The DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index; 2 of the 8 sources its claims cite were checked for support.
The epidemiological impact of the NHS COVID-19 app
Open accessAbstract
The COVID-19 pandemic has seen the emergence of digital contact tracing to help to prevent the spread of the disease. A mobile phone app records proximity events between app users, and when a user tests positive for COVID-19, their recent contacts can be notified instantly. Theoretical evidence has supported this new public health intervention 1-6, but its epidemiological impact has remained uncertain 7. Here we investigate the impact of the National Health Service (NHS) COVID-19 app for England and Wales, from its launch on 24 September 2020 to the end of December 2020. It was used regularly by approximately 16.5 million users (28% of the total population), and sent approximately 1.7 million exposure notifications: 4.2 per index case consenting to contact tracing. We estimated that the fraction of individuals notified by the app who subsequently showed symptoms and tested positive (the secondary attack rate (SAR)) was 6%, similar to the SAR for manually traced close contacts. We estimated the number of cases averted by the app using two complementary approaches: modelling based on the notifications and SAR gave an estimate of 284,000 (central 95% range of sensitivity analyses 108,000-450,000), and statistical comparison of matched neighbouring local authorities gave an estimate of 594,000 (95% confidence interval 317,000-914,000). Approximately one case was averted for each case consenting to notification of their contacts. We estimated that for every percentage point increase in app uptake, the number of cases could be reduced by 0.8% (using modelling) or 2.3% (using statistical analysis). These findings support the continued development and deployment of such apps in populations that are awaiting full protection from vaccines.
Overview
CoChat Summary
TL;DR
The NHS COVID-19 app for England and Wales was used regularly by approximately 16.5 million people and sent about 1.7 million exposure notifications from launch to end-December 2020, with an estimated secondary attack rate of 6% among notified users. Two complementary analyses estimated 284,000 cases averted by modelling and 594,000 by matched-neighbours regression, supporting digital contact tracing as an additional public-health intervention.
Key findings
- The app was regularly used by approximately 16.5 million users (28% of the total population) and sent approximately 1.7 million exposure notifications: 4.2 per index case consenting to contact tracing. The estimated secondary attack rate among notified individuals was 6.02% (confidence interval 5.96–6.09%; sensitivity analyses suggested roughly 5–7%).
- Cases averted during phases 1 and 2 combined were estimated at 284,000 (central 95% sensitivity range 108,000–450,000) using modelling and 594,000 (95% CI 317,000–914,000) using matched-neighbours regression. Corresponding deaths averted were 4,200 (1,600–6,600) and 8,700 (4,700–13,500), against 1,892,000 actual cases and 32,500 actual deaths in the period.
- For every percentage point increase in app uptake, cases were estimated to fall by 0.8% using modelling or 2.3% using statistical analysis. The effect was larger in phase 2, after an upgrade to the app’s risk-scoring function.
- Scenario projections for phase 2 included: increasing uptake to 35.9% for all lower tier local authorities (11% modelling, 21.0% statistical reduction), increasing uptake by 20 percentage points (24% modelling, 41.5% statistical), switching to opt-out notification (6.6% modelling), improving quarantine adherence by 20 percentage points (6.8% modelling), and reducing time to test result by one day (3.6% modelling).
Methods
- Observational analysis of NHS COVID-19 app data for England and Wales from launch on 24 September 2020 to 31 December 2020, aggregated at lower tier local authority level, with case data from coronavirus.data.gov.uk and analysis phases defined around app changes.
- Two complementary approaches: a mechanistic model linking notifications, secondary attack rate, timing of notification relative to transmission, and quarantine adherence (central assumption of 61% overall quarantine effectiveness) to estimate cases averted; and a matched-neighbours regression comparing adjacent LTLAs within the same phase-0 case quintile, adjusting for rural/urban score, local GDP, and poverty, with a placebo regression to assess confounding.
- The secondary attack rate was estimated with a probabilistic model using iPhone data from December 2020–January 2021 and a Poisson likelihood; deaths averted were derived using a case fatality rate of 1.47% (27,922 deaths / 1,891,777 cases).
Limitations
- The paper states the main limitation is that this is an observational study with no randomized or systematic experiment producing different app uptake rates; residual confounding could mean the statistical estimate is too high, while a genuine indirect effect of app use could mean the modelling estimate is too low, so an effect size between the two estimates seems most likely.
- The paper states that epidemic dynamics for individual LTLAs were difficult to interpret because the period included staggered local restrictions, a national lockdown, the Christmas holiday season, and emergence of the more infectious B.1.1.7 variant; it did not model time-varying reproduction numbers and suggests future hierarchical modelling.
- The text makes evident additional uncertainties: quarantine adherence is critical but difficult to assess reliably, with cited surveys ranging from 11% proper adherence to greater than 80% recent adherence; Android user-number fluctuations are described as unexplained; and sensitivity analyses suggested greater uncertainty in the SAR than the likelihood-profile confidence interval alone.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
2 Valid · 0 Not valid · 6 Not checked · 46 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“4.2 per index case consenting to contact tracing.”
- Also stated in: § The epidemiological impact of the NHS COVID-19 app · p. 1 · plain
“resulting in a mean number of notifications per tracing event of 4.2.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § The epidemiological impact of the NHS COVID-19 app · p. 1 · plain
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“or 2.3% (using statistical analysis)”
- Also stated in: § The epidemiological impact of the NHS COVID-19 app · p. 3 · plain
“2.26 (1.50-3.00)”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § The epidemiological impact of the NHS COVID-19 app · p. 3 · plain
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“To test for heterogeneity, it was also estimated as the regression of local deaths to cases, but no substantial heterogeneity was observed (not shown).”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It was used regularly by approximately 16.5 million users (28% of the total population)”
- Also stated in: § The epidemiological impact of the NHS COVID-19 app · p. 1 · plain
“is regularly used by at least 16.5 million people.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § The epidemiological impact of the NHS COVID-19 app · p. 1 · plain
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Seventy-two per cent of app-using index cases consented to digital tracing upon testing positive”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
26 more claims
Create a free account to see all 54 claims and to verify the 6 references still open.
- It may state two separate claims as one. 11
- Probably not a substantial claim. 16
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source says what the claim attributes to it.
In the citing paper
“One example of this is the United Kingdom’s National Health Service Covid-19 contact-tracing app, which helped prevent between 100,000 and 900,000 Covid-19 infections from October to December 2020.73”
From Addressing bias in big data and AI for health care: A call for open science (2021) · cites this paper as [73]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ The epidemiological impact of the NHS COVID-19 app · p. 3
“For comparison, the total number of cases and deaths that actually occurred in this period were 1,892,000 and 32,500, respectively.”
§ Case fatality rate · p. 7
“The case fatality rate was estimated as the ratio of total deaths (27,922) to cases (1,891,777) for phases 1 and 2 combined.”
Citations
Cites
Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe
Flaxman, S., Mishra, S., Gandy, A., Unwin, H. J. T., Mellan, T. A., Coupland, H., Whittaker, C., Zhu, H., Berah, T., Eaton, J. W., et al. (2020) · Nature · doi:10.1038/s41586-020-2405-7 · Cited by 3,350
Using Big Data to Emulate a Target Trial When a Randomized Trial Is Not Available: Table 1.
Hernán, M. A., & Robins, J. M. (2016) · American Journal of Epidemiology · doi:10.1093/aje/kwv254 · Cited by 3,115
Estimated transmissibility and impact of SARS-CoV-2 lineage B.1.1.7 in England
Davies, N. G., Abbott, S., Barnard, R. C., Jarvis, C. I., Kucharski, A. J., Munday, J. D., Pearson, C. A. B., Russell, T. W., Tully, D. C., Washburne, A. D., et al. (2021) · Science · doi:10.1126/science.abg3055 · Cited by 2,581
Estimated transmissibility and impact of SARS-CoV-2 lineage B.1.1.7 in England
Davies, N. G., Abbott, S., Barnard, R. C., Jarvis, C. I., Kucharski, A. J., Munday, J. D., Pearson, C. A. B., Russell, T. W., Tully, D. C., Washburne, A. D., et al. (2020) · doi:10.1101/2020.12.24.20248822 · Cited by 2,557
Quantifying SARS-CoV-2 transmission suggests epidemic control with digital contact tracing
Ferretti, L., Wymant, C., Kendall, M., Zhao, L., Nurtay, A., Abeler-Dörner, L., Parker, M., Bonsall, D., & Fraser, C. (2020) · Science · doi:10.1126/science.abb6936 · Cited by 2,512
Self-citationAssessing transmissibility of SARS-CoV-2 lineage B.1.1.7 in England
The COVID-19 Genomics UK (COG-UK) consortium, Volz, E., Mishra, S., Chand, M., Barrett, J. C., Johnson, R., Geidelberg, L., Hinsley, W. R., Laydon, D. J., Dabrera, G., et al. (2021) · Nature · doi:10.1038/s41586-021-03470-x · Cited by 1,611
Volz, E., Mishra, S., Chand, M., Barrett, J. C., Johnson, R., Geidelberg, L., Hinsley, W. R., Laydon, D. J., Dabrera, G., O’Toole, Á., et al. (2021) · doi:10.1101/2020.12.30.20249034 · Cited by 1,611
Impact of delays on effectiveness of contact tracing strategies for COVID-19: a modelling study
Kretzschmar, M. E., Rozhnova, G., Bootsma, M. C. J., van Boven, M., van de Wijgert, J. H. H. M., & Bonten, M. J. M. (2020) · The Lancet Public Health · doi:10.1016/s2468-2667(20)30157-2 · Cited by 697
Quantifying SARS-CoV-2 transmission suggests epidemic control with digital contact tracingPreprint
Ferretti, L., Wymant, C., Kendall, M., et al. (2020) · medRxiv · doi:10.1101/2020.03.08.20032946 · Cited by 528
Self-citationSmith, L. E., Potts, H. W. W., Amlôt, R., Fear, N. T., Michie, S., & Rubin, G. J. (2021) · BMJ · doi:10.1136/bmj.n608 · Cited by 297
Cited by
Addressing bias in big data and AI for health care: A call for open science
Norori, N., Hu, Q., Aellen, F. M., Faraci, F. D., & Tzovara, A. (2021) · Patterns · doi:10.1016/j.patter.2021.100347 · Cited by 903
“example of this is the United Kingdom’s National Health Service Covid-19 contact-tracing app, which helped prevent between 100,000 and 900,000 Covid-19 infections from October to December 2020.(73) Organizations like Omdena and the Alan Turing Institute are pioneers in developing ethical AI solutions in a humanitarian…”
BackgroundSonabend, R., Whittles, L. K., Imai, N., Perez-Guzman, P. N., Knock, E. S., Rawson, T., Gaythorpe, K. A. M., Djaafara, B. A., Hinsley, W., FitzJohn, R. G., et al. (2021) · The Lancet · doi:10.1016/s0140-6736(21)02276-5 · Cited by 161
“…in the forward projections we systematically varied four factors as follows: vaccine effectiveness against delta; cross-protection against delta from previous infection with non-delta variants; the duration of natural immunity; and the level of transmissibility after step four (appendix pp 47–48).”
Artificial intelligence for modelling infectious disease epidemics
Kraemer, M. U. G., Tsui, J. L.-H., Chang, S. Y., Lytras, S., Khurana, M. P., Vanderslott, S., Bajaj, S., Scheidwasser, N., Curran-Sebastian, J. L., Semenova, E., et al. (2025) · Nature · doi:10.1038/s41586-024-08564-w · Cited by 169
“During the COVID-19 pandemic in the UK, the National Health Service COVID-19 app was downloaded to >21 million mobile phone devices, providing users with timely notifications of infectious disease exposures and instructions in case their contact was diagnosed positive 90 .”
Self-citationEffectiveness evaluation of digital contact tracing for COVID-19 in New South Wales, Australia
VOGT, F. M., Haire, B. G., Selvey, L. A., et al. (2022) · The Lancet Public Health · doi:10.1016/s2468-2667(22)00010-x · Cited by 111
Effectiveness of contact tracing in the control of infectious diseases: a systematic review
Hossain, A. D., Jarolimova, J., Elnaiem, A. D., Huang, C. X., Richterman, A. G., & Ivers, L. C. (2022) · The Lancet Public Health · doi:10.1016/s2468-2667(22)00001-9 · Cited by 133
Metadata
- Authors
- Wymant, C., Ferretti, L., Tsallis, D., Charalambides, M., Abeler-Dörner, L., Bonsall, D., Hinch, R., Kendall, M., Milsom, L., Ayres, M., Holmes, C., Briers, M., & Fraser, C.
- Affiliations
- University of Oxford (GB)
- The Alan Turing Institute (GB)
- University of Warwick (GB)
- Corresponding author
- Fraser, Christophe
- ORCID
- Venue
- Nature
- Published
- 2021-05-12
- Volume
- 594
- Issue
- 7863
- Pages
- 408-412
- ISSN
- 0028-0836 (ISSN-L)
- 1476-4687
- OpenAlex
- W3163076983
- Semantic Scholar
- e1709067cc25c48e198745fa60f721379f0145d1
- Cited by
- 271
- References
- 30
- Citations by year
- 2026: 13 · 2025: 18 · 2024: 50 · 2023: 63 · 2022: 73 · 2021: 52
- FWCI
- 53.74
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- www.springer.com/tdm (tdm)
- www.springer.com/tdm (vor)
- Locations
- Nature (publishedVersion)
- PubMed (publishedVersion)
- Warwick Research Archive Portal (University of Warwick) (acceptedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- Lirias (KU Leuven) (submittedVersion)
- Funders
- Li Ka Shing Foundation
- National Institute for Health and Care Research (NIHR-INF-1687)
- Department of Health and Social Care
- Nature
- Medical Research Council (MR/R015600/1, MC_PC_19012)
- Engineering and Physical Sciences Research Council (EP/R018561/1)
- Topics
- Physical Sciences › Computer Science › Information Systems › COVID-19 Digital Contact Tracing
- Physical Sciences › Mathematics › Modeling and Simulation › COVID-19 epidemiological studies
- Health Sciences › Medicine › Epidemiology › Data-Driven Disease Surveillance
- Keywords
- digital contact tracing, COVID-19 pandemic, secondary attack rate, case averted, exposure notification, mobile application, contact tracing, SARS-CoV-2 transmission, app uptake, public health interventions
- MeSH terms
- COVID-19, England, Humans, Mortality, National Health Programs, Quarantine, Wales, Contact Tracing, Basic Reproduction Number, Mobile Applications
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-04, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-04 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: The epidemiological impact of the NHS COVID-19 app. https://app.cochat.ai/papers/10.1038/s41586-021-03606-z
- The paper
- Wymant, C., Ferretti, L., Tsallis, D., Charalambides, M., Abeler-Dörner, L., Bonsall, D., Hinch, R., Kendall, M., Milsom, L., Ayres, M., Holmes, C., Briers, M., & Fraser, C. (2021). The epidemiological impact of the NHS COVID-19 app. Nature. https://doi.org/10.1038/s41586-021-03606-z
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.