Research

The impact of COVID-19 lockdowns on primary care contact among vulnerable populations in England: a controlled interrupted time-series study

Scott R Walter, Chris Salisbury, Lauren J Scott, Frank de Vocht, John Macleod, Yoav Ben-Shlomo, Helen J Curtis, Aziz Sheikh, Srinivasa V Katikireddi, Amir Mehrkar, Sebastian Bacon, George Hickman, Ben Goldacre, The OpenSAFELY Collective and Maria Theresa Redaniel
BJGP Open 2025; 9 (4): BJGPO.2025.0017. DOI: https://doi.org/10.3399/BJGPO.2025.0017

Abstract

Background UK COVID-19 lockdowns significantly affected primary care access and delivery. Little is known about whether lockdowns disproportionally impacted vulnerable groups, including people who misuse substances, people who have experienced domestic violence or abuse, those with intellectual disability, and children with safeguarding concerns.

Aim To evaluate the impact of UK COVID-19 lockdowns on primary care contact rates among vulnerable groups.

Design & setting Natural experimental design using all registered patients in the OpenSAFELY platform.

Method With approval from NHS England, we conducted controlled interrupted time-series analyses on records from 24 million patients in England between September 2019 and September 2021.

Results Pre-pandemic, primary care contact rates were 110.1 per 1000 patients per week. Following the initiation of the first lockdown (23 March 2020), there was a large reduction of 29–61 contacts per 1000 patients per week among vulnerable and general population groups. For patients with alcohol misuse, those aged ≥14 years with intellectual disability, and children with safeguarding concerns, this reduction was significantly more extreme than corresponding general populations (relative rate difference -23.8 [95% confidence interval {CI} = -39.8 to -7.7, P = 0.003], -24.6 [95% CI = -38.8 to -10.5, P<0.001], and -15.4 [95% CI = -26.9 to -3.8, P = 0.009], respectively). Following the final lockdown (29 March 2021), all groups had contact rates exceeding pre-pandemic rates (with increases more marked in vulnerable populations), except those only including children.

Conclusion Our results suggested a larger short-term impact of the first COVID-19 lockdown on primary care contact for some vulnerable groups, compared with the general population, and differential impacts persisted through subsequent lockdowns and beyond for some vulnerable groups. There is a need to examine drivers of these differences to enable more equitable primary care access and provision.

How this fits in

There is limited knowledge on how the COVID-19 pandemic and resultant lockdowns affected primary care contact for vulnerable people. We used primary care records from 24 million patients in England to evaluate the impacts of the UK COVID-19 lockdowns on primary care service use in vulnerable groups, including patients with issues of substance misuse or dependence, those who had experienced domestic violence or abuse, patients with intellectual disability, and children with safeguarding concerns, compared with the rest of the population. Following the initiation of the first lockdown (23 March 2020), there was a large reduction in primary care contacts for all vulnerable and non-vulnerable patient groups. For patients with alcohol misuse issues, those aged ≥14 years with intellectual disability, and children with safeguarding concerns, this reduction was significantly larger than the corresponding general population. Differential effects persisted through subsequent lockdowns. It is important to understand the drivers of the observed differences to enable more equitable primary care access and provision, both now and in the event of a future pandemic or large-scale change in availability of primary care for other reasons.

Introduction

The impact of the COVID-19 pandemic has been profound and far-reaching, with around 70% of people in England infected and more than 200 000 COVID-19-related deaths.1–3 The inequity of those impacts has been well documented, where those with pre-existing morbidities or disabilities, and people from certain ethnic minorities, regions or age groups were disproportionately affected.4

The lockdowns in response to the pandemic in the UK dramatically changed health service provision and use. There is widespread evidence of reduced healthcare access at the outset of the first national UK lockdown, including a decrease in hospital admission rates, particularly cancer admissions,5,6 and accident and emergency presentations.7–9

In the initial lockdown, there was an accelerated shift towards remote primary care consultations,10 as well as an overall decrease in the number of consultations.10–13 Differential decreases in primary care contacts have been observed for patients depending on their condition, with diabetic emergency and depression among the most affected.12 Foley et al 11 reported a 41% reduction in contacts among children and young people. In contrast, people with poor mental health, those who were shielding, and those aged ≥85 years had similar or higher consultation rates through the initial lockdown period compared with the same period the previous year.10 A potential knock-on effect of reduced consulting is a decline in rates of diagnoses for physical and mental health conditions.13

On average, the general population has around five primary care appointments per year.14–16 However, those in vulnerable groups often have higher contact (for example, people who misuse substances often have higher rates of physical17 and mental18 health disorders than those who do not misuse substances, and people who use illicit drugs may require regular contact with primary care to be issued substitute drug prescriptions). There is little evidence on how the pandemic and resultant lockdowns affected primary care contact in vulnerable groups.

This study aimed to assess changes in primary care contact rates in England through changing lockdown states among vulnerable populations, including people who misuse substances, people who experience(d) domestic violence or abuse (DVA), those with intellectual disability, and children with safeguarding concerns, relative to the general population.

Method

Study design

This study used a controlled interrupted time-series (CITS) design applied to a cohort of approximately 24 million registered patients (approximately 40% of the population)19 from English primary care practices that use TPP SystmOne software.

Eight vulnerable populations were studied: patients with alcohol misuse; patients with drug misuse; those with opioid dependence; females who had experienced DVA; males who had experienced DVA; people with intellectual disability aged <14 years; those with intellectual disability aged ≥14 years; and children (aged <18 years) with safeguarding concerns. These populations were chosen as they represent groups that have not been previously studied in this context and who were possible to identify using codelists.

Two sets of analyses were performed exploring the first UK COVID-19 lockdown (23 March 2020–09 May 2020),20 and the second and third UK lockdowns combined as a single period (05 November 2020–28 March 2021).21,22 Both analyses included 6-months of data before and after these dates (Figure 1; some control periods were truncated by a few days so as not to include active lockdown periods).

Figure 1.
Figure 1. Timeline of UK national COVID-19 lockdowns and the analysis periods used in this study

Data source

The OpenSAFELY platform was established to support urgent research into the COVID-19 emergency with the Control of Patient Information (COPI) notice forming the legal basis for its establishment.23 All data were linked, stored and analysed securely within the OpenSAFELY platform.24 Data include pseudonymised data such as coded diagnoses, medications, and physiological parameters. All code is shared openly for review and re-use under an MIT open licence.25

Population

Codelists were used to identify vulnerable populations25 (Supplementary material 1). For a given codelist, any registered patient with a matching code in their primary care record during a 1-year lookback period (this was a moving period, with the cohort of interest updated each week) was classified as belonging to that vulnerable population, while all patients without that vulnerability formed the general comparison population. For vulnerable groups defined by age or sex, the comparison population were those within the same age or sex groups but without the relevant vulnerability.

Outcome

The primary outcome was the rate of all primary care contacts per 1000 patients per week, constructed as the weekly count of contacts divided by the number of registered patients at the beginning of each week. The data were derived from the TPP appointments table, which is not necessarily an exact record of face-to-face or virtual consultations, but is considered a better approximation than the alternative of deriving data from consultation tables.26

Statistical methods

Within a CITS framework, linear regression models analysed each of the eight vulnerable groups and two analysis periods (16 models in total). The models estimated a step change (the immediate change) and slope change (the average weekly change) in the vulnerable and general populations, at the start and end of the lockdown periods (Supplementary material 2 & 3). Results are presented as contact rate differences (RDs) per 1000 patients and their corresponding 95% confidence intervals (CIs) and P values for the vulnerable populations, the comparative general populations, and the relative RD between the two.

Models were adjusted for public holidays since contact rates were markedly lower during these periods. Seasonal effects were not accounted for since there was little evidence of seasonal change in the outcome (in sensitivity analyses, including calendar month had little impact on estimates). Although there may be differences in factors, such as age distribution and deprivation between vulnerable and general populations, these are reasonably stable over time so would not impact estimates of change when modelling weekly aggregated data.

Analyses were performed using Python (version 3.9) and Stata MP (version 17); all analysis codes are recorded on Github.25

Results

Rates of primary care contact before the first UK lockdown

In the pre-lockdown period, the crude primary care contact rate in the general population was 110.1 per 1000 patients per week; women were more likely to have contact with their GP than men, and adults more likely than children (Table 1).

View this table:
Table 1. Crude rates of primary care contact per registered 1000 patients per week, by vulnerable group and time period

Rates of primary care contact throughout the study period were generally higher among vulnerable groups compared with the rest of the population (Table 1). In particular, contact rates for patients with substance misuse, female patients with experience of DVA, and patients aged ≥14 years with intellectual disability were approximately 2–3 times higher than their comparative general populations. In contrast, children with safeguarding concerns had about half the number of primary care contacts compared with children who had no record of safeguarding concerns.

Following the final lockdown, primary care contact rates were slightly higher than pre-pandemic rates in most vulnerable and general population groups, but patients with substance misuse issues and females who experienced DVA had considerably higher contact rates than before the first lockdown (Table 1). For patients aged <14 years with intellectual disability and children with safeguarding concerns, post-lockdown contact rates remained lower than pre-lockdown rates, as did the rates in their corresponding general populations (Table 1).

Controlled interrupted time-series analysis

Analysis period one: first UK lockdown (23 September 2019–1 November 2020)

When the UK first locked down, there was a large reduction in primary care contacts of 29–61 contacts per 1000 patients per week among vulnerable groups and 29–39 contacts per 1000 patients in the general population (step change 1 in Table 2). For patients with alcohol misuse issues, those with intellectual disability aged ≥14 years, and children with safeguarding concerns, the decrease was more significant compared with general populations (relative RD -23.8 [95% CI = -39.8 to -7.7, P = 0.003], -24.6 [95% CI = -38.8 to -10.5, P<0.001], and -15.4 [95% CI = -26.9 to -3.8, P = 0.009], respectively; step change 1 in Table 2).

View this table:
Table 2. Rate difference estimates (per 1000 patients) of step and slope changes from controlled interrupted time-series modelling for the period of the first national UK lockdown

Following the initial sharp decrease, contact rates for most groups gradually rebounded through the first lockdown. Rates for those with alcohol misuse, drug misuse, opioid dependence, and those aged ≥14 years with intellectual disability increased significantly more throughout the first lockdown than their general population comparators (relative weekly change in RD 3.7 [95% CI = 0.5 to 7.0, P = 0.02], 4.9 [95% CI = 1.2 to 8.5, P = 0.009], 4.8 [95% CI = 1.2 to 8.4, P = 0.008], and 3.2 [95% CI = 0.7 to 5.6, P = 0.01], respectively; slope change 1 in Table 2), with contact rates for patients in all substance misuse categories exceeding pre-COVID rates following the first lockdown (Table 1). However, for males experiencing DVA, this rebound happened a little later following the end of the first lockdown (Table 1 and slope and step change 2 in Table 2). The rebound effect was less apparent for children, where there was a further decrease in contact rates at the end of the lockdown period (for children with safeguarding concerns, intellectual disability aged <14 years, and their corresponding general population comparators), but no significant difference between vulnerable patients and the general population were seen (step change 2 in Table 2).

Analysis period two: second and third UK lockdowns (11 May 2020–26 September 2021)

There was no evidence of a step change at the start of the second lockdown in any group. In drug misuse, opioid dependence, and female DVA groups, there was a significant increasing trend in contact rate during the period when the second and third lockdowns were in effect, compared with the general population (relative weekly chainge in RD 2.5 [95%CI = 1.1 to 3.9, P<0.001], 3.9 [95% CI = 2.4 to 5.4, P<0.001], and 2.0 [95% CI = 0.8 to 3.1, P = 0.001], respectively; slope change 3 in Table 3). At the end of the final lockdown, a step decrease was observed among alcohol misuse, drug misuse, opioid dependence, female DVA, and those aged ≥14 years with intellectual disability, in comparison to a step increase in the general population (relative RD -20.7 [95%CI = -34.4 to -7.0, P = 0.003], -33.9 [95% CI = -51.4 to -16.4, P<0.001], -41.3 [95% CI = -57.6 to -25.1, P<0.001], -23.2 [95% CI = -38.4 to -8.0, P = 0.003], and -40.4 [95% CI = -59.9 to -20.9, P<0.001], respectively; step change 4 in Table 3). For three of these groups — drug misuse, opioid dependence, and female DVA — this was followed by a final slope decrease over and above their corresponding general populations following restrictions easing after the final lockdown (relative weekly change inRD -3.1 [95%CI = -4.6 to -1.6, P<0.001], -4.1 [95% CI = -5.6 to -2.5, P<0.001], and -1.6 [95% CI = -2.8 to 0.4, P = 0.01], respectively; slope change 4 in Table 3).

View this table:
Table 3. Rate difference estimates (per 1000 patients) of step and slope changes from controlled interrupted time-series modelling for the period of the combined second and third national UK lockdowns

Discussion

Summary

Pre-pandemic rates of primary care contact for vulnerable groups were more frequent than the general population, except for children with safeguarding concerns who had lower contact rates than their equivalent general population. All patients experienced a significant drop in primary care contact at the start of the first lockdown, and for those groups with alcohol misuse, those aged ≥14 years with intellectual disability, and children with safeguarding concerns, this drop was significantly larger than their general population comparators. Rates then steadily increased, with timings differing slightly between groups, and by the end of final lockdown, most general populations and vulnerable groups had contact rates exceeding those of before the first lockdown (with increases more marked in the vulnerable populations). For vulnerable children and their general population counterparts, post-pandemic contact rates remained lower than pre-pandemic rates.

Strengths and limitations

This study used a large, nationally representative19 sample of 24 million patients (with the least populous vulnerable group comprising more than 8000 patients) and focused on vulnerable groups who have not previously been explored in this context. The CITS approach, with weekly-level temporal resolution, is a useful method for evaluating the effects of natural experiments on health outcomes.27

A potential limitation is that contact rates may not be reliably reported. For example, cancelled appointments may sometimes still be counted, and ad hoc patient contact, such as an unscheduled phone call to a patient, may not be recorded. However, the pre-pandemic rate of 5–6 contacts with primary care per patient per year in our study population is similar to the rates reported in other studies (around 4–5 contacts per patient per year).14–16,28 The issue with cancelled appointments may be more problematic in some vulnerable groups if they are more likely to cancel appointments than others; unfortunately we were unable to explore this further using the OpenSAFELY platform. The dynamic 1-year lookback period to define vulnerable groups may underestimate some cohorts, such as intellectual disability, where codes may not be continually recorded in primary care records following initial diagnosis. However, given that these issues, along with any other confounding issues, are likely to be reasonably constant over time, they should not have interfered with our analyses.

Primary care delivery and factors affecting access to care both changed during the pandemic in response to the pandemic itself and the lockdowns, which may have influenced the way data were recorded. The extent this impacted on results in unknown; however, we did not observe evidence of such non-lockdown changes affecting either vulnerable or general populations.

Comparison with existing literature

Several studies documented changes in care during the first UK lockdown, including decreases in hospital admissions,5 the extent of disruption experienced by patients,29 and the rapid shift to online consultations.10 In the primary care context, studies of lockdown-related impacts are broadly consistent with our observed decreases in contact. Mansfield et al 12 reported the odds ratios for post- versus pre-lockdown contact rates ranging from 0.98 for acute alcohol-related events to 0.35 for diabetic emergency, while Murphy et al reported an 11% decrease in GP consultations between April and July 2020,10 compared with the same period the previous year.

A potential knock-on effect of reduced contact is reduced diagnosis rates (for example Williams et al 13 estimated decreases in diagnosis rates ranging from 16% for cancer to 50% for common mental health conditions during the first lockdown), which could lead to higher morbidity and mortality.

Attempts were made to address anticipated reductions in contact in some vulnerable groups. For example, telephone consultation availability was increased for patients with issues of drug misuse or dependence, and enhanced outreach services enabled continued service provision, albeit modified.30 Some opioid substitution therapy programmes continued through the lockdowns with changes to therapy provision.30 These may explain the reduced differential impact of the first lockdown on these groups.

There was evidence of increased DVA incidence under lockdown conditions31 and reduced referrals from GPs to the Identification and Referral to Improve Safety (IRIS) programme.32 The shift to remote consultations created challenges for GPs to provide DVA support; in response, practices implemented adaptations around DVA support messaging along with integrated approaches to identifying potential DVA cases.31

Implications for research and practice

While the initial decreased contact for all groups was likely to be largely attributable to changes in how primary care was provided and accessed, along with government messaging to stay home and reduced infectious disease circulation, there are likely multiple drivers of the differential effects observed across vulnerable groups. As part of the rapid adaptation of primary care service provision with the first lockdown, there was a general shift towards prioritising the highest need patients,10,33 but the extent to which this prioritisation was proactive by healthcare providers or reactive based on patient demands is unclear and potentially varied widely.

Our findings indicate evidence of differential impacts of the UK COVID-19-related lockdowns on primary care contact for vulnerable groups. While the largest impacts were observed at the start of the first lockdown, evidence of differential impact persisted through subsequent lockdowns and beyond for some groups, with the most pronounced changes in the later phase occurring around end of the third lockdown. This study lays the foundation for exploring factors driving differences in primary care access and provision, not just during the pandemic but also examining inequities that may have persisted since.

Notes

Funding

This project was funded by the National Institute for Health and Care Research (NIHR) Applied Research Collaboration (ARC) West and the West of England Academic Health Sciences Network (WEAHSN) as part of the NHS Insight Prioritisation Programme (NIPP). Funding was also provided by the Longitudinal Health and Wellbeing National Core Study (MC_PC_20030, MC_PC_20059), and the CONVALESCENCE study (characterisation, determinants, mechanisms and consequences of the long-term effects of COVID-19: providing the evidence base for health care services) funded by NIHR (COV-LT-0009). The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care. Research in OpenSAFELY uses data assets made available as part of the Data and Connectivity National Core Study, led by Health Data Research UK in partnership with the Office for National Statistics and funded by UK Research and Innovation (grant ref MC_PC_20058). In addition, the OpenSAFELY Platform is supported by grants from the Wellcome Trust (222097/Z/20/Z); MRC (MR/V015737/1, MC_PC-20059, MR/W016729/1); NIHR (NIHR135559, COV-LT2-0073), and Health Data Research UK (HDRUK2021.000, 2021.0157). FDV is partly funded by the NIHR School for Public Health Research. SVK acknowledges funding from a NHS Research Scotland Senior Clinical Fellowship (SCAF/15/02), the Medical Research Council (MC_UU_00022/2) and the Scottish Government Chief Scientist Office (SPHSU17). CS (NIHR201314) acknowledges support as a NIHR Senior Investigator.

Ethical approval

The study was approved by the Health Research Authority (Reference: 22/PR/0095, IRAS project ID: 307541) and by the University of Bristol Faculty of Health Sciences Research Ethics Committee (REC reference: 117269). NHS England is the data controller of the NHS England OpenSAFELY COVID-19 Service; TPP is the data processor; all study authors using OpenSAFELY have the approval of NHS England. This implementation of OpenSAFELY is hosted within the TPP environment which is accredited to the ISO 27001 information security standard and is NHS IG Toolkit compliant. Patient data has been pseudonymised for analysis and linkage using industry standard cryptographic hashing techniques; all pseudonymised datasets transmitted for linkage onto OpenSAFELY are encrypted; access to the NHS England OpenSAFELY COVID-19 service is via a virtual private network (VPN) connection; the researchers hold contracts with NHS England and only access the platform to initiate database queries and statistical models; all database activity is logged; only aggregate statistical outputs leave the platform environment following best practice for anonymisation of results such as statistical disclosure control for low cell counts. The service adheres to the obligations of the UK General Data Protection Regulation (UK GDPR) and the Data Protection Act 2018. The service previously operated under notices initially issued in February 2020 by the Secretary of State under Regulation 3(4) of the Health Service (Control of Patient Information) Regulations 2002 (COPI Regulations), which required organisations to process confidential patient information for COVID-19 purposes; this set aside the requirement for patient consent. As of 1 July 2023, the Secretary of State has requested that NHS England continue to operate the Service under the COVID-19 Directions 2020. In some cases of data sharing, the common law duty of confidence is met using, for example, patient consent or support from the Health Research Authority Confidentiality Advisory Group. Taken together, these provide the legal bases to link patient datasets using the service. GP practices, which provide access to the primary care data, are required to share relevant health information to support the public health response to the pandemic, and have been informed of how the service operates.

Provenance

Freely submitted; externally peer reviewed.

Data

All data used in this study were accessed via the OpenSAFELY platform. All analysis code and primary care codelists used to perform the analyses are publicly available online: https://github.com/opensafely/lockdown-and-vulnerable-groups

Acknowledgements

We are very grateful for all the support received from the TPP Technical Operations team throughout this work, and for generous assistance from the information governance and database teams at NHS England and the NHS Transformation Directorate. The authors would like to thank the patient and public contributors who contributed to interpretation of the results.

Competing interests

SVK was co-chair of the Scottish Government's Expert Reference Group on Ethnicity & COVID-19 and a member of the UK Scientific Advisory Group on Emergencies (SAGE) subgroup on ethnicity. AS served on the Scottish Government’s COVID-19 Advisory Group and is a member of its Standing Committee on Pandemic Preparedness.

  • Received January 23, 2025.
  • Accepted April 17, 2025.

This article is Open Access: CC BY license (https://creativecommons.org/licenses/by/4.0/)

References

  1. 1.↵
    1. Massie L
    (2023) How many people have had COVID-19? The challenge of reinfections. accessed. .. https://blog.ons.gov.uk/2023/02/15/how-many-people-have-had-covid-19-the-challenge-of-reinfections. 3 Sep 2025.
  2. 2.
    (2022) Coronavirus (COVID-19) infection survey technical article: cumulative incidence of the number of people who have tested positive for COVID-19, UK, Office for National Statistics. accessed. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/conditionsanddiseases/articles/coronaviruscovid19infectionsurveytechnicalarticlecumulativeincidenceofthenumberofpeoplewhohavetestedpositiveforcovid19uk/22april2022. 3 Sep 2025.
  3. 3.↵
    (2024) UKHSA data dashboard: COVID-19, UK Health Security Agency. accessed. https://ukhsa-dashboard.data.gov.uk/respiratory-viruses/covid-19. 3 Sep 2025.
  4. 4.↵
    1. Hippisley-Cox J,
    2. Khunti K,
    3. Sheikh A,
    4. et al.
    (2023) Risk prediction of covid-19 related death or hospital admission in adults testing positive for SARS-CoV-2 infection during the omicron wave in England (QCOVID4): cohort study. BMJ 381 doi:10.1136/bmj-2022-072976, pmid:37343968. e072976.
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    1. Shah SA,
    2. Brophy S,
    3. Kennedy J,
    4. et al.
    (2022) Impact of first UK COVID-19 lockdown on hospital admissions: interrupted time series study of 32 million people. EClinicalMedicine 49 doi:10.1016/j.eclinm.2022.101462, pmid:35611160. 101462.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Shah SA,
    2. Quint JK,
    3. Nwaru BI,
    4. Sheikh A
    (2021) Impact of COVID-19 national lockdown on asthma exacerbations: interrupted time-series analysis of English primary care data. Thorax 76 (9):860–866, doi:10.1136/thoraxjnl-2020-216512, pmid:33782080.
    OpenUrlAbstract/FREE Full Text
  7. 7.↵
    1. Mulholland RH,
    2. Wood R,
    3. Stagg HR,
    4. et al.
    (2020) Impact of COVID-19 on accident and emergency attendances and emergency and planned hospital admissions in Scotland: an interrupted time-series analysis. J R Soc Med 113 (11):444–453, doi:10.1177/0141076820962447, pmid:33012218.
    OpenUrlCrossRefPubMed
  8. 8.
    1. Shah SA,
    2. Mulholland RH,
    3. Wilkinson S,
    4. et al.
    (2022) Impact on emergency and elective hospital-based care in Scotland over the first 12 months of the pandemic: interrupted time-series analysis of national lockdowns. J R Soc Med 115 (11):429–438, doi:10.1177/01410768221095239, pmid:35502909.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Wyatt S,
    2. Mohammed MA,
    3. Fisher E,
    4. et al.
    (2021) Impact of the SARS-CoV-2 pandemic and associated lockdown measures on attendances at emergency departments in English hospitals: a retrospective database study. Lancet Reg Health Eur 2 doi:10.1016/j.lanepe.2021.100034, pmid:34173630. 100034.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Murphy M,
    2. Scott LJ,
    3. Salisbury C,
    4. et al.
    (2021) Implementation of remote consulting in UK primary care following the COVID-19 pandemic: a mixed-methods longitudinal study. Br J Gen Pract 71 (704):e166–e177, doi:10.3399/BJGP.2020.0948, pmid:33558332.
    OpenUrlAbstract/FREE Full Text
  11. 11.↵
    1. Foley KA,
    2. Maile EJ,
    3. Bottle A,
    4. et al.
    (2022) Impact of COVID-19 on primary care contacts with children and young people in England: longitudinal trends study 2015–2020. Br J Gen Pract 72 (720):e464–e471, doi:10.3399/BJGP.2021.0643, pmid:35667683.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. Mansfield KE,
    2. Mathur R,
    3. Tazare J,
    4. et al.
    (2021) Indirect acute effects of the COVID-19 pandemic on physical and mental health in the UK: a population-based study. Lancet Digit Health 3 (4):e217–e230, doi:10.1016/S2589-7500(21)00017-0, pmid:33612430.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Williams R,
    2. Jenkins DA,
    3. Ashcroft DM,
    4. et al.
    (2020) Diagnosis of physical and mental health conditions in primary care during the COVID-19 pandemic: a retrospective cohort study. Lancet Public Health 5 (10):e543–e550, doi:10.1016/S2468-2667(20)30201-2, pmid:32979305.
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Hobbs FDR,
    2. Bankhead C,
    3. Mukhtar T,
    4. et al.
    (2016) Clinical workload in UK primary care: a retrospective analysis of 100 million consultations in England, 2007–14. Lancet 387 (10035):2323–2330, doi:10.1016/S0140-6736(16)00620-6, pmid:27059888.
    OpenUrlCrossRefPubMed
  15. 15.
    1. Kontopantelis E,
    2. Olier I,
    3. Planner C,
    4. et al.
    (2015) Primary care consultation rates among people with and without severe mental illness: a UK cohort study using the Clinical Practice Research Datalink. BMJ Open 5 (12), doi:10.1136/bmjopen-2015-008650, pmid:26674496. e008650.
    OpenUrlAbstract/FREE Full Text
  16. 16.↵
    1. Lay-Flurrie S,
    2. Mathieu E,
    3. Bankhead C,
    4. et al.
    (2019) Patient consultation rate and clinical and NHS outcomes: a cross-sectional analysis of English primary care data from 2.7 million patients in 238 practices. BMC Health Serv Res 19 (1), doi:10.1186/s12913-019-4036-y, pmid:30954074. 219.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Mertens JR,
    2. Weisner C,
    3. Ray GT,
    4. et al.
    (2005) Hazardous drinkers and drug users in HMO primary care: prevalence, medical conditions, and costs. Alcohol Clin Exp Res 29 (6):989–998, doi:10.1097/01.ALC.0000167958.68586.3D.
    OpenUrlCrossRefPubMed
  18. 18.↵
    1. Jané-Llopis E,
    2. Matytsina I
    (2006) Mental health and alcohol, drugs and tobacco: a review of the comorbidity between mental disorders and the use of alcohol, tobacco and illicit drugs. Drug Alcohol Rev 25 (6):515–536, doi:10.1080/09595230600944461, pmid:17132571.
    OpenUrlCrossRefPubMed
  19. 19.↵
    1. Andrews C,
    2. Schultze A,
    3. Curtis H,
    4. et al.
    (2022) OpenSAFELY: representativeness of electronic health record platform OpenSAFELY-TPP data compared to the population of England. Wellcome Open Res 7 doi:10.12688/wellcomeopenres.18010.1, pmid:35966958. 191.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. GOV.UK
    (2020) Prime Minister’s statement on coronavirus (COVID-19): 10 May 2020, accessed. https://www.gov.uk/government/speeches/pm-address-to-the-nation-on-coronavirus-10-may-2020. 3 Sep 2025.
  21. 21.↵
    1. GOV.UK
    (2020) Prime Minister’s statement on coronavirus (COVID-19): 23 March 2020, accessed. https://www.gov.uk/government/speeches/pm-address-to-the-nation-on-coronavirus-23-march-2020. 3 Sep 2025.
  22. 22.↵
    (2020) Gov.UK. Prime Minister’s statement on coronavirus (COVID-19): 5 November 2020. accessed. https://www.gov.uk/government/speeches/prime-ministers-statement-on-coronavirus-covid-19-5-november-2020. 30 Oct 2025.
  23. 23.↵
  24. 24.↵
    (2025) Secure analytics platform for NHS electronic health records. OpenSAFELY. accessed. https://www.opensafely.org. 3 Sep 2025.
  25. 25.↵
    1. Walter SR
    (2023) OpenSAFELY: Lockdown and vulnerable groups. accessed. https://github.com/opensafely/lockdown-and-vulnerable-groups. 3 Sep 2025.
  26. 26.↵
    1. Baird B,
    2. Charles A,
    3. Honeyman M,
    4. et al.
    Understanding pressures in general practice, accessed. https://www.kingsfund.org.uk/insight-and-analysis/reports/understanding-pressures-general-practice. 28 Oct 2025.
  27. 27.↵
    1. Craig P,
    2. Katikireddi SV,
    3. Leyland A,
    4. Popham F
    (2017) Natural experiments: an overview of methods, approaches, and contributions to public health intervention research. Annu Rev Public Health 38 39–56, doi:10.1146/annurev-publhealth-031816-044327, pmid:28125392.
    OpenUrlCrossRefPubMed
  28. 28.↵
    1. Salisbury C,
    2. Lay-Flurrie S,
    3. Bankhead CR,
    4. et al.
    (2021) Measuring the complexity of general practice consultations: a Delphi and cross-sectional study in English primary care. Br J Gen Pract 71 (707):e423–e431, doi:10.3399/BJGP.2020.0486, pmid:33824162.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Maddock J,
    2. Parsons S,
    3. Di Gessa G,
    4. et al.
    (2022) Inequalities in healthcare disruptions during the COVID-19 pandemic: evidence from 12 UK population-based longitudinal studies. BMJ Open 12 (10), doi:10.1136/bmjopen-2022-064981, pmid:36229151. e064981.
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    1. Kesten JM,
    2. Holland A,
    3. Linton M-J,
    4. et al.
    (2021) Living under coronavirus and injecting drugs in Bristol (LUCID-b): a qualitative study of experiences of COVID-19 among people who inject drugs. Int J Drug Policy 98 doi:10.1016/j.drugpo.2021.103391, pmid:34343945. 103391.
    OpenUrlCrossRefPubMed
  31. 31.↵
    1. Dixon S,
    2. De Simoni A,
    3. Szilassy E,
    4. et al.
    (2023) General practice wide adaptations to support patients affected by DVA during the COVID-19 pandemic: a rapid qualitative study. BMC Prim Care 24 (1), doi:10.1186/s12875-023-02008-6, pmid:36959527. 78.
    OpenUrlCrossRefPubMed
  32. 32.↵
    1. Panovska-Griffiths J,
    2. Szilassy E,
    3. Johnson M,
    4. et al.
    (2022) Impact of the first national COVID-19 lockdown on referral of women experiencing domestic violence and abuse in England and Wales. BMC Public Health 22 (1), doi:10.1186/s12889-022-12825-6, pmid:35291956. 504.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Turner A,
    2. Scott A,
    3. Horwood J,
    4. et al.
    (2021) Maintaining face-to-face contact during the COVID-19 pandemic: a longitudinal qualitative investigation in UK primary care. BJGP Open 5 (5), doi:10.3399/BJGPO.2021.0036, pmid:34257067. BJGPO.2021.0036.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

BJGP Open
Vol. 9, Issue 4
December 2025
Download PDF
Download PowerPoint
Email Article
Citation Tools
The impact of COVID-19 lockdowns on primary care contact among vulnerable populations in England: a controlled interrupted time-series study
Scott R Walter, Chris Salisbury, Lauren J Scott, Frank de Vocht, John Macleod, Yoav Ben-Shlomo, Helen J Curtis, Aziz Sheikh, Srinivasa V Katikireddi, Amir Mehrkar, Sebastian Bacon, George Hickman, Ben Goldacre, The OpenSAFELY Collective, Maria Theresa Redaniel
BJGP Open 2025; 9 (4): BJGPO.2025.0017. DOI: 10.3399/BJGPO.2025.0017
del.icio.us logo Facebook logo Mendeley logo Bluesky logo

Jump to section

Keywords

Intended for Healthcare Professionals
 
 

British Journal of General Practice