Perspective

Perspective in Health Economic Evaluation

Perspective determines whose costs and outcomes count in an economic evaluation, which is why two analyses of the same intervention can reach different conclusions. The perspectives most often used are the healthcare-system, payer, patient, societal, employer and provider viewpoints, and the NICE manual (PMG36), the Second Panel on Cost-Effectiveness in Health and Medicine and the Dutch guideline each require a different one. Because perspective changes the costs and outcomes included, the choice should be justified and reported. Productivity costs and double counting need particular care, since that is where the main perspectives diverge most.

Why perspective matters

Perspective defines the viewpoint from which costs and outcomes are identified, measured and valued. It affects which people, organisations and sectors are represented in the analysis. Two evaluations of the same intervention can therefore reach different results because they adopt different perspectives.

• A healthcare-system perspective includes consequences borne by the healthcare system.
• A payer perspective includes consequences relevant to a specified insurer, government programme or other payer.
• A patient perspective includes costs and outcomes experienced directly by patients.
• A societal perspective aims to include relevant consequences regardless of who experiences them.
• An employer perspective includes work-related costs and consequences borne by employers.
• A provider perspective includes resources, expenditure and operational consequences affecting healthcare providers.

Perspective should be specified before resource use, costs and outcomes are selected. Choosing the perspective after seeing the results can introduce bias by favouring the viewpoint that produces the preferred conclusion.

How common perspectives differ

Each perspective establishes a different analytical boundary. The appropriate perspective depends on the decision problem, the decision-maker and any applicable reference case or methodological guidance. The labels should not be treated as interchangeable.

A perspective label does not prove that every relevant consequence has been included. The evaluation should state the actual categories included and excluded.

Perspectives required by NICE, the Second Panel and the Dutch guideline

Reference cases fix the perspective so that analyses are comparable. The three sets of rules below show how far practice varies, from a health and social care budget in England to a full societal perspective in the Netherlands. Each rule is taken from the current text of the guidance named.

NICE (England). The NICE manual (PMG36) states two separate perspectives in its reference case. Section 4.2.7 sets the perspective on outcomes as all relevant health effects, whether for patients or, when relevant, other people (mainly carers), and the perspective on costs as that of the NHS and personal social services (PSS). Section 4.2.9 gives the reason. NICE does not set the NHS budget, and its evaluations aim to guide the efficient use of available NHS and PSS resources, so productivity costs are not included. Section 4.3.17 asks for evidence that a condition has a substantial effect on carers' health-related quality of life before carer health effects are presented.

The same manual allows some wider consequences outside the reference case. Under section 4.4.23, costs paid by patients may be included when the NHS or PSS reimburses them, unreimbursed patient costs may be presented separately, and productivity costs may be presented separately when they could be a critical component of the value of the technology. Section 4.4.24 allows the time of family members or friends to be costed, even under an NHS and PSS perspective, when that care might otherwise have been provided by the NHS or PSS, with such analyses shown separately. Sections 4.2.10 and 4.4.22 allow costs and benefits to other government bodies only when agreed with the Department of Health and Social Care, reported separately from the reference-case analysis.

Second Panel on Cost-Effectiveness in Health and Medicine. The Second Panel recommended that every cost-effectiveness analysis report two reference case analyses, one from a healthcare sector perspective and one from a societal perspective. It also recommended an impact inventory, a structured table of consequences inside and outside the formal healthcare sector, intended to make the scope and boundaries of the two analyses explicit.

Zorginstituut Nederland (the Netherlands). The Dutch guideline (version 2024) requires economic evaluations to be conducted and reported from the societal perspective, including all relevant costs and benefits regardless of who bears the costs or receives the benefits (section 2.2). Other perspectives, such as the healthcare perspective, may be presented as scenario analyses. Section 3.2 lists the cost categories as costs within healthcare, costs to patients and families such as informal care, and costs in other sectors, including productivity losses, which must be valued with the friction cost method.

How perspective changes the costs included

The same resource or expenditure may be included under one perspective and excluded under another. The analyst must determine who bears each cost and whether it falls within the stated analytical boundary. Transfers between parties should be distinguished from resource costs when that distinction matters to the analysis.

The following examples show how the boundary shifts.

• A patient copayment is a cost from the patient perspective.
• A reimbursed medicine is an expenditure from the payer perspective.
• Unpaid caregiver time may be included under a societal or household perspective.
• Productivity losses may be included under a societal or employer perspective.
• Hospital staff time may be included under healthcare-system and provider perspectives.
• Travel costs may be excluded from a healthcare-system perspective but included from a patient or societal perspective.

The analysis should avoid double counting when the same payment appears from more than one viewpoint. A patient payment and a payer reimbursement may finance the same healthcare service but represent different financial flows.

Productivity costs raise a second double-counting question. The original Panel on Cost-Effectiveness in Health and Medicine, often called the Washington Panel, treated lost productivity as a health effect captured in the QALY rather than as a cost, on the grounds that counting it in both places would count it twice. Empirical work, such as a survey of patients with rheumatic disorders in four western European countries, has not fully supported that position, and guidelines now differ. NICE excludes productivity costs from its reference case, while the Dutch guideline includes them in the base case. An analysis that includes productivity costs should state the valuation method, such as the human capital approach or the friction cost method, and whether the health-state values used could already reflect lost income.

How perspective changes the outcomes included

Perspective can also affect which outcomes are considered relevant. Most health economic evaluations include patient health outcomes, but broader perspectives may include consequences for caregivers, families, employers or other sectors. These wider outcomes should be measured and reported separately when combining them would obscure who experiences each effect.

• A patient perspective may include treatment burden, time costs and out-of-pocket consequences.
• A healthcare-system perspective may focus on patient health outcomes produced by healthcare resources, although some reference cases, such as NICE's, also count carers' health effects.
• A societal perspective may include caregiver health, productivity and consequences outside healthcare.
• An employer perspective may include work absence, presenteeism and return to work.
• A provider perspective may include service capacity, waiting times and implementation effects.

Including a wider range of outcomes does not automatically improve an evaluation. Each outcome must be relevant to the decision problem, measured credibly and included without overlap.

Selecting the appropriate perspective

The selected perspective should match the purpose of the evaluation and the budget and objectives of the decision-maker. A body responsible for one budget needs to know the consequences for that budget, which is the reason NICE gives for its NHS and PSS perspective on costs. Where the decision-maker is concerned with all consequences regardless of who bears them, as in the Dutch societal perspective, the analysis needs the consequences that fall on patients, families and other sectors as well. National or institutional reference cases may prescribe a primary perspective to improve consistency across submissions. A secondary perspective can be reported when important consequences fall outside the required primary viewpoint.

A clear selection process follows seven steps.

1. Define the decision-maker. Identify the person or organisation using the evaluation.
2. State the decision problem. Specify the population, intervention, comparator and decision being considered.
3. Check applicable guidance. Determine whether a reference case requires a particular perspective.
4. Identify affected parties. List the patients, caregivers, payers, providers, employers and sectors that may experience material consequences.
5. Select the primary perspective. Choose the viewpoint that directly supports the decision.
6. Consider secondary analyses. Add another perspective when excluded consequences could materially change interpretation.
7. Document inclusions and exclusions. Explain which cost and outcome categories enter each analysis.

The primary perspective should not be broadened or narrowed without explanation. When several perspectives are reported, their results should remain clearly separated.

Worked example (illustrative) with one treatment and two cost totals

This illustrative example uses invented figures to show how perspective alone can change a result. A treatment reduces hospital use but requires patients to travel regularly to a specialist clinic. It also allows patients to return to paid work sooner and reduces the time caregivers spend providing care.

• The healthcare-system perspective includes the treatment, clinic visits and avoided hospital care.
• The payer perspective includes reimbursed treatment and services covered by the payer.
• The patient perspective includes travel, out-of-pocket payments and patient time.
• The societal perspective includes healthcare resources, patient costs, caregiver time and productivity consequences.
• The employer perspective includes changes in absence and workplace productivity.

The table gives the incremental cost per patient compared with usual care. The health gain is the same under every perspective, at 0.05 QALYs per patient.

From the healthcare-system perspective, the incremental cost is £2,400 plus £600 minus £1,800, which is £1,200. Dividing by the health gain gives £1,200 / 0.05 = £24,000 per QALY gained. From the societal perspective, the incremental cost is £1,200 plus £300 minus £1,200 minus £400, which is -£100, meaning the treatment saves £100 per patient and also improves health. It therefore dominates usual care.

The two results answer different questions. A NICE submission would present the £24,000 per QALY figure as its reference case, with the productivity gain and unreimbursed patient costs, if presented at all, shown separately, while a Dutch submission would present the societal result as its base case. The patient's own position is different again. Patients bear the £300 of travel and time, and how much of the £1,200 productivity gain reaches them depends on who bears lost earnings, for example through sick pay arrangements. If the QALY gain partly reflects patients' valuation of regaining income, adding the £1,200 productivity gain as a cost saving could count part of the same benefit twice.

Using more than one perspective

Reporting more than one perspective can show whether a decision shifts costs or benefits between groups. This is especially useful when an intervention reduces healthcare expenditure but increases patient costs, or when it creates benefits outside the healthcare sector. Separate analyses are usually clearer than combining incompatible viewpoints into one total.

A multiple-perspective analysis should do each of the following.

• Apply the same population, intervention, comparator and time horizon wherever possible.
• Identify each cost and outcome category consistently.
• Show which party bears each consequence.
• Prevent the same resource or outcome from being counted more than once.
• Report results separately for each perspective.
• Explain why differences between perspectives arise.

Results from different perspectives should not be compared as if they were generated from identical analytical boundaries. The impact inventory recommended by the Second Panel is one structured way to show which consequences each analysis includes.

Common mistakes when applying perspective

An evaluation becomes difficult to interpret when its stated perspective does not match the costs and outcomes actually included. These inconsistencies can alter incremental results and lead readers to misunderstand who gains, who pays and which resources are affected.

Common mistakes include the following.

• The evaluation states a societal perspective but includes only healthcare costs.
• The evaluation includes productivity losses without identifying the valuation method.
• The evaluation includes patient payments and full healthcare costs without checking for overlap.
• The evaluation excludes caregiver consequences without explaining why they are irrelevant.
• The evaluation changes perspective between cost categories.
• The evaluation combines results from different perspectives into one unexplained total.
• The evaluation uses a payer perspective without identifying the payer.
• The evaluation selects its perspective after examining the results.

These problems should be addressed through a prespecified analytical plan and a transparent cost-and-outcome inventory.

What should be reported

Transparent reporting allows readers to determine whether the selected perspective fits the decision problem. It also makes clear which consequences are represented and which remain outside the analysis. The report should provide enough detail for the analytical boundary to be reproduced.

A report should state the following items.

• The primary perspective and why it was selected.
• The decision-maker associated with the perspective.
• Any reference case or guidance governing the selection.
• Every included cost category.
• Every included outcome category.
• The parties bearing each cost or receiving each outcome.
• Material categories that were excluded.
• Any secondary perspectives and why they were analysed.
• Methods used to value patient time, caregiver time or productivity.
• Steps taken to prevent double counting.
• The effect of perspective on incremental costs, outcomes and conclusions.
• Limitations created by the selected analytical boundary.

Perspective is a substantive design choice that changes which consequences an analysis can show. Decision-makers should consider whether consequences excluded from the primary analysis remain important to patients, caregivers, providers or society.

Sources

• Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes. 4th edition. Oxford: Oxford University Press; 2015.
• National Institute for Health and Care Excellence. NICE technology appraisal and highly specialised technologies guidance: the manual (PMG36). Published 31 January 2022, last updated 13 October 2026. Sections 4.2.7 to 4.2.10, 4.3.17 and 4.4.22 to 4.4.24.
• Sanders GD, Neumann PJ, Basu A, Brock DW, Feeny D, Krahn M, et al. Recommendations for conduct, methodological practices, and reporting of cost-effectiveness analyses: Second Panel on Cost-Effectiveness in Health and Medicine. JAMA. 2016;316(10):1093-1103.
• Zorginstituut Nederland. Richtlijn voor het uitvoeren van economische evaluaties in de gezondheidszorg (versie 2024). Zorginstituut Nederland; 16 January 2024. Sections 2.2 and 3.2.
• Knies S, Boonen A, Severens JL. Do the Washington Panel recommendations hold for Europe: investigating the relation between quality of life versus work-status, absenteeism and presenteeism. Cost Effectiveness and Resource Allocation. 2014;12:24.