Skip to content

Type 1 diabetes — screening and early detection

TL;DR — Screening can identify islet autoimmunity years before symptomatic diabetes and sharply changes the diagnostic pathway, but it is not a stand-alone blood test: positives require confirmation, staging, longitudinal metabolic monitoring, education, and psychosocial support (Phillip 2024, PMID 38912694). In Bavaria, 90,632 children were screened and 280 (0.31%, 95% CI 0.27–0.35) had presymptomatic T1D (Ziegler 2020, PMID 31990315). Multiple autoantibodies predict far more progression than a single antibody, while population programs find many children without a family history (Ziegler 2013, PMID 23780460). The unresolved policy question is whether reduced DKA and access to teplizumab justify the assay, confirmation, surveillance, anxiety, and opportunity costs across whole populations.

What screening measures

Marker Antigen Typical role Important limitation
IAA Insulin Often an early antibody in young children Insulin treatment complicates interpretation after diagnosis
GADA GAD65 Common in adult-onset autoimmune diabetes Single positivity can persist without rapid progression
IA-2A IA-2 Often marks a higher-risk multiple-antibody pattern Assay thresholds and combinations matter
ZnT8A Zinc transporter 8 Adds sensitivity beyond older panels Absence does not exclude autoimmunity
ICA Islet-cell cytoplasmic staining Historical composite assay Labor-intensive and less antigen-specific

Autoantibodies are risk biomarkers, not direct measures of β-cell mass. A confirmed single antibody denotes lower and heterogeneous risk; two or more persistent antibodies with normoglycemia define stage 1, and dysglycemia defines stage 2 (Insel 2015, PMID 26404926; Phillip 2024, PMID 38912694).

Who can be screened

Strategy Population Advantage Structural weakness
Family-risk programs First- and second-degree relatives Higher yield; prevention-trial infrastructure Miss most eventual cases because most lack an affected relative
Genetic prescreen Newborn or child risk score followed by antibodies Concentrates repeat testing Calibration varies by ancestry (Sharp 2019, PMID 30655379)
General population Children at selected ages Finds cases regardless of pedigree Low prevalence means many tests and a large follow-up system
Clinical case-finding Symptoms, dysglycemia, diagnostic uncertainty Immediately actionable Too late to prevent many DKA presentations

Risk scores can improve newborn enrichment and classification, but European-derived scores transfer imperfectly to other ancestries; a Chinese validation illustrates the need for population-specific calibration (Sharp 2019, PMID 30655379; Hu 2025, PMID 40569436).

Evidence from Fr1da

Fr1da embedded capillary antibody testing in Bavarian well-child visits. Its design used a multiplex screen, reference-assay retesting, venous confirmation, metabolic staging, education, and assessment of family distress; 99.46% of the first 26,760 capillary samples were adequate (Raab 2016, PMID 27194320).

Fr1da measure Result
Children in the 2015–2019 report 90,632
Presymptomatic T1D 280; 0.31% (95% CI 0.27–0.35)
Stage 1 at detection 196; 0.22%
Stage 2 at detection 17; 0.02%
Stage 3 at detection 26; 0.03%
Median follow-up 2.4 years

The program establishes feasibility and yield, not universal cost-effectiveness. Participation by pediatricians and families, age windows, background incidence, assay logistics, and available follow-up all affect transportability (Raab 2016, PMID 27194320; Ziegler 2020, PMID 31990315).

Confirmation and staging pathway

  1. An initial positive result should be confirmed on a second sample, preferably in a reference laboratory (Phillip 2024, PMID 38912694).
  2. Confirmed antibody identity and multiplicity separate lower-risk single-antibody status from stage 1/2 disease.
  3. Glucose, HbA1c, and when appropriate oral-glucose-tolerance testing establish metabolic stage.
  4. Monitoring intensity is individualized by age, antibody pattern, glucose trajectory, and symptoms.
  5. Every pathway requires DKA-symptom education, a rapid route to clinical review, and psychosocial support.
  6. People with stage 2 should be offered information about trials and approved delay therapy where eligible (Phillip 2024, PMID 38912694).

The progression-likelihood score combines longitudinal glucose and C-peptide data and can stratify presymptomatic children more finely than stage alone, but implementation thresholds remain program-dependent (Weiss 2022, PMID 36028774).

Benefits that should be measured

Outcome Why it matters Design problem
DKA at stage-3 diagnosis Immediate preventable morbidity Awareness and access programs also change DKA
HbA1c and clinical condition at diagnosis Measures metabolic decompensation Follow-up intensity can confound screening effect
Time from symptoms to insulin Tests pathway responsiveness Requires linked clinical data
Family anxiety and distress Captures harm as well as reassurance Distress changes over time and by counseling quality
Trial or teplizumab uptake Converts detection into disease modification Eligibility and affordability vary
Cost per DKA avoided or QALY gained Enables policy comparison Long horizons and uncertain therapy durability
Equity of uptake and retention Detects selective benefit Participation can reproduce access disparities

DKA remains common at diagnosis worldwide: a 233-study pediatric meta-analysis estimated 41.9% (95% CI 39.7–44.0), with wide between-country variation (Zhang 2026, PMID 42303108). Screening should therefore be compared with symptom-awareness campaigns, primary-care education, and combinations rather than with no intervention alone.

Ethics and communication

Presymptomatic diagnosis converts a healthy-seeming person into someone living with an uncertain disease clock. Communication must distinguish single from multiple antibodies, relative from absolute risk, and delay from prevention. The monitoring consensus explicitly includes psychosocial support because repeated testing, future-treatment decisions, and fear of acute onset can create burden (Phillip 2024, PMID 38912694).

Consent is especially complex in children: parents authorize testing, but the result creates years of obligations for the child. Programs should predefine data governance, recontact, confirmatory testing, withdrawal, and what happens when families move or lose insurance.

Screening after teplizumab

Teplizumab changed the value proposition because stage 2 became actionable. In 76 relatives, one 14-day course shifted median diagnosis from 24.4 to 48.4 months and reduced the hazard of stage 3 (HR 0.41, 95% CI 0.22–0.78), with rash and transient lymphopenia among expected adverse events (Herold 2019, PMID 31180194).

The trial was small, enriched for relatives, and does not show that every screen-detected child benefits equally. Screening policy must incorporate age eligibility, confirmatory staging, infusion capacity, safety monitoring, treatment preferences, and the fact that delay is not established prevention.

Screening economics: assumptions drive the answer

The ASK model estimated $4,700 per case detected under research-program delivery and $14,000 under a routine-care scenario. Avoiding DKA alone did not reach $50,000–$150,000 per QALY thresholds; cost-effectiveness required a 20% DKA reduction plus a sustained 0.1 percentage-point HbA1c improvement (McQueen 2020, PMID 32327420).

Model/input Estimate Sensitivity
ASK research cost per detected case $4,700 Existing study infrastructure
ASK routine-care scenario $14,000 Staff, testing, confirmation, follow-up
Australian newborn bloodspot genetic-first ICER $50,682/QALY Time horizon, discounting, assay cost
Australian population antibody screening ICER $133,285/QALY Lower prevalence at each test and repeat costs
German modeled DKA reduction with screening at ages 2 and 6 61% Assumes monitoring adherence and health-system response

In an Australian 100,000-person microsimulation, newborn bloodspot genetic risk stratification followed by antibody testing was the most cost-effective of three modeled strategies, but cost $480,798 per screen-detected T1D and $12,183 per DKA episode avoided (Chen 2025, PMID 41362419). Model ranking is not a universal policy result: genetics, incidence, DKA prevalence, test price, and treatment availability differ.

Psychological consequences are outcomes

Among 280 confirmed antibody-positive children enrolled in ASK follow-up, mean parental State Anxiety Inventory score at the first visit was 46.1±11.2, above the clinical cutoff of 40, and remained elevated at the next visit despite beginning to decline. Only 48.9% initially understood their child as having increased risk, demonstrating that anxiety and accurate risk perception can diverge (O'Donnell 2023, PMID 37673098).

Screening programs should therefore measure:

  • immediate and longitudinal anxiety in the screened person and caregivers;
  • risk comprehension, including uncertainty and time horizon;
  • school, insurance, and treatment-access consequences;
  • monitoring attendance and loss to follow-up;
  • stage-3 condition, DKA, hospitalization, and HbA1c;
  • treatment uptake and regret where disease modification is offered.

Monitoring intensity modifies apparent benefit

TEDDY diagnosed 379 children during prospective follow-up; DKA occurred in 23 (6.1%), including 1/103 (0.97%) with an affected first-degree relative and 22/276 (8.0%) without. Median time since the last study visit was 10.2 months in those with DKA versus 2.0 months without DKA (Jacobsen 2022, PMID 35043162). Screening itself is therefore not the active ingredient; retained monitoring, education, and rapid diagnostic access are.

A single-center TrialNet pathway screened 4,046 relatives, followed participants for a median 9.9 years, and documented no DKA among 51 progressors. Fifteen-year stage-3-free survival was 99.5% in antibody-negative, 87.3% in single-positive, and 45.9% in multiple-positive participants (Martinenghi 2025, PMID 40439773). This intensive specialist result may not transport to population programs with attrition.

Assay and threshold problems

Problem Quantified evidence Operational response
Initial low-risk positives In DAISY, 31% false on blinded duplicate and 31% transient Confirm on a separate sample
Cross-laboratory units GADA/IA-2A absolute values differed despite WHO calibration Use harmonized protocols and proficiency testing
Titer ignored Five-year risk ranged 6–75% after type-specific titer thresholds Preserve quantitative values, not positive/negative alone
Adult HbA1c aging Standard ≥5.7% threshold yielded lower one-year risk in adults than children Validate age-adjustment or ≥6.0% in adults ≥30

Harmonized IA-2A assays achieved >99% specificity and 64% sensitivity; GADA discordance in retested TEDDY samples fell from 15.4% to 2.7% (Bonifacio 2010, PMID 20444913). Across three laboratories, positive/negative agreement was high while absolute WHO-unit values still differed markedly (Bingley 2010, PMID 20693189).

In 24,662 at-risk children, antibody-type-specific titer thresholds stratified five-year risk from 6% to 75%; single screening at age 10 in an enriched adolescent cohort had 90% sensitivity (95% CI 86–95) and 66% positive predictive value for diabetes by 18, while adding age 14 raised sensitivity to 93% but lowered PPV to 55% (Ng 2022, PMID 34758977; Ghalwash 2023, PMID 36681087).

Adult thresholds require separate calibration. In 5,024 antibody-positive TrialNet relatives, standard HbA1c ≥5.7% produced one-year risks of 38% in children versus 13% in adults with one antibody; age adjustment reduced the difference, and ≥6.0% produced more comparable risk (Templeman 2026, PMID 42090204).

Screening interpretation also depends on when autoimmunity begins and which antibody appears. DAISY documented late-onset childhood autoimmunity that would be missed by a single early screen (Frohnert 2017, PMID 28314946). TEDDY found distinct predictors for initiation, antibody spreading, and progression (Krischer 2022, PMID 36150053), while TrialNet showed that IA-2A positivity increases progression risk within and across established stages (Sims 2025, PMID 40016443). A binary “any antibody” endpoint therefore loses information relevant to rescreening and monitoring intensity.

Open questions

  • Which ages and repeat-testing intervals capture the most seroconversions per test without missing rapid progressors? (Ziegler 2013, PMID 23780460)
  • Does population screening reduce DKA beyond high-quality awareness and rapid-access programs, and at what incremental cost? (Zhang 2026, PMID 42303108)
  • How should ancestry-calibration errors in genetic prescreening be measured and corrected? (Sharp 2019, PMID 30655379; Hu 2025, PMID 40569436)
  • What is the long-term psychological trajectory after a positive screen outside specialist research programs? (Phillip 2024, PMID 38912694)
  • Will repeat or combination disease-modifying therapy make screening more cost-effective, or increase treatment burden? (Herold 2019, PMID 31180194)

References

  1. Insel RA, et al. Staging presymptomatic type 1 diabetes. Diabetes Care. 2015;38:1964-1974. PMID 26404926
  2. Ziegler AG, et al. Seroconversion to multiple islet autoantibodies and risk of progression to diabetes in children. JAMA. 2013;309:2473-2479. PMID 23780460
  3. Raab J, et al. Capillary blood islet autoantibody screening for identifying pre-type 1 diabetes. BMJ Open. 2016;6:e011144. PMID 27194320
  4. Ziegler AG, et al. Yield of a Public Health Screening of Children for Islet Autoantibodies in Bavaria, Germany. JAMA. 2020;323:339-351. PMID 31990315
  5. Weiss A, et al. Progression likelihood score identifies substages of presymptomatic type 1 diabetes. Diabetologia. 2022;65:2121-2131. PMID 36028774
  6. Phillip M, et al. Consensus Guidance for Monitoring Individuals With Islet Autoantibody-Positive Pre-Stage 3 Type 1 Diabetes. Diabetes Care. 2024;47:1276-1298. PMID 38912694
  7. Sharp SA, et al. Development and Standardization of an Improved Type 1 Diabetes Genetic Risk Score. Diabetes Care. 2019;42:200-207. PMID 30655379
  8. Hu J, et al. A type 1 diabetes genetic risk score discriminates between type 1 and type 2 diabetes in a Chinese population. Diabetologia. 2025. PMID 40569436
  9. Herold KC, et al. An Anti-CD3 Antibody, Teplizumab, in Relatives at Risk for Type 1 Diabetes. N Engl J Med. 2019;381:603-613. PMID 31180194
  10. Zhang T, et al. Worldwide prevalence of diabetic ketoacidosis at diagnosis of type 1 diabetes. Prev Med. 2026;210:108625. PMID 42303108
  11. McQueen RB, et al. Cost and Cost-effectiveness of Large-scale Screening for Type 1 Diabetes in Colorado. Diabetes Care. 2020;43:1496-1503. PMID 32327420
  12. Chen W, et al. Economic evaluation of potential national childhood screening strategies for type 1 diabetes in Australia. Lancet Reg Health West Pac. 2025;65:101755. PMID 41362419
  13. Rewers M. Health economic considerations of screening for early type 1 diabetes. Diabetes Obes Metab. 2025;27 Suppl 6:69-77. PMID 40555704
  14. O'Donnell HK, et al. Anxiety and Risk Perception in Parents of Children Identified by Population Screening. Diabetes Care. 2023;46:2155-2161. PMID 37673098
  15. Jacobsen LM, et al. Heterogeneity of DKA Incidence in Children Diagnosed in TEDDY. Diabetes Care. 2022;45:624-633. PMID 35043162
  16. Martinenghi S, et al. Prevention of diabetic ketoacidosis in relatives screened and followed in TrialNet. Diabetologia. 2025;68:1889-1898. PMID 40439773
  17. Barker JM, et al. Prediction of autoantibody positivity and progression to type 1 diabetes: DAISY. J Clin Endocrinol Metab. 2004;89:3896-3902. PMID 15292324
  18. Bonifacio E, et al. Harmonization of glutamic acid decarboxylase and islet antigen-2 autoantibody assays. J Clin Endocrinol Metab. 2010;95:3360-3367. PMID 20444913
  19. Bingley PJ, et al. Measurement of islet cell antibodies in the Type 1 Diabetes Genetics Consortium. Clin Trials. 2010;7:S56-S64. PMID 20693189
  20. Ng K, et al. Islet Autoantibody Type-Specific Titer Thresholds Improve Risk Stratification. Diabetes Care. 2022;45:160-168. PMID 34758977
  21. Ghalwash M, et al. Islet autoantibody screening in at-risk adolescents to predict type 1 diabetes. Lancet Child Adolesc Health. 2023;7:261-268. PMID 36681087
  22. Templeman EL, et al. Accounting for Age-Related Increases in HbA1c Quantifies Progression Risk in Adults. Diabetes Care. 2026;49:1262-1269. PMID 42090204
  23. Frohnert BI, et al. Late-onset islet autoimmunity in childhood: DAISY. Diabetologia. 2017;60:998-1006. PMID 28314946
  24. Krischer JP, et al. Predictors of initiation of islet autoimmunity and progression. Diabetes Care. 2022;45:2271-2281. PMID 36150053
  25. Sims EK, et al. IA-2A positivity and progression within and across T1D stages. Diabetologia. 2025;68:993-1004. PMID 40016443