| Verified against primary record | |
| What a discharge is | A hospital patient being formally released after an episode of care[1] |
|---|---|
| European residence variable | Non-residents among all hospital discharges, per cent, collected since 2010[2] |
| US residence variable | Patient state postal code; no foreign country codes exist in the field[3] |
| Records read | Four statistical agency sources and two national reports, 2 October 2026 |
| Independently reported | |
| England | The NHS does not routinely or systematically capture data on use by overseas visitors and migrants, as assessed in 2013[4] |
| Bands apply only to the rows beneath them. No national discharge collection recording patient country of residence was located for this entry. | |
Hospital discharge data is the largest and most systematic body of hospital activity data that exists, and it is the source most often assumed to be able to count foreign patients. It mostly cannot, for reasons that are visible in the structure of the datasets themselves.
What a discharge record is
A hospital discharge occurs when a hospital patient is formally released after an episode of care, and an in-patient is a patient who is formally admitted and stays for a minimum of one night or more than 24 hours.[1] The record is generated by the act of discharge, not by the arrival of a person, which is the origin of the counting problem described in international patient visits and unique patients.
European figures come from hospital administrative data collected through a joint questionnaire operated by Eurostat, the OECD and the WHO since 2010, with Commission Regulation 2022/2294 adopted in November 2022 as the legal basis for the current collection.[2] The standard variables are discharges by age, sex and disease.[5]
The European non-resident percentage, and what it does not define
One collected metric does bear directly on this subject. Eurostat’s methodology page lists, among the indicators gathered, the share of non-residents among all hospital discharges, expressed as a percentage.[2] That is the closest thing to an official cross-border hospital use indicator in Europe.
Two problems limit what it can support. The first is that no definition of resident or non-resident appears anywhere in the methodology documentation examined for this entry. The second is comparability, which the agency itself states plainly: the comparability of data across EU member states is limited by the fact that national data is subject to the way in which information is available.[2] The metadata for the discharge dataset makes the same point about health system organisation, and records that the collection rests on a gentlemen’s agreement rather than on compulsion for every element.[5]
The country notes make the inconsistency concrete. Germany and Austria state that residents and non-residents are both included. Romania notes that a person could be included multiple times for every episode of discharge. There is no standardised wording across countries for how residence is established.[6]
The United States records a state, not a country
The American inpatient databases are large and detailed, and they are structurally unable to identify an international patient. The patient residence variable in the state inpatient databases is labelled patient state postal code. Where the data source provided the state of the patient’s residence it is assigned to the reported state, and otherwise it is assigned by mapping the patient’s postal code to a state.[3] The field holds United States state codes. There is no foreign country value.
Even domestic travel is imperfectly recorded: out-of-state patients were incorrectly assigned to the reporting state for Georgia through 2010 and for North Carolina through 2009.[3]
The national sample adds a second limitation. It contains discharge-level records, not patient-level records, so individual patients who are hospitalised multiple times in one year may be present multiple times, and there is no uniform patient identifier available that allows a patient-level analysis.[7] It also eliminates state and hospital identifiers, which removes the geography that a travel analysis would need.
England did not collect it, and modelled instead
The clearest statement of what happens when a government needs this number and the hospital data cannot produce it comes from a 2013 study commissioned by the Department of Health. It records that the NHS does not routinely or systematically capture data on the use overseas visitors and migrants make of the NHS in England, and that the systems are not centrally managed and are run differently by trusts.[4]
Unable to count, the study modelled, building a top-down estimate from the 2011 Census and the 2012 International Passenger Survey. Its central estimate for deliberate health tourism was 60 to 80 million pounds within a range of 20 to 100 million, accompanied by the authors’ own instruction that these numbers should be used with caution.[4] The spread of the range is the honest measure of what the underlying records could support.
See also
- Administrative data in medical tourism research, the general case this is an instance of
- International patient visits and unique patients, why discharge counts exceed people
- Medical travelers and resident foreign patients, the distinction the residence field is meant to make
- Medical tourism statistics, the overview of available measurement
References
- Eurostat. Hospital discharges and length of stay statistics. Statistics Explained, 2023 reference data. Verified against primary record: statistical authority page opened and read. Retrieved 2 October 2026.
- Eurostat. Healthcare non-expenditure statistics, methodology. Statistics Explained. Verified against primary record: collected indicator list, legal basis and comparability caveat read in full. Retrieved 2 October 2026.
- Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project. PSTATE, patient state postal code. State Inpatient Databases documentation. Verified against primary record: variable documentation opened and read. Retrieved 2 October 2026.
- Prederi, for the Department of Health. Quantitative assessment of visitor and migrant use of the NHS in England. Published 3 October 2013. Independently reported: government-commissioned study. Retrieved 2 October 2026.
- Eurostat. Hospital discharges and patient mobility, ESMS metadata, hlth_act_h. Verified against primary record: quality and comparability metadata opened and read. Retrieved 2 October 2026.
- Eurostat. Country-specific notes on hospital discharges. ESMS annex. Verified against primary record: national notes read for residence treatment. Retrieved 2 October 2026.
- Agency for Healthcare Research and Quality. Introduction to the HCUP National Inpatient Sample, 2021. Verified against primary record: database documentation opened and read. Retrieved 2 October 2026.
Sourcing note: the European definitions, methodology, metadata and country annex, the American variable documentation and database introduction, and the English 2013 assessment were opened and read for this entry on 2 October 2026. No national hospital discharge collection recording patient country of residence was located. That is stated as a search result rather than as a finding that none exists.