Patient, Procedure, and Service-Line Costing

Introduction

Healthcare organizations increasingly compete and contract on the basis of value rather than volume. Yet “value” cannot be evaluated without credible knowledge of cost. A hospital may know the total cost of its cardiac service line, but managers also need to know the cost of a particular bypass procedure, the resources consumed by an individual patient, and whether reimbursement covers the full economic cost of care. These questions require costing systems that connect clinical activity with financial resources.

Patient, procedure, and service-line costing are related but distinct levels of analysis. Patient-level costing assigns resources to a specific episode of care. Procedure costing estimates the resources required for a defined intervention, such as a cataract operation or a magnetic-resonance-imaging examination. Service-line costing aggregates costs and revenues for a clinically coherent group of services, such as oncology, orthopaedics, or maternity care.

The central managerial problem is attribution. Direct items such as an implant or a drug may be traced to a patient with reasonable accuracy. Shared resources—operating theatres, nursing units, information systems, utilities, and senior administration—must be assigned using cost drivers. The choice of driver affects reported margins, service-line comparisons, pricing, capacity decisions, and assessments of cross-subsidization.

This chapter develops a practical framework using relative value units, case-mix measures, diagnosis-related groups, clinical pathways, activity-based costing (ABC), and time-driven activity-based costing (TDABC). It also evaluates the limitations of each method and explains how governance, data quality, and clinical engagement determine whether a costing system is useful.

Learning objectives

By the end of this chapter, the reader should be able to:

1. Distinguish patient, procedure, and service-line costing.
2. Explain how RVUs, case-mix indices, and DRGs support cost allocation and comparison.
3. Construct a patient-level cost from direct costs, overhead pools, and activity drivers.
4. Apply ABC and TDABC to clinical pathways.
5. Interpret service-line margins and identify possible cross-subsidization.
6. Assess cost accuracy, limitations, ethical risks, and governance requirements.
7. Recommend improvements to a healthcare costing system for managerial decision-making.

1. Levels and purposes of healthcare costing

Costing has different purposes, and no single system is optimal for all of them.

A cost object is the item for which cost is measured. It may be a patient, visit, procedure, diagnosis, pathway, department, or service line. A cost pool is a grouping of costs with a common allocation logic. A cost driver is a measurable factor that causes or reasonably explains variation in a cost pool.

Costs may be classified as:

• Direct costs: traceable economically to a cost object, such as a patient-specific implant.
• Indirect costs: shared by several cost objects, such as ward supervision.
• Variable costs: change substantially with activity in the relevant range, such as disposable supplies.
• Fixed costs: do not change immediately with volume, such as building depreciation.
• Capacity-sustaining costs: support the existence of a department but are not caused by an individual encounter, such as compliance administration.

The distinction between direct and indirect costs is not absolute. A cost is direct only relative to a specified cost object. Nursing may be indirect to a procedure but direct to a ward or service line.

A useful conceptual identity is:

This identity is an accounting structure, not a claim that every assigned cost is causally caused by the patient. It is therefore essential to disclose whether reported cost means full absorption cost, incremental cost, or avoidable cost. Short-run pricing and capacity decisions often require incremental or avoidable cost, whereas financial reporting and long-run sustainability analysis may require full cost.

2. Measurement foundations: from clinical activity to cost

A costing system normally follows five stages:

1. Define the cost object and time period.
2. Capture resource quantities and prices.
3. Trace direct resources to the cost object.
4. Assign shared costs through cost pools and drivers.
5. Validate results against operational and financial totals.

For a patient episode i, a general cost model is:

where Cᵢ is the total cost of episode i; Dᵢ is any directly assigned cost not represented in the other terms; qᵢⱼ is the quantity of resource j used by the patient; pⱼ is the unit cost of resource j; aᵢₖ is the amount of activity-driver k consumed; and rₖ is the cost per unit of that driver. Quantities and driver units must be measured consistently—for example, minutes with minutes, tests with tests, and doses with doses.

The formula assumes that resource quantities are observable or estimable and that unit prices are appropriately calculated. It does not prove causality. A driver such as length of stay may be a useful proxy for ward cost, but it can over- or under-assign cost if patients with equal days consume materially different nursing time.

Direct costing and allocation

A patient’s direct cost may include:

• pharmaceuticals and blood products;
• implants and prostheses;
• laboratory and imaging tests;
• procedure-specific supplies;
• directly recorded clinician or therapist time; and
• external services purchased for that patient.

Indirect costs may be allocated in stages. For example, facilities costs may first be assigned to departments using floor area, information-technology costs using users or devices, and operating-theatre costs using minutes. Departmental costs can then be assigned to patients using procedure minutes, nursing hours, tests, or encounters.

Sequential allocation recognizes that support departments serve one another. Reciprocal allocation is more theoretically complete when support departments have material interdependence, but it requires more data and often provides little practical benefit if the amounts are small.

3. Relative value units and case-mix measures

Relative value units

A relative value unit (RVU) expresses the relative resource intensity of a service compared with a reference service. In physician payment systems, RVUs may include components for professional work, practice expense, and professional liability. In internal costing, an organization may construct its own procedural or nursing RVUs.

If a cost pool has total cost P and the services consume R total RVUs, the cost per RVU is:

The assigned cost for service s with rₛ RVUs is:

The method assumes that RVUs reasonably represent relative resource consumption within the pool and that the pool is sufficiently homogeneous. It is inappropriate to use one RVU schedule to allocate costs that have fundamentally different drivers, such as laboratory reagents and inpatient accommodation.

Illustrative example. A rehabilitation department has monthly costs of $120,000 and produces 6,000 therapy RVUs. Its rate is $20 per RVU. A patient receiving 18 RVUs is assigned $360 of rehabilitation cost. The result is an average resource assignment, not a measurement of the therapist’s exact cost for that patient.

RVUs are useful where detailed time data are unavailable, but they may become outdated, embed professional payment priorities rather than actual resource use, and conceal variation in complexity within a code.

Case-mix index

A case-mix index (CMI) summarizes the average relative complexity of a hospital’s inpatient cases. If each case has a relative weight wᵢ and there are N cases:

The CMI is dimensionless. A higher CMI generally indicates more resource-intensive cases, but it is not itself a cost measure. Coding practices, documentation, comorbidities, transfer patterns, and grouping rules affect CMI.

For a department or service line, the average cost per weighted case can be calculated as:

This measure is useful for comparing units with different case mixes, but comparisons require similar scope, accounting policies, wage markets, technology, and patient populations.

Diagnosis-related groups

Diagnosis-related groups (DRGs) classify inpatient episodes into clinically and economically similar categories using diagnosis, procedures, age, complications, discharge status, and other variables. DRGs support prospective payment, benchmarking, utilization management, and analysis of cost variation.

A DRG should not be treated as a patient cost. It is a classification category. Two patients in the same DRG may differ in complications, social needs, length of stay, or discharge destination. Conversely, an individual patient may be grouped differently because of documentation or coding changes rather than a change in clinical resources.

DRG-based analysis commonly compares:

• actual cost per case with payment per case;
• length of stay with an expected or benchmark length of stay;
• complication rates and readmissions; and
• cost variation within the same DRG.

A hospital should distinguish between a payment weight, which supports reimbursement, and a resource-consumption weight, which supports internal costing. They may be correlated without being identical.

4. Clinical pathways as costing frameworks

A clinical pathway specifies the expected sequence and timing of care for a condition or procedure. It may include assessment, diagnostic testing, treatment, nursing care, rehabilitation, discharge planning, and follow-up.

Pathway costing connects activities to expected resource quantities. For pathway p:

The estimate assumes that the pathway is defined clearly, resource quantities are based on credible evidence or local observation, and deviations are separately recorded. A pathway is not a rigid prescription. Clinical exceptions may be appropriate, and costing should not penalize necessary individualized care.

Pathway costing is particularly valuable for bundled payments and service redesign. It can reveal that a low-cost procedure becomes expensive because of delayed imaging, avoidable inpatient days, repeated tests, or poor coordination after discharge.

Variation analysis compares actual patient cost with expected pathway cost:

A positive variance may indicate complications, inefficient processes, incomplete pathway design, or appropriate complexity. Interpretation requires clinical review rather than automatic judgment.

5. Activity-based costing

ABC assigns resource costs to activities and then assigns activity costs to cost objects using cost drivers. Typical healthcare activities include registration, triage, medication administration, imaging, surgery, discharge planning, and claims processing.

The basic ABC rate is:

The cost assigned to patient i is:

Using practical rather than theoretical capacity avoids inflating rates when resources are temporarily idle. For example, if a diagnostic unit costs $300,000 and can practically perform 10,000 examinations, its rate is $30 per examination. If theoretical capacity were 12,000 examinations, the rate would be $25, but that rate would fail to show the cost of unused capacity.

ABC is most useful when:

• overhead is substantial;
• services consume support activities differently;
• managers need process-level information; and
• cost drivers can be measured reliably.

ABC can become burdensome when the number of activities is excessive or when staff must enter unreliable time estimates. It also risks confusing correlation with causation. A driver should be selected because it explains resource consumption, not merely because it is convenient.

6. Time-driven activity-based costing

TDABC simplifies ABC by estimating two quantities: the cost of supplying capacity per unit of time and the time required for each activity.

The capacity cost rate is:

The cost of an activity is:

For patient i using several resources:

The capacity cost rate has units such as dollars per available clinical minute. Practical capacity should deduct breaks, training, meetings, leave, maintenance, and reasonable nonproductive time. If a clinic’s monthly cost is $90,000 and practical capacity is 3,600 clinical minutes, the rate is $25 per minute. A 40-minute visit therefore consumes $1,000 of capacity cost before separately traced supplies.

TDABC can use time equations to represent variation. For example:

Here, 12 is the base time in minutes; each indicator I equals 1 when the condition is present and 0 otherwise; and the added times represent an interpreter and a complexity feature. The equation is valid only if the organization has observed and validated these time increments.

TDABC makes unused capacity visible. If supplied capacity is 4,000 minutes but actual patient time is 3,200 minutes, 800 minutes are unused. At $25 per minute, unused-capacity cost is $20,000. It should not automatically be charged to patients, because doing so obscures operational capacity decisions.

7. Worked patient-level costing example

Consider an illustrative outpatient endoscopy episode. The organization uses TDABC for clinical labor and ABC for shared support activities.

Thus:

Suppose reimbursement is $1,050. The episode’s accounting contribution before service-line fixed costs is $146. If $100 of service-line capacity-sustaining cost is assigned, the full-cost margin becomes $46. These are different decision measures. The $146 contribution may be relevant to a short-run acceptance decision when capacity is available; the $46 full-cost margin is more relevant to long-run sustainability.

The example also illustrates why a charge-based allocation may be misleading. If all episodes were assigned overhead according to billed charges, a high-priced supply might receive a disproportionate share of administrative cost even though administrative work is driven mainly by encounters and documentation.

8. Service-line costing and cross-subsidization

A service line normally includes clinical revenue, direct clinical cost, allocated support cost, and capacity-sustaining cost. A simplified statement is:

For decision analysis, managers should also calculate contribution margin:

A service line may show a loss after full-cost allocation but generate positive contribution toward unavoidable hospital costs. Conversely, a service line may appear profitable under a crude allocation while consuming substantial scarce capacity.

Cross-subsidization occurs when one service, patient group, payer, or product is charged more than its resource cost to support another. It may be intentional—for example, using profitable elective services to sustain emergency care—or unintentional, arising from inaccurate allocation.

A practical cross-subsidy analysis should:

1. calculate patient-level or procedure-level cost using resource-based drivers;
2. compare cost with net revenue by payer and service;
3. separate avoidable, fixed, and capacity-sustaining costs;
4. examine margins under alternative allocation methods; and
5. assess whether differences reflect mission, access obligations, quality, or measurement error.

A simple subsidy indicator for service s can be expressed as:

A positive value indicates that revenue is below estimated resource-based cost. It does not establish that another service is subsidizing it, because the organization may be using philanthropy, public funding, reserves, or uncompensated capacity. Causal claims require analysis of funding flows and management decisions.

9. Integrated illustrative case: Riverside Medical Centre

The following is a fictional illustrative case.

Background

Riverside Medical Centre operates orthopaedics, cardiology, emergency medicine, and diagnostic imaging. Its finance department allocates most overhead using charges. Orthopaedics appears highly profitable, while emergency medicine appears to be loss-making. Clinicians argue that the results do not reflect actual resource use.

Problem

Orthopaedic procedures often use expensive implants but relatively little administrative time. Emergency patients may require multiple consultations, imaging studies, social-work intervention, and discharge coordination. A charge-based system assigns a large share of overhead to orthopaedics because implant charges are high, while it may understate the complexity of emergency episodes.

Analysis

The centre forms a multidisciplinary costing team. It defines the episode as the period from registration to discharge or transfer. Direct implant and drug costs are traced to patients. Operating-room costs are assigned using room minutes, imaging using examination minutes and modality-specific rates, and nursing using observed patient-care minutes supplemented by acuity categories. Registration and billing use encounter counts and documentation time.

For inpatient comparison, the team retains DRG and CMI information but does not treat DRG weight as actual patient cost. It develops pathway templates for hip replacement and selected emergency presentations. TDABC interviews and observations establish practical capacity rates for theatre staff, imaging technologists, and care coordinators.

Results and approach

The revised model shows three findings:

• Orthopaedics remains profitable for several procedures, but its margin is lower than previously reported because theatre capacity and post-acute coordination are recognized.
• Emergency medicine has positive contribution margin for some cases but a substantial structural funding gap because emergency readiness is maintained even when demand fluctuates.
• Within hip replacement, cost variation is driven more by length of stay, rehabilitation needs, and complications than by implant cost alone.

Management therefore rejects a simple “close the emergency service” conclusion. It negotiates public funding for readiness, redesigns discharge planning, reviews implant purchasing, and uses pathway variance reports for clinical improvement.

Learning points

The case demonstrates that costing is not merely an accounting exercise. Driver selection changes managerial interpretation. A credible system distinguishes reimbursement from resource consumption, contribution from full cost, and intentional mission subsidy from accidental cross-subsidy.

10. Evaluating cost accuracy

Cost accuracy is multidimensional. A system may reconcile perfectly to the general ledger yet provide poor patient-level information if allocation drivers are weak. Evaluation should address:

• Completeness: Are all material resources included?
• Traceability: Can a reported cost be linked to source records?
• Causal relevance: Does the driver explain resource consumption?
• Timeliness: Is information available before decisions are made?
• Consistency: Are definitions stable across periods and departments?
• Clinical validity: Do clinicians recognize the pathway and resource assumptions?
• Reconciliation: Do aggregated patient costs agree with departmental and organizational totals?
• Sensitivity: Do conclusions change materially under plausible alternative drivers?

Useful validation techniques include shadow costing, chart review, time observation, variance analysis, benchmark comparison, and sensitivity analysis. If a service-line conclusion changes from surplus to deficit when room minutes replace charges as the driver, management should investigate the reason rather than selecting the preferred result.

Costing systems should report confidence and limitations. A patient cost estimated from exact implant records and observed operating time has a different evidential basis from one allocated using average length of stay.

11. Governance, ethics, and regulation

Cost data can influence access, staffing, clinical autonomy, and payer negotiations. Governance is therefore essential.

First, patient-level data must be protected through role-based access, minimum necessary use, secure data exchange, and compliance with applicable privacy law. Second, costing definitions should be documented in a data dictionary. Third, changes in coding, DRG grouper rules, wage rates, or allocation bases should be version-controlled.

Ethically, cost information must not be used to deny clinically necessary care solely because an individual appears expensive. High cost may reflect severity, disability, language needs, housing insecurity, or unequal access to earlier treatment. Equity analysis should examine whether costing methods systematically make some populations appear less desirable.

Managers should also avoid gaming. If departments are judged on allocated cost, they may shift activity, delay documentation, or avoid complex patients. Performance systems should combine cost with outcomes, safety, access, experience, and equity.

12. Technology, analytics, and contemporary developments

Electronic health records, computerized provider-order entry, pharmacy systems, theatre-management systems, and time stamps enable more granular costing. Data warehouses can link clinical events to general-ledger transactions. Process mining can identify actual pathways and deviations from designed pathways.

Artificial intelligence may assist in predicting length of stay, identifying unusual cost patterns, extracting time intervals from records, or estimating resource requirements. However, predictive accuracy is not equivalent to causal validity. An algorithm trained on historical utilization may reproduce inequitable access patterns or label under-served patients as expensive. AI-generated cost estimates require human review, transparent features, monitoring for drift, and auditability.

Contemporary payment models—including bundled payments, accountable care, and value-based contracts—make episode and pathway costing more important. Organizations increasingly need longitudinal costing across hospital, outpatient, rehabilitation, and home care. This creates interoperability and attribution challenges: an apparent hospital saving may simply shift cost to another provider or to the patient.

13. Best-practice implementation roadmap

A practical implementation can proceed in stages:

1. Clarify purpose. Decide whether the system supports pricing, contracting, pathway improvement, budgeting, or external reporting.
2. Define boundaries. Specify episode start and end points, included services, and treatment of complications and follow-up.
3. Map processes. Engage clinicians to identify activities, handoffs, and sources of variation.
4. Trace high-value items. Begin with implants, drugs, procedures, imaging, and laboratory tests.
5. Select drivers carefully. Prefer causal, observable, stable drivers over convenient proxies.
6. Use practical capacity. Separate unused capacity from patient consumption.
7. Pilot and reconcile. Test one service line and reconcile outputs to the ledger.
8. Validate clinically. Review outliers and pathway assumptions with multidisciplinary teams.
9. Report multiple views. Show direct, contribution, and full cost where decision purposes differ.
10. Maintain the model. Review rates, pathways, coding rules, and capacity assumptions regularly.

The goal is not maximal complexity. It is decision-relevant accuracy at a sustainable cost of measurement.

Challenges and limitations

No costing method eliminates uncertainty. Patient-level systems may be expensive to implement and can create false precision. RVUs may reflect payment conventions rather than actual resource use. DRGs group cases but cannot capture every clinical difference. CMI is sensitive to coding and should not be used as a standalone efficiency measure. ABC requires many drivers, while TDABC depends on credible time estimates and practical-capacity assumptions.

Shared costs remain difficult. Leadership, education, research, emergency readiness, and unused capacity may benefit the whole institution without being caused by individual patients. Allocating them can support full-cost accountability but may distort operational decisions. Finally, cost is only one dimension of value. A low-cost intervention that worsens outcomes is not efficient care.

Chapter Summary

Patient, procedure, and service-line costing connect financial resources with clinical activity. Patient-level costing combines direct tracing with allocation of shared costs; procedure costing estimates the resources required by a defined intervention; and service-line costing evaluates the sustainability of a clinical portfolio. RVUs provide relative measures of service intensity, while CMI and DRGs support case-mix adjustment and benchmarking but do not automatically equal actual cost.

Clinical pathways provide an episode framework and expose variation. ABC assigns resource pools through activity drivers, whereas TDABC estimates capacity cost per minute and multiplies it by observed or estimated time. Practical capacity is important because unused capacity should be visible rather than silently charged to patients. Service-line analysis should distinguish contribution margin from full-cost margin and should investigate whether cross-subsidization is intentional, mission-related, or caused by weak measurement.

Reliable costing requires causal drivers, clinical validation, reconciliation, privacy protection, version control, and sensitivity analysis. Technology and AI can improve timeliness and granularity but cannot replace governance or clinical judgment. The best costing system is not the most elaborate system; it is the one whose assumptions are transparent and whose results support better decisions about value, access, quality, and sustainability.

Key Terms

Cost object: The patient, procedure, service, or other unit for which cost is measured.

Cost pool: A grouping of costs assigned through a common allocation logic.

Cost driver: A measurable factor used to explain or assign resource consumption.

Direct cost: A cost economically traceable to a specified cost object.

Indirect cost: A shared cost requiring allocation to a cost object.

Relative value unit (RVU): A relative measure of the resource intensity of a service.

Case-mix index (CMI): The average relative complexity of a set of inpatient cases.

Diagnosis-related group (DRG): A classification of clinically similar inpatient episodes used for payment and analysis.

Clinical pathway: An expected sequence and timing of care for a condition or procedure.

Activity-based costing (ABC): A method assigning costs through activities and their drivers.

Time-driven activity-based costing (TDABC): A method using capacity cost per time unit and estimated activity time.

Capacity cost rate: Cost of supplied capacity divided by practical capacity units.

Practical capacity: Capacity available after reasonable allowances for breaks, leave, training, and nonproductive time.

Contribution margin: Revenue less avoidable or variable costs.

Full-cost margin: Revenue less direct and allocated indirect costs.

Cross-subsidization: Support of one service or group by surplus generated elsewhere.

Cost variance: Difference between actual and expected cost.

Resource consumption: The quantity of labor, supplies, space, equipment, or services used.

Cost-to-charge ratio: A ratio used to convert charges into estimated costs; it is less resource-sensitive than activity-based methods.

Capacity-sustaining cost: Cost supporting the existence or readiness of a service rather than a specific encounter.

Review Questions

Short Answer Questions

1. Distinguish patient-level costing from service-line costing.
2. What is the purpose of a cost driver?
3. Why is a DRG not the same as a patient’s actual cost?
4. Define practical capacity in TDABC.
5. Distinguish contribution margin from full-cost margin.

Long Answer Questions

1. Explain how RVUs, CMI, and DRGs can support healthcare cost analysis, including their limitations.
2. Describe the stages involved in constructing a patient-level costing system.
3. Compare ABC and TDABC in terms of data requirements, strengths, and weaknesses.
4. Explain how clinical pathways can be used to analyze cost variation and improve care.
5. Discuss the governance and ethical risks of using patient-level cost data for performance management.

Analytical / Application Questions

1. A diagnostic unit costs $240,000 per month and has practical capacity of 8,000 examinations. Calculate its capacity cost rate and the capacity cost of a 35-minute examination if each examination consumes 35 minutes.
2. A service line reports revenue of $4.2 million, direct cost of $2.6 million, and allocated indirect cost of $1.9 million. Calculate its full-cost margin and explain why its contribution margin cannot be determined from these data alone.
3. A hospital’s orthopaedic margin changes substantially when overhead is allocated by charges rather than operating-room minutes. Design a sensitivity and governance review to determine whether the apparent cross-subsidy is real or an allocation artifact.

Multiple Choice Questions with Answers

Question 1

Which statement best describes a relative value unit (RVU)?

A. The exact dollar cost of a patient episode

B. A relative measure of the resource intensity of a service

C. The reimbursement amount paid by every payer

D. The average length of stay for a DRG

Correct Answer: B. A relative measure of the resource intensity of a service

Question 2

A department has costs of $120,000 and produces 6,000 RVUs. What is its cost per RVU?

A. $5

B. $15

C. $20

D. $720

Correct Answer: C. $20

Question 3

Which formula correctly represents a case-mix index?

A. Total revenue ÷ total cost

B. Total relative weights ÷ number of cases

C. Number of cases ÷ total relative weights

D. Total cost ÷ length of stay

Correct Answer: B. Total relative weights ÷ number of cases

Question 4

In TDABC, the capacity cost rate is calculated as:

A. Revenue ÷ number of patients

B. Cost of supplied capacity ÷ practical capacity units

C. Total charges ÷ total RVUs

D. Fixed cost ÷ reimbursement rate

Correct Answer: B. Cost of supplied capacity ÷ practical capacity units

Question 5

Why is practical capacity preferred to theoretical capacity in calculating a TDABC rate?

A. It excludes all fixed costs

B. It reflects reasonable nonproductive time and makes unused capacity visible

C. It guarantees that every patient receives the same cost

D. It converts revenue into cost

Correct Answer: B. It reflects reasonable nonproductive time and makes unused capacity visible

Question 6

A clinic has a capacity cost rate of $25 per minute. What capacity cost is assigned to a 40-minute visit, before separately traced supplies?

A. $65

B. $100

C. $625

D. $1,000

Correct Answer: D. $1,000

Question 7

Which is the strongest example of a direct patient cost?

A. Hospital chief executive salary

B. Building depreciation

C. An implant recorded for a specific operation

D. General information-technology support

Correct Answer: C. An implant recorded for a specific operation

Question 8

A positive subsidy gap defined as resource-based cost minus net revenue indicates that:

A. Revenue exceeds resource-based cost

B. Resource-based cost exceeds net revenue

C. The service has no fixed costs

D. The DRG weight is incorrect

Correct Answer: B. Resource-based cost exceeds net revenue

Question 9

Which measure is most appropriate for comparing the average complexity of inpatient cases?

A. Case-mix index

B. Contribution margin

C. Capacity cost rate

D. Cost-to-charge ratio

Correct Answer: A. Case-mix index

Question 10

Which statement about a DRG is most accurate?

A. It is always the exact cost of an individual patient

B. It is a classification category that may support payment and benchmarking

C. It measures only outpatient nursing time

D. It eliminates the need for clinical validation

Correct Answer: B. It is a classification category that may support payment and benchmarking