| Verified against primary record | |
| Governing standard | The OECD and Joint Research Centre handbook, 2008, setting out ten construction steps[1] |
|---|---|
| Why normalisation is unavoidable | Indicators in a data set often have different measurement units[1] |
| Audit practice | Over 100 statistical audits of composite indicators by the Joint Research Centre’s competence centre[2] |
| Records read | The handbook, two audit sources and two sector index publications, 2 October 2026 |
| Provider-supplied | |
| Leading sector index | Describes a rigorous eight-step procedure across three dimensions and 34 indicators[3] |
| Bands apply only to the rows beneath them. The provider-supplied row reports what the publisher states about its own method, which is not independently verifiable from its published pages. | |
Medical tourism indexes are composite indicators: single scores assembled from many separate measurements of a country, city or hospital. Composite indicator construction is a mature statistical discipline with a published standard and an established audit practice, which makes it unusually easy to assess whether a given index has followed it.
The ten steps
The reference work in this field is a handbook jointly produced by the OECD and the European Commission’s Joint Research Centre, which sets out ten steps: a theoretical framework, data selection, imputation of missing data, multivariate analysis, normalisation, weighting and aggregation, uncertainty and sensitivity analysis, returning to the data, links to other indicators, and visualisation.[1] It has become the reference report for practitioners, and the competence centre that maintains it has carried out over 100 statistical audits of composite indicators.[2] The stated objective of such an audit is to ensure the transparency of the composite indicator methodology and the reliability of its outcomes.[4]
Four of the ten steps are where indexes in this sector tend to fail.
Normalisation. It is required prior to any data aggregation, because the indicators in a data set often have different measurement units, and the method chosen changes the result. Of the common minimum to maximum method the handbook warns that extreme values or outliers could distort the transformed indicator.[1]
Imputation. The handbook’s warning here is worth quoting at length, because filling gaps is routine and rarely disclosed. The idea of imputation is described as both seductive and dangerous: seductive because it can lull the user into the pleasurable state of believing that the data are complete after all, and dangerous because it lumps together situations where the problem is sufficiently minor that it can be legitimately handled in this way and situations where standard estimators applied to real and imputed data have substantial bias.[1]
Weighting. Regardless of which method is used, weights are essentially value judgements. Equal weighting does not mean no weights but implicitly implies that the weights are equal, and if variables are grouped into dimensions which are further aggregated into the composite, then applying equal weighting to the variables may imply an unequal weighting of the dimension.[1]
Uncertainty and sensitivity analysis. A combination of uncertainty and sensitivity analysis can help gauge the robustness of the composite indicator and improve transparency, and ideally all potential sources of uncertainty should be addressed.[1] This is the step that converts a single score into a defensible one, and it is the step most often absent.
What the handbook says an index can do badly
The same handbook lists the drawbacks of composite indicators, and they read as a description of how such scores are used in destination marketing. A composite may send misleading policy messages if poorly constructed or misinterpreted, may invite simplistic policy conclusions, may be misused for example to support a desired policy if the construction process is not transparent or lacks sound statistical or conceptual principles, and may disguise serious failings in some dimensions.[1]
The last of those is structural rather than a matter of bad faith. Aggregation averages, so a very weak score on one dimension can be offset by strong scores elsewhere and disappear into the total.
What an audited index publishes
A worked benchmark is useful. The statistical audit attached to one well-documented international index publishes the distributional tests applied, with skewness above 2.25 and kurtosis above 3.5 triggering winsorisation or logarithmic transformation; principal component analysis per pillar with an eigenvalue threshold above 1.0; an internal consistency figure, the Input Sub-Index reliability measured by Cronbach’s alpha reported as very high at 0.94, well above the 0.70 threshold; 4,000 Monte Carlo simulations varying three modelling choices simultaneously; 90 per cent confidence intervals on the published ranks; data envelopment analysis as an alternative benchmark; and a documented review chain. It also notes that weights there function as scaling coefficients rather than as statements of importance.[5]
That is the standard of disclosure against which any index in this sector can be read.
What the leading sector index publishes
The most widely cited index in this field is published by a private research centre alongside an industry association. Its own pages describe a rigorous eight-step methodological, statistical and index construction procedure, claim psychometrically sound measurement, state that the index comprises three dimensions and 34 indicators, say that it focuses on supply-side factors or pull factors, and note that the results are drawn from American feedback.[3]
Checked against the ten steps on 2 October 2026, those pages did not publish numerical weights, named data sources for any indicator, a survey sample size, sampling frame or fieldwork year, a normalisation formula, an aggregation formula, any uncertainty or sensitivity analysis, any external audit, or a downloadable technical report. Six of the ten steps therefore cannot be assessed from what is public.
There is also an internal tension worth noting on its own terms. An index that describes itself as measuring supply-side factors while drawing its results from the feedback of one nationality is using perception data to score facts such as accreditation status, and those are different quantities.
The academic literature is only partially better. A peer-reviewed index proposal for this sector publishes its structure of three dimensions and eleven sub-dimensions and the analytic hierarchy process weights derived from a panel of fifteen experts, which is more than the commercial index discloses. It publishes no data sources, no normalisation method, no aggregation formula, no sensitivity analysis and no reference to the governing handbook, and its authors note the limitation that the expert panel sat within one cultural context.[6]
See also
- Medical tourism rankings, how the resulting ordered lists behave
- Medical tourism market forecasts, the same disclosure test applied to projections
- Medical tourism statistics, the quality of the underlying indicators
- Patient satisfaction in medical tourism, a common index input and its limits
References
- Nardo M, Saisana M, Saltelli A, Tarantola S, Hoffman A, Giovannini E. Handbook on constructing composite indicators: methodology and user guide. OECD and European Commission Joint Research Centre, 2008, ISBN 978-92-64-04345-9, JRC47008. Verified against primary record: the ten steps, normalisation, imputation, weighting, sensitivity and drawbacks sections read. Retrieved 2 October 2026.
- European Commission Joint Research Centre. Competence Centre on Composite Indicators and Scoreboards. Verified against primary record: audit count and handbook status read. Retrieved 2 October 2026.
- International Healthcare Research Center and Medical Tourism Association. Medical Tourism Index methodology, with the research centre’s own research page. Provider-supplied: publisher’s description of its own method; both pages read for construction detail on 2 October 2026.
- European Commission Joint Research Centre. Competence Centre on Composite Indicators and Scoreboards: five-year review 2021-2025. JRC144475. Verified against primary record: stated audit objective read. Retrieved 2 October 2026.
- European Commission Joint Research Centre. Joint Research Centre statistical audit of the Global Innovation Index 2023. Verified against primary record: full audit methodology read as a disclosure benchmark. Retrieved 2 October 2026.
- Lee M, Li C. A study on the development of a medical tourism competitiveness index. International Journal of Environmental Research and Public Health, 2019;16(22):4579. Independently reported: peer-reviewed article read for published construction detail. Retrieved 2 October 2026.
Sourcing note: the handbook, the competence centre pages, the statistical audit and the two sector index publications were opened and read on 2 October 2026. The finding about the leading sector index is scoped to what its two public pages contained on that date, and is a statement about public auditability rather than a claim that no such analysis exists internally. No index scores or positions are reproduced in this entry. The audited international index is cited only as a benchmark for disclosure, and its own results are not reported here.