ICHOM Standard Set

ICHOM Standard Set in Value-Based Health Care

An ICHOM Standard Set fixes, for one medical condition, which outcomes every provider records in everyday practice, which instruments and time points are used, and which patient characteristics are captured so that results can be adjusted for case mix. The sets are published by the International Consortium for Health Outcomes Measurement (ICHOM), which grew out of Michael Porter's argument that health care should be judged by the outcomes achieved for patients relative to the cost of achieving them. For a health economist, such a set decides what outcome data will exist in registries and hospital systems, and therefore what real-world data can later support benchmarking, payment schemes or a cost-utility analysis. Porter, Larsson and Lee reported in 2016 that only 139 (7 per cent) of the 1,958 quality indicators in the US National Quality Measures Clearinghouse were actual outcomes, and only 32 (under 2 per cent) were patient-reported outcomes, which is the gap standard sets were designed to close.

Why value-based health care needed standard outcome sets

Porter defined value in health care in 2010 as health outcomes achieved per dollar spent and argued that achieving high value for patients must become the overarching goal of care delivery. Costs already had common accounting units, but outcomes were measured differently by each hospital, registry and specialty, so the numerator of the value ratio could not be compared across providers.

Porter, Larsson and Lee argued in their 2016 article in the New England Journal of Medicine that accelerating outcomes measurement requires a commitment to measuring a minimum sufficient set of outcomes for every major medical condition, with defined methods of collection and risk adjustment, standardised nationally and globally. They also observed that for many conditions death is a rare outcome whose measurement fails to separate excellent providers from merely competent ones. A standard set therefore reaches beyond mortality to the results patients notice, such as function, symptoms and return to work.

ICHOM uses two names for the same product. Its About page speaks of Standard Sets, and its main site describes them as Sets of Patient-Centered Outcome Measures.

What an ICHOM set contains

An ICHOM set has two halves. The outcome half lists what is measured, how and when, and the case-mix half lists the baseline information needed to compare like with like. ICHOM's reference guide for hip and knee osteoarthritis describes the second half as baseline conditions and risk factors to enable meaningful case-mix adjustment, together with high-level treatment variables for stratifying outcomes by type of treatment.

The hip and knee osteoarthritis set, reported by Rolfson and colleagues in 2016, shows the outcome half in practice. A working group representing 10 countries held 8 teleconferences and used a modified Delphi process, with a 67 per cent majority vote, to reach consensus on measures of joint pain, physical functioning, health-related quality of life, work status, mortality, reoperations, readmissions and overall satisfaction with the result of treatment. It also defined baseline characteristics and risk factors and recommended annual outcome measurement for all patients. ICHOM's reference guide for the set tracks function with the HOOS-PS or KOOS-PS and quality of life with the EQ-5D-3L or SF-12.

Sets are not limited to adults or to a single questionnaire. The overall paediatric health set, for example, assigns PROMIS parent proxy instruments to children aged 6 to 12 and other instruments to those aged 13 to 17. ICHOM's FAQ notes that developments in outcome measurement, such as new tools or updated versions, may prompt revision of a set.

How working groups agree a set

ICHOM describes each set as produced by a group of leading physicians, measurement experts and patients. The working group meets over months and votes on what to include, using structured consensus methods.

McNamara and colleagues, reporting the coronary artery disease set in 2015, described a working group that held a series of conference calls, 10 of which were followed by surveys on the decision points, and used a modified Delphi method to decide what to measure. ICHOM's About page records that it was founded in 2012 as a non-profit organisation by Michael Porter of Harvard Business School, Stefan Larsson of the Boston Consulting Group and Martin Ingvar of the Karolinska Institute.

ICHOM's FAQ says it prioritises the development of new sets on three factors, burden, clinical engagement and funding. Its About page reported 48 Standard Sets in October 2026, while an ICHOM introduction flyer dated March 2026 counted 47 Sets. The sets are published on the ICHOM website for anyone to see.

Where standard sets meet economic evaluation

An ICHOM Standard Set is built for comparing providers, but its data feed several kinds of economic analysis. The link is strongest when the set includes a preference-based instrument.

When a set lists the EQ-5D among its options, as the hip and knee set does, providers that choose it collect utilities at fixed time points as part of routine care. Those data can supply health state utility values for a model, or before and after measurements for a cost-utility analysis of a service change, without a separate trial. When a provider chooses the SF-12 instead, a mapping or a separate utility algorithm is needed before QALYs can be calculated, so the choice of instrument within a set matters for later economic use.

Standard sets also supply the outcome side of provider-level value comparisons. Set-based outcomes can be paired with costs per patient over the cycle of care, estimated for example by time-driven activity-based costing, to give the two halves of Porter's value ratio. Payers can tie payment to set-based outcomes in value-based purchasing schemes, and providers can use the same data for benchmarking against peers. Each of these uses depends on the case-mix variables, because unadjusted outcomes reward providers who treat healthier patients.

Illustrative worked example of a case-mix adjusted comparison

All numbers in this example are illustrative. Two hospitals each perform 200 hip replacements in a year and record the set's function outcome, counting a patient as improved when the score at 12 months passes a pre-agreed level. A risk model built on the set's case-mix variables, such as age, body mass index and baseline function, gives each patient a predicted probability of improvement, and the sum of those probabilities is the expected number improved.

1. Crude results. Hospital A has 150 improved patients out of 200, which is 75 per cent. Hospital B has 140 out of 200, which is 70 per cent. On crude results A looks better.

2. Observed to expected ratios.

$$R = \frac{O}{E}$$

where $R$ is the observed to expected ratio, $O$ is the observed number of improved patients and $E$ is the expected number given the hospital's case mix. Hospital A treats fitter patients, with $E$ equal to 160, and Hospital B treats patients with worse baseline function, with $E$ equal to 126.

$$R_A = \frac{150}{160} = 0.9375 \qquad R_B = \frac{140}{126} = 1.1111$$

where $R_A$ and $R_B$ are the ratios for Hospitals A and B, rounded to four decimal places.

3. Cost per improved patient. If an episode costs GBP 6,000 at Hospital A and GBP 6,300 at Hospital B, total costs are 200 × 6,000 = 1,200,000 and 200 × 6,300 = 1,260,000. Dividing by improved patients gives GBP 8,000 per improved patient at A (1,200,000 ÷ 150) and GBP 9,000 at B (1,260,000 ÷ 140).

After adjustment the ranking reverses on outcomes, since B achieves about 11 per cent more improvements than its case mix predicts while A achieves about 6 per cent fewer. The crude cost per improved patient still favours A, which shows why a value comparison that ignores case mix can reward the wrong provider. A fair cost comparison would also need costs adjusted for case mix, and the conclusion would depend on the risk model being well specified, which is why the set fixes the variables in advance.

How an ICHOM Standard Set differs from neighbouring ideas

An ICHOM Standard Set is easy to confuse with other kinds of outcome list. The boundaries turn on who uses the data and for what decision.

A core outcome set states the minimum outcomes to measure and report in research on a condition, so that trials can be compared and pooled, and its development follows its own methods. An ICHOM Standard Set serves everyday practice and provider comparison, fixes instruments and time points as well as domains, and adds case-mix variables for adjustment.

A single outcome measure such as a function score or the EQ-5D is one component of a set, not the set itself. A quality indicator used in performance management may be a process or structure measure, while a standard set is limited to outcomes and the variables needed to interpret them. A patient-reported outcome instrument often fills one or more domains of a set, but a set can also include clinician-reported outcomes such as complications or survival.

Limitations and common misreadings

Standard sets raise practical and analytical problems once they meet real services. Several are easy to overlook in an economic analysis that draws on set-based data.

• Many sets for one patient. Blom and colleagues harmonised the measures in all 28 sets published by 2020 and reported that consolidation reduced the measures used across the sets by 15 per cent, from 3,178 to 2,712. A patient with several conditions could otherwise be asked overlapping questions from several sets.
• Choice of instrument within a set. Where a set allows alternatives, such as the EQ-5D-3L or SF-12, providers using different options cannot be compared without a crosswalk, and only some options yield utilities directly.
• Adjustment is only as good as the variables. A set's case-mix list is agreed by consensus, and characteristics left off it cannot be adjusted for later.
• Consensus is not validation. A Delphi vote shows that a group agreed an outcome matters, not that the chosen instrument is responsive in a given setting or that a provider's score reflects its quality of care.
• Routine data are not trial data. Before and after comparisons from set-based registries lack a randomised control, so an economic evaluation built on them needs methods for confounding and missing follow-up.
• Counts change over time. ICHOM's About page gave 48 Standard Sets and its March 2026 flyer 47 Sets, so a cited total should carry its date and source.

Sources

• Blom MC, Khalid M, Van-Lettow B, Hutink H, Larsson S, Huff S, Ingvar M. Harmonization of the ICHOM quality measures to enable health outcomes measurement in multimorbid patients. Frontiers in Digital Health. 2020;2:606246. doi:10.3389/fdgth.2020.606246
• International Consortium for Health Outcomes Measurement. About us, FAQs and Sets of Patient-Centered Outcome Measures web pages, accessed October 2026. https://www.ichom.org/about-us/ and https://www.ichom.org/faqs/ and https://ichom.org/standard-sets
• International Consortium for Health Outcomes Measurement. ICHOM introduction flyer, March 2026. https://www.ichom.org/wp-content/uploads/2026/03/ICHOM-Introduction-Flyer.pdf
• International Consortium for Health Outcomes Measurement. Hip and knee osteoarthritis data collection reference guide, accessed October 2026. https://ichom.org/files/medical-conditions/hip-knee-osteoarthritis/hip-knee-osteoarthritis-reference-guide.pdf
• McNamara RL, Spatz ES, Kelley TA, et al. Standardized outcome measurement for patients with coronary artery disease: consensus from the International Consortium for Health Outcomes Measurement (ICHOM). Journal of the American Heart Association. 2015;4(5):e001767. doi:10.1161/JAHA.115.001767
• Porter ME. What is value in health care? New England Journal of Medicine. 2010;363:2477-2481. doi:10.1056/NEJMp1011024
• Porter ME, Larsson S, Lee TH. Standardizing patient outcomes measurement. New England Journal of Medicine. 2016;374:504-506. doi:10.1056/NEJMp1511701
• Rolfson O, Wissig S, van Maasakkers L, et al. Defining an international standard set of outcome measures for patients with hip or knee osteoarthritis: consensus of the International Consortium for Health Outcomes Measurement Hip and Knee Osteoarthritis Working Group. Arthritis Care & Research. 2016;68(11):1631-1639. doi:10.1002/acr.22868