| Verified against primary record | |
| When it occurs | When individuals or groups in a study differ systematically from the population of interest[1] |
|---|---|
| Records read | Four bias reference entries, an epidemiology textbook chapter and three medical travel studies, 9 October 2026 |
| Independently reported | |
| Representativeness in this field | It is difficult to find representative samples of patients who travel for care[2] |
| Two national counting methods | Thailand counts foreigners obtaining care at hospitals; Singapore uses airport exit polls[2] |
| A documented ascertainment limit | Five case series on bariatric surgery travel included only emergency, urgent or acute admissions[3] |
| Bands apply only to the rows beneath them. The application of general bias constructs to medical travel is this entry’s reasoning, not a sourced claim. | |
Selection bias is the reason most published findings about medical travel describe a narrower group of people than they appear to. It occurs when individuals or groups in a study differ systematically from the population of interest, producing a systematic error in an association or outcome.[1] In this field it is not an occasional flaw but a property of nearly every available data source.
Where the sample comes from decides what it measures
Four recognised mechanisms account for most of the problem, and each maps onto a specific way medical travel data is collected.
Volunteering. Participants volunteering to take part in a study intrinsically have different characteristics from the general population, with the risk that they are not representative of it.[4] A satisfaction form offered at discharge is answered by those willing to answer. One caution belongs with this: the reference entry’s claim about volunteers being healthier is made of females specifically, citing one source, and should not be generalised.
Referral. Referral of any group of unwell people from primary to secondary to tertiary care causes an increase in the concentration of rare cases, more complex cases or people with worse outcomes.[5] A specialist destination hospital sits at the end of such a chain, so its caseload is not the caseload of the condition.
Self-selection and the healthy worker analogy. A standard epidemiology text records that selection bias will occur with volunteers, which it names self-selection bias, and separately that only relatively healthy people are able to remain in employment, which it calls the healthy worker effect.[6]
It is tempting to posit a healthy traveller effect by analogy, since funding, researching, organising and flying to elective surgery abroad selects for people with money, information and the physical capacity to travel. Searches for this entry did not surface any established epidemiological term of that name, so it is not presented as one here. The reasoning is this site’s, built from adjacent constructs, and should be read as analysis rather than as a sourced finding.
Incomplete follow-up. The same text makes an important conditional point: non-response, refusal to participate and withdrawals cause bias only if the degree of incompleteness differs between the groups being compared.[6] Loss of participants is not automatically fatal, which is why the question in patient attrition in medical tourism studies is differential loss rather than total loss.
Counted at the destination, counted at the gate
National counting methods illustrate the mechanism at the level of official statistics. One review records that Thailand counts the number of foreigners obtaining medical care at hospitals, and that Singapore collects data on medical travel using exit polls at the airport.[2]
Both capture the traveller at or before departure from the destination. Neither can observe anything that happens after the patient goes home, and the same review notes that complications may arise after medical travellers return home.[2] A measurement system that closes at the airport cannot produce an outcome, only an arrival.
The mirror-image bias at the home end
Research conducted in the departure country has the opposite distortion, and it is unusually well documented. A commissioned review of complications and costs falling on the United Kingdom’s health service records that its evidence is limited to the cases that have been reported in the literature, often from patients who presented as emergencies, and that five case series reporting on bariatric surgery tourism only included emergency, urgent or acute admissions. It states the consequence directly: case numbers are under-reported, and costs are under-estimated.[3]
So destination-side data over-represents the satisfied and the still-present, and home-side data over-represents the acutely unwell. Neither describes the population of people who travelled, and averaging them does not produce it.
How thin the primary base is
One reason selection bias dominates this literature is that so little of it collects data at all. A scoping review of the patient experience of medical tourism found that only a small minority of sources reported on empirical studies that involved the collection of primary data, which it put at five.[7] That review is from 2010 and describes the literature of its time rather than the present state of the field, but the structural point survives: where primary data is scarce, each study’s sampling route carries disproportionate weight in what the field believes.
The broader verdict from the same period is that data are largely limited to anecdotal projects from the perspective of a single country, and that it is difficult to find representative samples of patients who travel for care.[2]
See also
- Medical tourism surveys, the three sampling frames and what each can support
- Patient attrition in medical tourism studies, loss after recruitment
- Case studies in medical tourism research, the design most exposed to this problem
- Medical tourism research methods, which designs are feasible here
References
- Catalogue of Bias Collaboration, Nunan D, Bankhead C, Aronson JK. Selection bias. Catalogue of Bias, 2017. Independently reported: academic reference resource opened and read. Retrieved 9 October 2026.
- Ruggeri K, Zalis L, Meurice CR, Hilton I, Ly TL, Zupan Z, Hinrichs S. Evidence on global medical travel. Bulletin of the World Health Organization, 2015;93:785-789. Independently reported: peer-reviewed paper in a World Health Organization journal. Retrieved 9 October 2026.
- England C, Bromham N, Needham-Taylor A, Hounsome J, Gillen E, Ingram BJ, Davies J, Edwards A, Lewis R. Complications and costs to the UK National Health Service due to outward medical tourism for elective surgery, a rapid review. Health and Care Research Wales Evidence Centre; published as BMJ Open 2026;16(1):e109050, 13 January 2026. Independently reported: publicly commissioned review, preprint full text read. Retrieved 9 October 2026.
- Catalogue of Bias Collaboration, Brassey J, Mahtani KR, Spencer EA, Heneghan C. Volunteer bias. Catalogue of Bias, 2017. Independently reported: academic reference resource opened and read. Retrieved 9 October 2026.
- Catalogue of Bias Collaboration, Bankhead C, Nunan D, Aronson JK. Referral filter bias. Catalogue of Biases, 2019. Independently reported: academic reference resource opened and read. Retrieved 9 October 2026.
- dos Santos Silva I. Cancer Epidemiology, Principles and Methods, chapter 13, interpretation of epidemiological studies. International Agency for Research on Cancer, 1999. Verified against primary record: chapter opened and read on the agency’s publications site. Retrieved 9 October 2026.
- Crooks VA, Kingsbury P, Snyder J, Johnston R. What is known about the patient’s experience of medical tourism? A scoping review. BMC Health Services Research, 2010;10:266. Independently reported: peer-reviewed scoping review, abstract read. Retrieved 9 October 2026.
Sourcing note: the four bias reference entries, the epidemiology chapter and the three medical travel sources were opened and read on 9 October 2026, the scoping review at abstract level only. The transfer of volunteer, referral and healthy worker constructs to medical travel is this entry’s own reasoning; none of those sources discusses travel. No established term for a healthy traveller effect was located, and the idea is therefore presented as analysis rather than as an accepted construct. The 2010 scoping review’s count describes the literature of its time.