Original ResearchName analysis to classify populations by ethnicity in public health: Validation of Onomap in Scotland
Section snippets
Background
Health inequalities exist between ethnic minorities and indigenous populations.1, 2, 3, 4, 5, 6, 7 In order to address the underlying causes of these inequalities, it is essential to systematically identify and classify individuals into population groups defined by ethnicity. To the authors’ knowledge, there are currently limited means by which to identify such groups between decennial censuses. Information on ethnicity is usually collected at the level of the individual, typically being
Methods
A diagnostic accuracy study was carried out to evaluate the Onomap-assigned classification of a person’s cultural ethnic or linguistic origins with three independent reference datasets that contained information on migration origin or ethnicity collected in Scotland and Poland:
- ●
birth registration data collected between 2004 and 2008 in the region of Lothian, Scotland;
- ●
the City of Edinburgh Council pupil census data collected between 2005 and 2008;
- ●
a registration database of Polish healthcare
Birth registration database
Onomap was unable to classify 307 records (0.4%). The most common reason for this was names not present in the dictionary (n = 231, 75%). There was greater likelihood of fathers’ names being classified by Onomap than mothers’ names (unclassified fathers’ names: n = 142, 0.4%; unclassified mothers’ names: n = 185, 0.5%). This difference was statistically significant (0.09%, 95% CI 0.08–0.1%; P < 0.05).
Parents were predominantly classified in the British group (86.8%), with South Asians and
Discussion
Onomap Version 2 is a quick, effective and user-friendly tool for identifying population groups in health-related and educational datasets, categorizing populations into a variety of Onomap ethnic groups and languages of origin.
There are some limitations intrinsic to the nature of the three independent datasets evaluated. Within the birth registration dataset, country of birth of the parents is not obviously representative of a person’s ethnicity, especially amongst established migrant
Application to public health practice
Similarly to other name-based methodologies, Onomap may be most useful when looking at population health rather than individual’s health. Within Lothian, Onomap has been used to monitor changes since the 2001 Census. The influx of 25,000 Poles since 2004 has been identified, making Poles the biggest ethnic minority within Lothian, and explaining some new demands placed on health and council services. Onomap has identified Poles registered with primary care, allowing mapping of the population to
Conclusion
There is a policy drive to improve ethnicity recording in routinely collected health datasets.3, 11, 12, 13 Nevertheless, improvements are unlikely to extend to all disease registers and administrative datasets in the near future, and the need for alternative means to identify population groups by ethnicity remain. Besides contemporary datasets, past health registers remain largely unanalysed by ethnicity because of their lack of classification and the impracticality to contact individuals to
Ethical approval
Use of the data was approved by NHS Lothian Caldicott Guardian.
Funding
None declared.
Competing interests
Pablo Mateos holds the copyright of the Onomap classification and may receive royalties for its licensing in the future.
Acknowledgements
The authors would like to thank Caroll Brown (NHS Lothian) for her assistance in executing Onomap, Sanjeev Paul (City of Edinburgh Council) for his assistance with the pupil census dataset, Mette Tranter (NHS Lothian) for her assistance with maps, Evropi Theodoratou (University of Edinburgh) for statistical advice, and the Health Promotion Foundation, Warsaw, Poland for provision of the Polish dataset.
References (36)
The mandatory collection of data on ethnic group of inpatients: experience of NHS trusts in England in the first reporting years
Public Health
(2000)- et al.
Determining aspects of ethnicity amongst persons of south Asian origin: the use of a surname-classification programme (Nam Pehchan)
Public Health
(2007) Independent inquiry into inequalities in health
(1998)- et al.
Musculoskeletal pain is more generalised among people from ethnic minorities than among white people in Greater Manchester
Ann Rheum Dis
(2002) - et al.
Hospital care among ethnic minorities in Britain
Health Trends
(1991) What is the risk of coronary heart disease in south Asians? a review of UK research
J Public Health Med
(2000)- et al.
Ethnic differences in consultation rates in urban general practice
BMJ
(1989) - et al.
Psychiatric illness among British Afro-Caribbeans
BMJ
(1988) Cervical cancer screening: meeting the needs of minority ethnic women
Br J Cancer Suppl
(1996)Collective terminology to describe the minority ethnic population: the persistence of confusion and ambiguity in usage
Sociology
(2002)
The future of ethnicity classifications
J Ethn Migr Stud
Ethnicity, race and health in multicultural societies
Ethnicity recording in general practice computer systems
J Public Health
Can we implement ethnic monitoring in primary health care and use the data?
J Public Health Med
Asian american ethnic identification by surname
Popul Res Policy Rev
An ontology of ethnicity based upon personal names. Implications for neighbourhood profiling
A review of name-based ethnicity classification methods and their potential in population studies
Popul Space Place
The cultural, ethnic and linguistic classification of populations and neighbourhoods using personal names. CASA working paper 116
Cited by (72)
Racialised people in clinical guideline panels
2022, The LancetNames-based ethnicity enhancement of hospital admissions in England, 1999–2013
2021, International Journal of Medical InformaticsCitation Excerpt :It should in this context be mentioned that classifications based on groups of closely associated forenames and surnames are also available, i.e. the methodology used for creating the related Onomap software [10,15]. Onomap was validated against the Scottish birth registration database for 2004–2008, with slightly higher sensitivity for White, South Asian, and Chinese names than found in this study [15]. The reported sensitivity for Black African names was however as low as 25 % (compared with nearly 50 % in this study).
Quality of ethnicity data within Scottish health records and implications of misclassification for ethnic inequalities in severe COVID-19: a national linked data study
2024, Journal of Public Health (United Kingdom)