Population genetics applications: carrier screening and population-screening ethics

Short version. Several Mendelian-disease alleles segregate at substantially elevated frequencies in specific populations — Tay-Sachs disease in Ashkenazi Jewish, Sephardi, French-Canadian, and Cajun populations; cystic fibrosis in northern European populations; sickle-cell anaemia in populations of African origin; the α- and β-thalassaemias in Mediterranean, South Asian, and South-East Asian populations. The published literature has historically organised carrier-testing panels around population-of-ancestry. Over the past decade the published consensus has moved towards pan-ethnic expanded carrier panels (Bell et al. 2011; ACOG Committee Opinion 691 2017). The Wilson and Jungner 1968 criteria remain the canonical starting point for evaluating any population-screening programme, and the ELSI literature on consent, equity, stigma, and discrimination has shaped programme design throughout.

Mendelian-disease epidemiology by population

Recessive Mendelian-disease alleles vary substantially in frequency between human populations. The patterns reflect demographic history (founder effects, drift in small populations) and, in a smaller number of well-documented cases, balancing selection. The published epidemiology informs both research and the design of any screening programme; it does not in itself prescribe whether a programme should exist.

Tay-Sachs disease

Tay-Sachs disease (HEXA, OMIM 606869) is a lysosomal storage disorder caused by deficiency of β-hexosaminidase A. Carrier frequency in Ashkenazi Jewish populations is approximately 1 in 27, compared with approximately 1 in 250 to 300 in non-founder populations. Elevated carrier frequencies are also documented in Sephardi Jewish, French-Canadian (notably in the Saguenay-Lac-Saint-Jean and Bas-Saint-Laurent regions of Québec), and Cajun (Louisiana, descended from Acadian French settlers) populations. The Tay-Sachs carrier-screening programmes initiated in the 1970s in the United States and Israel are among the earliest published examples of community-led carrier testing.

Cystic fibrosis

Cystic fibrosis (CFTR, OMIM 602421) carrier frequency in northern European populations is approximately 1 in 25; in Latino populations, approximately 1 in 60 to 70; in African-ancestry and East Asian populations, lower still. The most common pathogenic variant in northern European populations is F508del (c.1521_1523delCTT, p.Phe508del), which accounts for approximately 70 per cent of pathogenic CFTR alleles in that population; the variant spectrum differs across populations. Population-genetics models for the elevated cystic-fibrosis frequency in northern Europe have included historical balancing selection, but the published evidence is mixed.

Sickle-cell anaemia

Sickle-cell anaemia (HBB Glu6Val, OMIM 603903) carrier frequency in populations of African origin is approximately 1 in 12 in West Africa and 1 in 13 in the African-American population, with regional variation. The elevated frequency reflects balancing selection through heterozygote protection from severe Plasmodium falciparum malaria, documented in the classical work of Allison 1954 (Br Med J 1:290). Sickle-cell trait is also at elevated frequency in some Mediterranean, Middle Eastern, and South Asian populations whose history overlaps with malaria endemic regions.

Thalassaemias

The α-thalassaemias and β-thalassaemias are at elevated frequency in Mediterranean (Cyprus, Sardinia, Greece, southern Italy), Middle Eastern, South Asian, and South-East Asian populations — the geographical distribution again overlapping with historical malaria endemicity. Published carrier frequencies in Cyprus and Sardinia exceed 10 per cent in some regions; the Cyprus β-thalassaemia programme initiated in the 1970s is one of the most-cited examples of a sustained, community-supported carrier-testing programme in the literature. Carrier frequencies in South-East Asia for haemoglobin E (HbE) are also substantial.

Other conditions with documented population-specific elevation include Canavan disease, Niemann-Pick disease type A, Gaucher disease, familial dysautonomia, and several others in Ashkenazi Jewish populations; congenital adrenal hyperplasia in Yupik Eskimo populations; and a long list of conditions in the Finnish disease heritage covered on the demography page. The OMIM and GeneReviews catalogues are the canonical references for population-specific epidemiology.

Carrier-screening models

The published literature distinguishes several models of carrier-testing panel design.

Single-condition testing

Testing for a single condition for which there is a population-specific elevated risk and an established programme — for example sickle-cell disease, cystic fibrosis, or Tay-Sachs — is the historical model. Single-condition testing reflects the era when sequencing was expensive and panel design was constrained. The model is well-documented in the published literature on the early Tay-Sachs and Cyprus β-thalassaemia programmes.

Ancestry-based panels

Ancestry-based panels group conditions by population-of-ancestry and offer a different panel to individuals from different ancestral backgrounds. The classical example is the multi-condition Ashkenazi panel offered through community programmes in the 1990s and 2000s. The model is internally consistent — it offers each individual a panel calibrated to the highest-prevalence conditions in their population — but it depends on accurate self-reported ancestry, which is unreliable in admixed populations and increasingly inadequate as populations mix.

Expanded pan-ethnic carrier panels

Bell et al. 2011 (Sci Transl Med 3:65ra4) demonstrated the technical feasibility of testing simultaneously for hundreds of recessive conditions on a pan-ethnic panel using next-generation sequencing. The published consensus has subsequently moved towards pan-ethnic panels: ACOG Committee Opinion 691 (2017) in the United States explicitly endorses ethnic-and-pan-ethnic carrier-screening models, and the European Society of Human Genetics has issued comparable guidance. The arguments are that pan-ethnic panels avoid the practical and ethical problems of using self-reported ancestry to allocate testing, that they are equitable across admixed populations, and that the marginal cost of additional conditions on a sequencing-based panel is low. The arguments against are that pan-ethnic panels can produce more findings of uncertain significance, raise concerns about pre-test counselling burden, and may reduce attention to high-frequency conditions in specific populations. The published literature is active and evolving on these trade-offs.

Pre-conception, prenatal, and newborn timing

The published literature on screening-programme design distinguishes timing relative to reproduction:

• Pre-conception testing identifies carriers before pregnancy. The full range of reproductive options remains available to a couple identified as both carriers, including (per the published literature) natural conception with prenatal testing, donor conception, in vitro fertilisation with pre-implantation genetic testing, adoption, or no further intervention.
• Prenatal testing identifies affected pregnancies. The available options are narrower than at the pre-conception stage and the time pressure greater. Prenatal testing has its own established literature on consent, counselling, and choice.
• Newborn screening identifies affected newborns for conditions where early intervention can change outcomes (the published examples include phenylketonuria, congenital hypothyroidism, sickle-cell disease, and a range of others). Newborn-screening panels in the United Kingdom, the United States, and other jurisdictions are programmatic, jurisdiction-specific, and are not addressed further on this page.

This page does not advise the reader on whether or when to seek any of these. The literature on the timing trade-offs is a matter for textbook and guideline reading; the choice belongs to individuals and couples, in conversation with their healthcare provider, in their jurisdiction.

Wilson and Jungner: criteria for screening programmes

The canonical published framework for evaluating whether a screening programme should exist is Wilson and Jungner 1968 (WHO Public Health Papers 34, Principles and Practice of Screening for Disease), which sets out ten classical criteria. In summary, a population-screening programme is evaluated against:

1. The condition is an important health problem.
2. An accepted intervention exists for affected individuals.
3. Diagnostic and treatment facilities are available.
4. The condition has a recognisable latent or early symptomatic stage.
5. A suitable test or examination exists.
6. The test is acceptable to the population.
7. The natural history of the condition is adequately understood.
8. An agreed policy for whom to treat exists.
9. The cost of case-finding is balanced against the overall expenditure on healthcare.
10. Case-finding is a continuing process, not a one-off campaign.

The criteria have been updated and elaborated — notably in the 2008 WHO update by Andermann and colleagues to take account of genomic technology, asymptomatic carrier status, and the ELSI considerations that did not feature prominently in the 1968 framing — but the ten Wilson-Jungner criteria remain the starting point for any published evaluation of a screening programme.

Ethical, legal, and social issues

The ELSI literature on population genetic screening is substantial. The principal threads:

Consent

Consent for screening must be informed and voluntary. The published debate has covered the appropriate threshold for "informed" in the context of a screening test offered population-wide (where opt-out designs may dilute genuine informed consent), the boundaries of consent for incidental findings, and the special considerations for testing in children and in research. The Council of Europe Convention on Human Rights and Biomedicine and the UNESCO Universal Declaration on Bioethics and Human Rights are international reference frameworks; specific national legislation varies.

Equity

Equity considerations have been central since the early Tay-Sachs programmes and have intensified with expanded pan-ethnic panels. Equity issues include access to testing, access to follow-up testing and counselling, the diversity of reference cohorts and variant catalogues, the portability of polygenic risk scores across populations (substantially lower in non-European populations than in European-ancestry populations), and the historical under-representation of non-European populations in genetic research. The H3Africa initiative and All of Us cohort (covered on the demography page) are responses to the diversity gap.

Stigma

The published literature on population genetic screening includes documented cases where carrier-testing programmes produced stigma against carriers within communities. The Cyprus β-thalassaemia programme, the early Tay-Sachs programmes, and the United States sickle-cell trait screening of the 1970s are case studies. Subsequent programme design has emphasised pre-test education and the carrier-status framing (carrier status is not disease) in part as a response to the stigma literature.

Insurance and discrimination

Genetic information has at various times been used as a basis for insurance underwriting and employment decisions. The United States Genetic Information Nondiscrimination Act of 2008 (GINA) prohibits discrimination on the basis of genetic information in health insurance and employment, with exceptions; the Patient Protection and Affordable Care Act 2010 narrowed the gaps. The United Kingdom operates a moratorium on the use of predictive genetic test results by life and health insurers, agreed between the Association of British Insurers and the UK Government. Other jurisdictions (Canada, Australia, several European countries) have differing legal frameworks. The published literature notes that the protections vary by jurisdiction, by line of insurance, and over time.

National-scale programmes

Two large national programmes are well-documented in the published literature:

The 100,000 Genomes Project (United Kingdom, 2013 to 2018), described in Caulfield et al. 2017 (BMJ 361:k1687) and successor papers, sequenced approximately 100,000 whole genomes from National Health Service patients with rare disease and cancer. The project established the infrastructure for the subsequent NHS Genomic Medicine Service, which integrated whole-genome sequencing into NHS clinical care for specified indications from 2018 onwards. The project's published outputs include diagnostic yields by disease category, infrastructure papers, and a substantial ELSI literature on consent, recontact, and findings.

The NHS Genomic Medicine Service, the operational successor, delivers genomic testing across England with a published National Genomic Test Directory listing approved indications and panel content. Comparable national-scale programmes exist in France (Plan France Médecine Génomique 2025), Germany (genomDE), Australia (Australian Genomics), and other jurisdictions, with documented differences in scope, governance, and integration with healthcare systems.

The 1000 Genomes Project (Auton et al. 2015, Nature 526:68) sits alongside as a research-cohort reference rather than a screening programme. Its published variant catalogues underpin much of the technical infrastructure for both clinical variant interpretation and research-cohort analysis worldwide.

Pedigree documentation in the population-screening context

Family-history documentation pre-dates and exists alongside population genetic screening. A pedigree records biological relationships and observed disease occurrences in a family; it is the canonical artefact of family-history-based assessment regardless of whether population screening or targeted testing is also under consideration. Pedigree documentation is described on the pedigree chart, clinical pedigree drawing, and pedigree drawing tool pages.

Evagene is an academic, research, and educational pedigree-modelling platform. Its implementations of published risk-model algorithms (Claus 1994, Couch 1997, Frank 2002, Tyrer / Duffy / Cuzick 2004 in IBIS-style approximation, BayesMendel BRCAPRO / MMRpro / PancPRO, family-history scoring) consume the carrier-frequency, penetrance, and prior-probability inputs that the population-genetics literature has produced. Outputs are illustrative and for educational and research use; Evagene is not a screening tool, not a diagnostic tool, and not a medical device. Where a clinical computation exists off-platform — for example BOADICEA at canrisk.org, hosted by the University of Cambridge and not bundled in Evagene — Evagene exports a CanRisk 2.0 pedigree file that the user can choose to upload there, an architectural separation that this page documents rather than minimises.

Frequently asked questions

Related reading

• Population and evolutionary genetics (pillar)
• Allele frequency dynamics
• Demography and population structure
• Mendelian inheritance calculator
• Hereditary cancer risk assessment
• Autosomal recessive calculator