCullough, E. C., Hughes, J. S., Tang, A. M., Eastman, B., Rawlins, L. K., & Averill, R. F. (2008). Identifying potentially preventable readmissions. Health care financing review, 30(1), 75.
Health Data Agency Belgium. (2026). About us; The European Health Data Space (EHDS).
Hughes, J. S., Averill, R. F., Goldfield, N. I., Gay, J. C., Muldoon, J., McCullough, E., & Xiang, J. (2006). Identifying potentially preventable complications using a present on admission indicator. Health Care Financing Review, 27(3), 63.
Kaplan, A. L., Agarwal, N., Setlur, N. P., Tan, H. J., Niedzwiecki, D., McLaughlin, N., ... & Saigal, C. S. (2015, March). Measuring the cost of care in benign prostatic hyperplasia using time-driven activity-based costing (TDABC). In Healthcare (Vol. 3, No. 1, pp. 43-48). Elsevier.
Keel, G., Savage, C., Rafiq, M., & Mazzocato, P. (2017). Time-driven activity-based costing in health care: a systematic review of the literature. Health policy, 121(7), 755-763.
Krafft, T., Castrillo-Riesgo, L. G., Edwards, S., Fischer, M., Overton, J., Robertson-Steel, I., & König, A. (2003). European emergency data project (EED Project) EMS data-based health surveillance system. The European Journal of Public Health, 13(suppl_1), 85-90.
Lancry, P. J., & Sandier, S. (1999). Rationing health care in France. Health Policy, 50(1-2), 23-38.
Martini, J., & Davesne, A. (2024). Health and health care targets: What lessons can Belgium learn from other countries? European Social Observatory.
NHS England. (2024). Integrated information requirements, costing processes and costing methods 2024.
OECD, European Commission, and European Observatory on Health Systems and Policies. (2025). State of Health in the EU: Country Health Profile 2025 – Belgium.
Oostenbrink, J. B., Koopmanschap, M. A., & Rutten, F. F. (2002). Standardisation of costs: the Dutch Manual for Costing in economic evaluations. Pharmacoeconomics, 20(7), 443-454.
Office of Strategic Services (1944). Simple Sabotage Field Manual by United States.
Simonet, D. (2010). Healthcare reforms and cost reduction strategies in Europe: the cases of Germany, UK, Switzerland, Italy and France. International Journal of Health Care Quality Assurance, 23(5), 470-488.
Savva, N., Debo, L., & Shumsky, R. A. (2023). Hospital reimbursement in the presence of cherry picking and upcoding. Management Science, 69(11), 6777-6799.
Shleifer, A. (1985). A theory of yardstick competition. The RAND journal of Economics, 319-327.
Sloan, F. A., & Steinwald, B. (1980). Effects of regulation on hospital costs and input use. The Journal of Law and Economics, 23(1), 81-109.
Tan, S. S., Bouwmans, C. A., Rutten, F. F., & Hakkaart-van Roijen, L. (2012). Update of the Dutch manual for costing in economic evaluations. International journal of technology assessment in health care, 28(2), 152-158.
Versteegh, M., Knies, S., & Brouwer, W. (2016). From good to better: new Dutch guidelines for economic evaluations in healthcare. Pharmacoeconomics, 34(11), 1071-1074.
World Health Organization. (2023). Can people afford to pay for health care? New evidence on financial protection in Belgium. Belgium, Office for Health Systems Financing (Barcelona) (HSF).
World Health Organization. (2020). Diagnosis-Related Groups (DRG): A Question & Answer Guide on Case-Based Classification and Payment Systems (Health Financing Guidance No. 10).
Van Osta, Peter. An essay concerning a new Healthcare.