Skip to main content
JJOS Demo Journals
HomeJournalsCascadia Journal of Biomedical ResearchVolume 17, Issue 110.58302/35pmjq57
Cascadia Journal of Biomedical Research
Article

The Eat-Out-to-Help-Out incentive: A trigger for gastrointestinal infections in England, 2020?

Reece Jarratt1*, Helen Clough2, Ewan Wilkinson3, Roberto Vivancos4, Valérie Decraene5

1https://ror.org/018h10037United Kingdom Health Security Agency, London, UK

2Institute of Infection, Veterinary and Ecological Sciences, https://ror.org/04xs57h96University of Liverpool, Liverpool, UK; National Institute for Health and Care Research Health Protection Research Unit in Gatrointestinal Infections, Norwich, UK

3Institute of Medicine, https://ror.org/01drpwb22University of Chester, Chester, UK

4https://ror.org/018h10037United Kingdom Health Security Agency, London, UK; National Institute for Health and Care Research Health Protection Research Unit in Gatrointestinal Infections, Norwich, UK; https://ror.org/05vvt7a66National Institute for Health and Care Research Health Protection Research Unit in Emerging and Zoonotic Infections, Liverpool, UK; Warwick Medical School, https://ror.org/01a77tt86University of Warwick, Coventry, UK

5https://ror.org/018h10037United Kingdom Health Security Agency, London, UK; National Institute for Health and Care Research Health Protection Research Unit in Gatrointestinal Infections, Norwich, UK

* Correspondence: reece.jarratt@ukhsa.gov.uk

Cascadia J. Biomed. Res. 2025, 17(1), e3; https://doi.org/10.58302/35pmjq57

Abstract

Our study assessed the link between gastrointestinal (GI) infections in England and the Eat Out to Help Out scheme (EOHO), a government subsidy created to encourage people to eat out during COVID-19 pandemic (03–30 August 2020). We studied national laboratory data between January 2015 and December 2020. We used time series change point analysis to see if there were shifts in reported cases of specific GI infections (Campylobacter spp., Escherichia coli O157, and non-typhoidal Salmonella spp.) associated with the timing of the scheme. Our analysis uniquely applied the Pruned Exact Linear Time method, with generalized linear models to a national dataset of GI infections. This revealed increases in cases closely aligned to the timing of the easing of COVID-19 restrictions, prior to the introduction of the EOHO scheme. Our study showed the scheme had no measurable impact, as there was no significant change on reported cases. Substantial reductions in cases after the first lockdown, followed by an increase as restrictions were phased out, show the wider impact of COVID-19 control measures, for example, public information campaigns aimed at improving hand-hygiene. These findings highlight the complicated interactions between COVID-19 control measures, the public’s behaviour, and the spread of GI infections.

Keywords:Food-borne infectionsCampylobacterEscherichia coli (E. coli)SalmonellaChange point

Introduction

Gastrointestinal (GI) infections result in health and public health challenges globally, causing considerable morbidity and placing a burden on healthcare systems [1, 2].

In England, an estimated 17 million GI cases occur annually [3], resulting in a societal cost of more than £9 billion, primarily due to loss of earnings [4]. These infections follow seasonal patterns, with bacterial pathogens peaking in summer months, linked to increasing temperatures and overseas travel [5].

Foodborne infections account for an estimated 2.4 million GI cases annually in England [4]. Up to 80% of reported Campylobacter spp., Salmonella spp., and Shiga-toxin producing Escherichia coli (including O157) infections can be attributed to contaminated food products [6, 7].

Deprivation influences exposure to risks of foodborne infections, with lower rates of Campylobacter and non-typhoidal Salmonella reported in more deprived areas [8]. This trend may partly reflect reduced opportunities for international travel and dining out, the former being a key risk factor for the development of GI infections and more common in affluent populations [9]. However, underreporting due to disparities in healthcare-seeking behaviour within deprived populations may distort the true incidence [10].

Geographic location influences infection patterns. Urban areas present higher exposure risks due to the dense populations and non-compliant food establishments [11]. Rural areas report more Campylobacter cases, likely from increased environmental exposure to livestock excretions [12]. However, this may reflect deprivation classification rather than true geographic transmission differences [13].

Due to the emergence of COVID-19, governments worldwide introduced non-pharmaceutical interventions (NPIs) to reduce the transmission and burden of the virus. These interventions ranged from social distancing, closure of businesses, international-travel restrictions, and an enforceable nationwide ‘lockdowns’ [14, 15].

Such interventions, coupled with improved hand hygiene through public health information campaigns, significantly reduced rates of endemic communicable diseases, as transmission routes were removed or interrupted [9]. For GI infections this was clearly demonstrated, most notably in infections driven by person-to-person transmission. The reductions in social interactions halted the spread of infections [16].

A national lockdown left the hospitality sector financially vulnerable. Over 80% of businesses were temporarily closed [17], resulting in consumer spending in the hospitality sector falling by 56% in spring 2020 compared to pre-pandemic levels [18]. In response, the UK government introduced a fiscal policy which aimed to stimulate demand for the sector through the introduction of the eat out to help out scheme (EOHO). The scheme directly subsidized 50% of cost of meals and non-alcoholic drinks in participating venues on Mondays to Wednesdays between 3 August 2020 and 31 August 2020 [19]. The EOHO scheme was the first initiative globally to encourage individuals to dine out following the gradual easing of national lockdown regulations (10 May 2020), which allowed restaurants to reopen on 04 July 2020 [15].

The pandemic brought significant workforce challenges for the hospitality sector, with 18% of furloughed jobs originating from the industry, resulting in difficulties in retaining and recruiting trained staff [20, 21]. Together with understaffing, high customer volumes have been cited by food workers as a factor to maintaining food safety practices [22]. During the EOHO scheme, there was marked increase in restaurant visits, exceeding volumes during the same period in the previous year [23]. Meanwhile, routine hygiene inspections in England decreased by approximately a third during the pandemic compared to pre-pandemic levels [21]. Together, these factors provided opportunities for lapses in compliance to safe food handling and preparation practices, as well as increased exposure to foodborne pathogens.

Surveillance of GI infections during events that may have increased public dining behaviours (e.g., 2012 London Olympics and 2022 Birmingham Commonwealth Games) found no outbreaks or public health incidents explicitly linked to dining activities [24, 25]. Unlike sporting or cultural events, EOHO was unique in its scale, intensity, and duration. By its design, it aimed to create a significant shift in dining behaviours, differing from other temporary influences.

To date, there has been no analysis of the effects of the EOHO scheme other than assessing its impact on the incidence of COVID-19 infections [23]. Our study aimed to assess if the EOHO scheme was associated with changes in the incidence of selected GI infections during the COVID-19 pandemic in 2020 among residents of England.

Methods

We conducted a retrospective ecological study by performing secondary analyses on routinely collected national data.

Foodborne pathogens, including Campylobacter spp., non-typhoidal Salmonella spp., and E. coli O157, are lawfully notifiable [26]. Laboratories in England with a primary diagnostic role are required to report cases of these pathogens to the UK Health Security Agency (UKHSA).

Our study population consisted of all residents in England who had a laboratory diagnosis of one of the three aforementioned GI infections, recorded between 1 January 2015 and 31 December 2020. Laboratory diagnoses were reviewed to ensure accurate case number calculations. Multiple diagnoses for the same individual, were grouped as a single infection episode within a pathogen-specific episode length (28 days for Campylobacter spp., 43 days for E. coli O157, and 93 days for non-typhoidal Salmonella spp.). Individuals were grouped using a combination of NHS number, date of birth, name, and sex. Diagnoses reported outside the specified episode length were classified as reinfections, representing new cases.

Data were collected from UKHSAs routine laboratory surveillance system (Second-Generation Surveillance System). English Index of Multiple Deprivation (2019; IMD) [27] and urban–rural residency classification (2011) [28] was linked to laboratory notifications on lower layer super output areas [29].

Statistical analysis

A weekly mean of cases was calculated for each GI infection for each of the five years (2015–2019) and presented with 95% confidence intervals. The weekly mean provided a comparator for cases registered in 2020 and was examined to provide insights into seasonality of each pathogen.

Time-series data (2015–2019) were decomposed by infection into trend, seasonal, and residual components. An additive decomposition was applied to visually inspect seasonal patterns. The long-term trend was estimated using seasonal-trend decomposition based on Loess (STL) [30] to obtain a more robust estimate.

A generalized linear model (GLM) was produced for each infection using negative-binomial regression to predict the number of cases. Each infection was further assessed for evidence of a trend using time as a predictor within the model. Seasonality was addressed via the addition of sine and cosine pairs and construction of periodograms. The autocorrelation and partial-autocorrelation functions were plotted to identify if the number of laboratory notifications were correlated with previous values in time series. Lag terms were added to the model to address autocorrelation. Likelihood ratio tests [31] were used to determine if additional predictors provided a significantly better fit to the observed data. A stepwise selection procedure based on the Akaike Information Criterion [32] was applied to determine an appropriate parsimonious model for each infection.

Additional GLM models were developed using Poisson fit and a log-link function to predict the rate of cases of each infection, stratified by urban–rural residency and IMD. These models utilized the log-transformed population of each stratum as offset variables.

Sudden deviations between the observed cases and each of the model’s predictions were examined to identify any unexplained factors influencing the incidence of laboratory notifications (change points). We used the pruned exact linear time method [33], assessing both the mean and variance of the deviations, to identify change points.

Specifically, we assessed whether these change points corresponded to the introduction of the national lockdown or the EOHO scheme. If a change point was identified at these time points, an interruption was added to the model. If no change points were identified corresponding to the EOHO scheme, we added a step change for the period following the beginning of the scheme (3 August 2020) to assess the schemes impact on predicting cases.

Statistical analyses were performed using R (version 4.3.2) [34]. A value of p < 0.05 was taken as statistically significant.

Results

There were 309,466 Campylobacter spp. infections, 42,018 non-typhoidal Salmonella spp. infections and 3,282 E. coli O157 infections, notified across England between 1 January 2015 and 31 December 2020, combining the pre-pandemic and pandemic phases. The majority of cases were notified pre-pandemic (1 January 2015 and 31 December 2019), accounting between 86% and 90% of cases across all three infections (Table 1). Cases from residents of urban areas formed between 69% and 84% of notifications for all three infections. The highest proportion cases for both Campylobacter spp. (23%) and E. coli O157 (26%) were from individuals living in the most deprived IMD quintile. In contrast, the highest proportion of cases for non-typhoidal Salmonella spp. (22%) were from those residing in the least deprived IMD quintile notified during 2020 (Table 1).

Table 1. Demographics of the population with gastrointestinal infections captured in England between 1 January 2015 and 31 December 2020.

Period Campylobacter spp. n (%) Escherichia coli O157 n (%)non-typhoidal Salmonella spp. n (%)
2015–201920202015–201920202015–20192020
Total266,521 (86)42,9452,974 (90)30837,989 (90)4,029
Residency
   Urban208,089 (78)33,414 (78)2,318 (78)212 (69)32,065 (84)3,330 (83)
   Rural58,4329,531656965,924699
IMD
   140,785 (15)7,049 (16)404 (14)35 (11)7,635 (20)877 (22)
   248,236 (18)8,034 (19)577 (19)52 (17)7,925 (21)850 (21)
   356,437 (21)9,174 (21)611 (21)66 (21)7,837 (21)866 (21)
   460,059 (23)9,338 (22)653 (22)76 (25)7,362 (19)782 (19)
   5 (most deprived)61,004 (23)9,350 (22)729 (25)79 (26)7,230 (19)654 (16)
Over the study period (1 January 2015 and 31 December 2020), we detected a significant decreasing trend in the number of cases of E. coli O157 (−0.002, 95% CI: −0.003 to −0.002, p = <0.001) (Figure 1). No significant trends were observed in case numbers of both Campylobacter spp. and non-typhoidal Salmonella spp. over the same period. A single seasonal peak was observed during the summer months for all three infections. The seasonal peak for non-typhoidal Salmonella spp. occurred 12 weeks later (Figure 2) than for the other two infections.
Figure 2.

Figure 2. Long-term trend components of weekly laboratory confirmed gastrointestinal infections in England (2015–2019), by infection. Trend components were derived using Seasonal-Trend Decomposition based on LOESS (locally estimated scatterplot smoothing; SLT), with a periodic seasonal window to account for annual patterns.

Cases across all infections in 2020 were lower than the historical mean over the previous five years for the majority of weeks in 2020 (between 69% and 79%). The lowest number of cases was observed between weeks 7 and 20 (10 February 2020 and 17 May 2020) (Figure 3). Cases of Campylobacter spp. and E. coli O157 did increase to levels consistent with the historical mean during summer (between 10 August 2020 and 30 August 2020), corresponding to the period of EOHO. In contrast, over the same period, non-typhoidal Salmonella spp. cases remained substantially below the historic mean.
Figure 3.

Figure 3. Weekly incidence of laboratory confirmed gastrointestinal infections in England during 2020, compared to the historical weekly average (2015–2019) with 95% confidence intervals.

We did not find any significant association between the implementation of the EOHO scheme and a change in the number of cases for any of the infections (Table 2).

Table 2. Coefficient estimates from a negative binomial regression model examining the impact of the ‘Eat Out to Help Out’ scheme on selected gastrointestinal infections in England in 2020

InfectionEstimate95% confidence interval P-value
LowerHigher
Campylobacter spp.−0.028−0.0920.038>0.05
Escherichia coli O1570.025−0.1950.242>0.05
Non-typhoidal Salmonella spp.−0.187−0.313−0.061<0.01
Our change point analysis showed distinct shifts in the incidence of Campylobacter spp. and non-typhoidal Salmonella spp. in 2020. A significant reduction in cases occurred between week 10 (02 March 2020) and week 11 (09 March 2020), followed by a subsequent increase between week 18 (27 April 2020) and week 20 (17 May 2020) (Figure 4). No significant change points were detected for E. coli O157.
Figure 4.

Figure 4. Weekly laboratory confirmed Campylobacter spp. and non-typhoidal Salmonella spp. infections in England during 2020, with identified change-points (detected using the Pruned Exact Linear Time method). Change-points reflect shifts in the mean and variance of residual deviations between observed counts and expected counts from a negative binomial generalized linear model.

The analysis confirmed there were no unexpected changes in the incidence of any of the infections during the EOHO scheme.

Analysis of the dataset split by rural and urban residency, found no subsequent increase in cases of non-typhoidal Salmonella spp. among rural residents following the reduction in week 10 (02 March 2020) (Figure 5).
Figure 5.

Figure 5. Weekly laboratory confirmed Campylobacter spp. and non-typhoidal Salmonella spp. infections in England during 2020, with identified change-points stratified by rural and urban residency. Change-points, identified using the Pruned Exact Linear Time method, reflect shifts in the mean and variance of residual deviations between observed rates and expected rates from a negative binomial generalized linear model, using population of each stratum as an offset variable.

When stratified by IMD, change point in cases captured in 2020 for both Campylobacter spp. and non-typhoidal Salmonella spp. were inconsistent.

Discussion

Our analysis did not identify a rise in the incidence of selected GI infections following the introduction of the EOHO scheme which encouraged dining out in August 2020. Instead, the observed reductive change points demonstrated the broader impact of NPIs implemented to mitigate COVID-19 on communicable diseases. While significant increases in GI infections were detected, these were aligned with the easing of NPIs rather than the EOHO scheme.

Public health measures aimed at controlling COVID-19 have had widespread unintended consequences, influencing the incidence of infectious diseases with varying transmission routes [35, 36]. A key contributor to the decreases in GI observed in early 2020 will have been substantial changes in healthcare seeking behaviours as a result of the first national lockdown [37]. Notably, the impact of healthcare seeking behaviours on GI cases were less pronounced in subsequent lockdowns (5 November 2020 to 2 December 2020). These effects align with documented findings [16, 38] and are supported by our study. The observed increase in GI infections coinciding with the easing of the national lockdown [38], reinforces the plausibility that GI infection resurgence was influenced by changes in population movement [16, 38].

The temporal and spatial overlap in the implementation and easing of NPIs aimed at controlling COVID-19 posed a challenge in determining any specific impact of a single event on GI infections over this period in time. For instance, while restaurants were permitted to reopen on 4 July 2020 with social distancing, which was subsequently eased on 19 July 2020, the scheme did not begin until August 2020 [15]. While we did not identify any direct impact of the EOHO scheme, evaluating its effect with sufficient statistical power was challenging because of the difficulty in disentangling the scheme’s specific influence from other concurrent public health control measures.

Multifactorial COVID-19 control measures disrupted multiple communicable disease transmission routes beyond foodborne infections, including person-to-person spread and importation [35, 36]. The true impact of the EOHO scheme on GI infection transmission may have been confounded by these control measures, notably social distancing and improved hand hygiene [39]. Furthermore, it is plausible that genuine increases attributable to the scheme were offset by control measures that significantly disrupted transmission routes prior to its implementation. For example, substantial declines in international travel during 2020 likely reduced travel associated infections, inclusive of Campylobacter spp., E. coli O157, and non-typhoidal Salmonella spp. [2, 40]. In addition, high-risk settings associated with E. coli O157 outbreaks, specifically nurseries and petting farms [40], were closed as part of the national lockdown and did not reopen until June 2020 [15].

The seasonality of GI infections, typically peaking during the summer months [5], coincided with the introduction of the EOHO scheme. It is plausible that true increases in incidence of laboratory notifications following the scheme’s launch (between 10 August 2020 and 30 August) (Figure 1) may have been masked by natural seasonal trends.

Figure 1.

Figure 1. Seasonal component of decomposed additive time series of weekly laboratory confirmed gastrointestinal infections in England (2015–2019), by infection.

National surveillance data does not capture all GI infections in the community as healthcare-seeking behaviours vary, which was notable during the COVID-19 pandemic [38, 41]. However, given the EOHO scheme lasted only four weeks, it is unlikely that significant changes in healthcare-seeking behaviours occurred.

While we identified change points for both Campylobacter spp. and non-typhoidal Salmonella spp., none were detected for E. coli O157. This may reflect limited statistical power, as annual case numbers for E. coli O157 were steadily declining while other GI infections remained stable prior to the COVID-19 pandemic. The observed decreasing trend in E. coli O157 cases over the study period aligns with broader epidemiological patterns [42]. Alternative modelling approaches may clarify whether changes in NPIs affected the incidence of E. coli O157.

The absence of a prompt resurgence in non-typhoidal Salmonella spp. cases among rural residents post-lockdown may be partly explained by differences in food consumption habits compared to urban areas. Rural populations generally consume less food prepared outside the home [43], which is notable given the majority of foodborne disease outbreaks can be attributed to Non-typhoidal Salmonella spp. [44]. Additionally, the rise of ‘dark kitchens’, (restaurants operating exclusively via delivery services) in urban areas, during COVID-19 [45] may have further influenced infection trends. These establishments have been associated with reduced trust in food safety and hygiene standards [46]. Managing food safety is a key challenge for such businesses, as they often operate in shared kitchens [47].

Our change point analysis allowed the data to determine where interruptions should be added, in contrast to prespecified interruptions. Compared to traditional methods which impose interruptions a priori, this approach objectively identified deviations between the model and the observed data. This offered an evidence-based assessment of the EOHO scheme’s impact on GI infections.

Our study used national data from a routine laboratory surveillance system, providing a large sample. The use of historical data over an extended period allowed for a robust baseline comparator to assess laboratory notifications in 2020 amid system changes. The inclusion of multiple GI infections offered a comprehensive evaluation of the EOHO scheme’s potential impact on the incidence of foodborne pathogens.

Stratifying the analysis by IMD reduced statistical power, limiting our ability to detect change-points. Moreover, IMD reflects socioeconomic conditions and is weighted on living environment, consequently making it confounded by urban and rural residency [27]. This confounding poses limits on our interpretation of the two characteristics independently.

Our study only analysed case data and did not review whether the EOHO scheme had an effect on the number of reported foodborne GI outbreaks. As not all GI infections are foodborne, incorporating such data may have provided a clearer assessment of the schemes impact. However, prior studies demonstrate that restaurant-associated foodborne outbreaks declined in conjunction with the introduction of NPIs [16].

Our models did not account for the phased implementation and easing of COVID-19 control measures beyond the first national lockdown, prior to the introduction of the scheme. These measures may have influenced GI infection trends (e.g., phased reopening of schools (1 June 2020), international travel corridors (8 June 2020), and reopening of restaurants and pubs (4 July 2020)).

Conclusion

We found no evidence that the introduction of the EOHO scheme impacted the incidence of Campylobacter spp., E. coli O157, and non-typhoidal Salmonella spp. cases. The introduction of the first nationwide lockdown was followed by significant reductions in cases, with a prompt increase observed as enforceable restrictions eased. These findings emphasize the complex interactions between public health measures, behavioural changes, and the incidence of GI pathogens.

Acknowledgements

The authors would like to thank Vicky Dowling from Field Services at UK Health Security Agency for their support in during in conceptualising this work.

Funding statement

The protocol, analysis, and manuscript of this study were the product of a training workshop on operational research for frontline public health practitioners and attended by the first author. The workshop was funded by the National Institute for Health Research Health Protection Research Unit (NIHR HPRU) in Gastrointestinal Infections. The views expressed are those of the authors and not necessarily those of the NIHR, UK Health Security Agency, or the Department of Health and Social Care. The work associated with the study was carried out as part of routine public health work with no additional funding.

Competing interests

The authors declare none.

Ethical standard

Research governance approval for this study was granted by UKHSA Research Ethics and Governance Group (reference R&D602) on 9th December 2024. All data were handled and stored in accordance with UKHSA information governance and securities policies.

References

  1. [1] Pires SM and Devleesschauwer B (2020) Chapter 1 - estimates of global disease burden associated with foodborne pathogens. In Morris JG and Vugia DJ (eds.), Foodborne Infections and Intoxications, 5th Edn. Florida: Academic Press, pp. 3–17.
  2. [2] Lee H and Yoon Y (2020) Etiological agents implicated in foodborne illness world wide. Food Science of Animal Resources 41, 1–7.10.5851/kosfa.2020.e75PMC781039533506212
  3. [3] Tam CC, et al. (2012) Longitudinal study of infectious intestinal disease in the UK (IID2 study): Incidence in the community and presenting to general practice. Gut 61, 69.21708822 10.1136/gut.2011.238386PMC3230829
  4. [4] Nicholas D, et al. (2020) The Burden of Foodborne Disease in the UK 2018: Analytics Unit, Food Standards Agency and London School of Hygiene & Tropical Medicine. https://www.food.gov.uk/sites/default/files/media/document/the-burden-of-foodborne-disease-in-the-uk_0.pdf (accessed 27 June 2024).
  5. [5] Cherrie MPC, et al. (2018) Pathogen seasonality and links with weather in England and Wales: A big data time series analysis. BioMed Central Public Health 18, 1067.30153803 10.1186/s12889-018-5931-6PMC6114700
  6. [6] Interagency Food Safety Analytics Collaboration (2023) Foodborne Illness Source Attribution Estimates for 2021 for Salmonella, Escherichia coli O157, and Listeria monocytogenes Using Multi-Year Outbreak Surveillance Data, United States. Georgia and Washing D.C.: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Food and Drug Administration, U.S. Department of Agriculture’s Food Safety and Inspection Service. https://www.cdc.gov/ifsac/media/pdfs/P19-2021-report-TriAgency-508.pdf (accessed 27 June 2024).
  7. [7] Food Standards Agency (2022) Enhanced Molecular-Based Surveillance and Source Attribution of Campylobacter Infections in the UK. Oxford University. 10.46756/sci.fsa.ksj135.
  8. [8] Buczkowska M, et al. (2023) Association between socioeconomic deprivation and incidence of infectious intestinal disease by pathogen and linked transmission route: An ecological analysis in the UK. Epidemiology and Infection 151, 109.10.1017/S0950268823000869PMC1036942537313601
  9. [9] Love NK, et al. (2024) International travel as a risk factor for gastrointestinal infections in residents of north East England. Epidemiology and Infection 27, 152.10.1017/S0950268824000827PMC1173644738801061
  10. [10] Adams NL, et al. (2018) Socioeconomic status and infectious intestinal disease in the community: A longitudinal study (IID2 study). European Journal of Public Health 28, 134–138.29016791 10.1093/eurpub/ckx091PMC5965370
  11. [11] Oldroyd RA, Morris MA and Birkin M (2020) Food safety vulnerability: Neighbourhood determinants of non-compliant establishments in England and Wales. Health & Place 63, 102325.32217280 10.1016/j.healthplace.2020.102325
  12. [12] Bessell PR, et al. (2010) Geographic determinants of reported human campylobacter infections in Scotland. BioMed Central Public Health. 10, 423.20633277 10.1186/1471-2458-10-423PMC2918555
  13. [13] Newman KL, et al. (2015) The impact of socioeconomic status on foodborne illness in high-income countries: A systematic review. Epidemiology and Infection. 143, 2473–2485.25600652 10.1017/S0950268814003847PMC4508232
  14. [14] Dunn P, et al. (2021) A Timeline of National Policy and Health System Responses to COVID-19 in England in 2020. The Health Foundation. https://www.health.org.uk/reports-and-analysis/analysis/covid-19-policy-tracker-2020.
  15. [15] Institute for Government (2022) Timeline of UK government coronavirus lockdowns and measures, March 2020 to December 2021. https://www.instituteforgovernment.org.uk/data-visualisation/timeline-coronavirus-lockdowns (accessed 29 June 2024).
  16. [16] Love NK, et al. (2023) Understanding the impact of the COVID-19 pandemic response on GI infection surveillance trends in England, January 2020–April 2022. Epidemiology and Infection 151, 147.10.1017/S095026882300136XPMC1054016837622322
  17. [17] Ramuni L (2021) Coronavirus and its impact on UK hospitality: January 2020 to June 2021. Office for National Statistics.
  18. [18] Office for National Statistics (2021) Coronavirus: how people and businesses have adapted to lockdowns. https://www.ons.gov.uk/economy/economicoutputandproductivity/output/articles/coronavirushowpeopleandbusinesseshaveadaptedtolockdowns/2021-03-19 (accessed 25 June 2024).
  19. [19] Hutton G (2020) Research Briefing Eat out to Help out Scheme. Houses of Commons Library UK Parliament.
  20. [20] Hutton G, Irvine S, Foley N (2022) Hospitality industry and Covid-19. In House of Commons Library. London: UK Parliament.
  21. [21] Food Standards Agency (2022) Our Food 2021: An Annual Review of Food Standards across the UK. Food Standards Agency, London. https://www.food.gov.uk/our-work/annualreviewoffoodstandards2021 (accessed 3 July 2024).
  22. [22] Clayton ML, et al. (2015) Listening to food workers: Factors that impact proper health and hygiene practice in food service. International Journal of Occupational Environmental Health. 21, 314–327.26243248 10.1179/2049396715Y.0000000011PMC4727591
  23. [23] Fetzer T (2021) Subsidising the spread of COVID-19: Evidence from the UK’S eat-out-to-help-out scheme*. The Economic Journal. 132, 1200–1217.
  24. [24] McCloskey B, et al. (2014) London 2012 Olympic and paralympic games: Public health surveillance and epidemiology. The Lancet. 383, 2083–2089.10.1016/S0140-6736(13)62342-9PMC713802224857700
  25. [25] Hughes H, et al. (2023) Development of an enhanced surveillance and epidemiology program to support the Birmingham 2022 commonwealth games in England: Outcomes and lessons learned. Prehospital and Disaster Medicine. 38, s69–s39.
  26. [26] The National Archives (2010) The Health Protection (Notification) Regulations. https://www.legislation.gov.uk/uksi/2010/659/contents (accessed 4 July 2024).
  27. [27] Office for National Statistics (2019) The English Indices of Deprivation 2019. https://www.gov.uk/government/statistics/english-indices-of-deprivation-2019 (accessed 24 July 2025).
  28. [28] Office for National Statistics (2016) 2011 rural/urban classification. https://www.ons.gov.uk/methodology/geography/geographicalproducts/ruralurbanclassifications/2011ruralurbanclassification (accessed 24 July 2024).
  29. [29] Office for National Statistics (2021) Census 2021 geographies. Available at https://www.ons.gov.uk/methodology/geography/ukgeographies/censusgeographies/census2021geographies (accessed 24 July 2024).
  30. [30] Cleveland R, et al. (1990) STL: A seasonal-trend decomposition procedure bases on loess. Journal of Official Statistics 6, 1.
  31. [31] Washington S, et al. (2020) Chapter 13 - discrete outcome models. In Statistical and Econometric Methods for Transportation Data Analysis (3rd Edn). New York: Chapman and Hall/CRC Press, pp. 255–285.
  32. [32] Sutherland C, et al. (2023) Practical advice on variable selection and reporting using Akaike information criterion. Proceedings of the Royal Society B: Biological Sciences 290, 20231261.10.1098/rspb.2023.1261PMC1052307137752836
  33. [33] Gachomo D (2015) The power of the pruned exact linear time(PELT) test in multiple changepoint detection. American Journal of Theoretical and Applied Statistics 4, 581.
  34. [34] R Development Core Team (2023) R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing.
  35. [35] Nash K, et al. (2022) Impact of national COVID-19 restrictions on incidence of notifiable communicable diseases in England: An interrupted time series analysis. BioMed Central Public Health 22, 2318.36510205 10.1186/s12889-022-14796-0PMC9742638
  36. [36] Hayes LJ, et al. (2023) Impact of the COVID-19 pandemic on the circulation of other pathogens in England. Journal of Medical Virology 95, e28401.36512325 10.1002/jmv.28401PMC9877778
  37. [37] Elliot AJ, et al. (2020) The COVID-19 pandemic: A new challenge for syndromic surveillance. Epidemiology and Infection 148, e122.32614283 10.1017/S0950268820001314PMC7338397
  38. [38] Love NK, et al. (2022) Impact of the COVID-19 pandemic on gastrointestinal infection trends in England, February-July 2020. British Medical Journal Open 12, e050469.10.1136/bmjopen-2021-050469PMC896811135314468
  39. [39] Lazarakou A, Mughini-Gras L and Pijnacker R (2024) Global impact of COVID-19 pandemic on gastrointestinal infections: A scoping review. Foodborne Pathogens and Disease, 22(10), 669–683. 10.1089/fpd.2024.0047.39588900 PMC12629675
  40. [40] King G, et al. (2025) Epidemiology of Shiga toxin-producing Escherichia coli other than serotype O157:H7 in England, 2016-2023. Journal of Medical Microbiology 74, 001947.39791999 10.1099/jmm.0.001947PMC11721027
  41. [41] Hughes HE, et al. (2020) Emergency department use during COVID-19 as described by syndromic surveillance. Emergency Medicine Journal 37, 600–604.32948621 10.1136/emermed-2020-209980PMC7503196
  42. [42] Butt S, et al. (2022) The epidemiology of Shiga toxin-producing Escherichia coli serogroup O157 in England, 2009–2019. Epidemiology and Infection 150, e52.35241189 10.1017/S0950268822000206PMC8915195
  43. [43] d’Angelo C, et al. (2020) Food Consumption in the UK: Trends, Attitudes and Drivers. Santa Monica, California: RAND Corporation. https://www.rand.org/pubs/research_reports/RR4379.html (accessed 17 March 2025).
  44. [44] Department for Environment Food and Rural Affairs (2024) United Kingdom Food Security Report 2024: Theme 5: Food Safety and Consumer Confidence: GOV.UK. https://www.gov.uk/government/statistics/united-kingdom-food-security-report-2024/united-kingdom-food-security-report-2024-theme-5-food-safety-and-consumer-confidence#foodborne-pathogen-surveillance (accessed 17 March 2025).
  45. [45] Gargiulo AH, et al. (2022) Food Safety Issues Related to Eating in and Eating Out. Microorganisms. 10.3390/microorganisms10112118.PMC969555936363709
  46. [46] Nield L, et al. (2025) Consumer knowledge of and engagement with traditional takeaway and dark kitchen food outlets. National Institue for Health and Care Research Open Research 4, 64.10.3310/nihropenres.13735.2PMC1192068640109407
  47. [47] Laheri Z, Ferris I and Soon-Sinclair JM (2025) The rise of dark kitchens: Characteristics and operational challenges. International Journal of Gastronomy and Food Science 40, 101142.

Declarations

Data Availability

The data used in this study is sensitive and was collected under Regulation 3 of the Health Service (Control of Patient Information) Regulations 2002, to process patient confidential information for national surveillance of communicable diseases. It is therefore not publicly available. Applications for relevant anonymized data should be submitted to the UKHSA Office for Data Release.

Download
PDFJATS XML