key: cord-0687208-k0qtjmwn
authors: Holland, David; Heald, Adrian H; Stedman, Mike; Hanna, Fahmy; Wu, Pensee; Duff, Christopher; Green, Lewis; Robinson, Sarah; Halsall, Ian; Gaskell, Neil; Pemberton, John; Bloor, Christine; Fryer, Anthony A
title: Assessment of the effect of the COVID-19 pandemic on UK HbA1c testing: implications for diabetes management and diagnosis
date: 2021-10-13
journal: J Clin Pathol
DOI: 10.1136/jclinpath-2021-207776
sha: dac7cfadca492e47f1bdf3bde22a6ac5a7119511
doc_id: 687208
cord_uid: k0qtjmwn

AIMS: The COVID-19 pandemic, and the focus on mitigating its effects, has disrupted diabetes healthcare services worldwide. We aimed to quantify the effect of the pandemic on diabetes diagnosis/management, using glycated haemoglobin (HbA1c) as surrogate, across six UK centres. METHODS: Using routinely collected laboratory data, we estimated the number of missed HbA1c tests for ‘diagnostic’/‘screening’/‘management’ purposes during the COVID-19 impact period (CIP; 23 March 2020 to 30 September 2020). We examined potential impact in terms of: (1) diabetes control in people with diabetes and (2) detection of new diabetes and prediabetes cases. RESULTS: In April 2020, HbA1c test numbers fell by ~80%. Overall, across six centres, 369 871 tests were missed during the 6.28 months of the CIP, equivalent to >6.6 million tests nationwide. We identified 79 131 missed ‘monitoring’ tests in people with diabetes. In those 28 564 people with suboptimal control, this delayed monitoring was associated with a 2–3 mmol/mol HbA1c increase. Overall, 149 455 ‘screening’ and 141 285 ‘diagnostic’ tests were also missed. Across the UK, our findings equate to 1.41 million missed/delayed diabetes monitoring tests (including 0.51 million in people with suboptimal control), 2.67 million screening tests in high-risk groups (0.48 million within the prediabetes range) and 2.52 million tests for diagnosis (0.21 million in the pre-diabetes range; ~70 000 in the diabetes range). CONCLUSIONS: Our findings illustrate the widespread collateral impact of implementing measures to mitigate the impact of COVID-19 in people with, or being investigated for, diabetes. For people with diabetes, missed tests will result in further deterioration in diabetes control, especially in those whose HbA1c levels are already high.

The COVID-19 pandemic, and the focus on mitigating its effects, has disrupted healthcare systems across the world, 1 2 including those for the diagnosis and management of people with diabetes. 3 4 Routine blood testing, a mainstay of diabetes diagnosis and management, became challenging, not least because of the potential risk it posed to facilitating transmission of the virus and associated public concerns regarding attending for tests. 5 Hence, some healthcare services adopted a pragmatic approach to long-term monitoring, prioritising those at highest risk. 5 Reflecting this, the demand for blood testing, including for diabetes diagnosis and monitoring, in some countries dropped during lockdowns. 6 Carr et al, 7 for example, showed that testing for the key diabetes marker, glycated haemoglobin (HbA1c), for primary care patients with diabetes reduced in the months following the first UK lockdown. However, the scale and impact of these changes across primary and secondary care diabetes services for both monitoring and diagnosis has not been investigated.

We previously showed a link between HbA1c testing frequency and diabetes control expressed as change in HbA1c 8 and likelihood of achieving target, 9 highlighting the importance of regular monitoring in maintaining diabetes control. 10 11 This is particularly important given that those with suboptimally controlled diabetes have poorer outcomes in the event of SARS-CoV-2 infection. [12] [13] [14] [15] [16] HbA1c is also frequently used in diagnosis in symptomatic patients 17 and as a screening tool in high-risk groups (~5 million people in England alone 18 ), where annual HbA1c testing is generally recommended. [18] [19] [20] [21] HbA1c forms part of the English Health Check programme for people aged 40-74. 22 As in those with diabetes, there is evidence that people with prediabetes, obesity and other causes of dysglycaemia are also at risk of poorer outcomes following SARS-CoV-2 infection. 12 13 23 We assessed the impact of the pandemic on diabetes diagnosis and management, using HbA1c as a surrogate, across regions covered by six UK laboratories. We estimated the number of missed HbA1c diabetes monitoring and diagnosis tests over a 3-year period and investigated the potential impact of these missed tests in terms of (1) effect on diabetes control in people with diabetes and (2) detection of new diabetes cases. 24 .

Data on the following standardised set of parameters were extracted: unique patient ID (anonymised), test result, date of request, age, sex, source of request (general practitioner or not).

Characteristics of the cohorts are shown in table 1. Data on the areas covered by the laboratories were obtained from National Health Service Digital, 25 based on the GP practices served by each laboratory. The six sites were selected to cover a range of population demographics.

HbA1c was measured using standard laboratory procedures. For all laboratories, the assay was within the scope of the laboratory's ISO 15189 accreditation, as overseen by the United Kingdom Accreditation Service. 26 Throughout the study period, the assay demonstrated acceptable performance on routine Internal Quality Control and External Quality Assurance parameter across all six sites.

All laboratories used a reference range of <48 mmol/mol as a diagnostic cut-off, as recommended by the National Institute for Health and Care Excellence (NICE) 10 and WHO. 17 The use of cut-offs for 'prediabetes' (42-47 mmol/mol) used across the centres was based on NICE guidance. 10 Comments provided on reports varied between centres, from minimal (reference range only), to providing some information on HbA1c targets for different type 1 and type 2 diabetes mellitus groups based on the result. Similarly, local guidance varied across sites but generally referred to NICE guidance. 10

We examined test volume changes over the course of the pandemic, compared with the prior 2 years. This enabled calculation of the number of missed tests during the COVID-19 impact period (CIP; 23 March 2020 to 30 September 2020). We categorised tests into three groups: 'monitoring' (those in people with diabetes), 'screening' (those in high-risk groups) and 'diagnostic' (testing to see if a person has diabetes) using the algorithm shown in figure 1 . This categorisation was used as each of these groups require different intervals for follow-up testing (see below), and this information was critical in the calculations of potential missed tests.

We also stratified tests by HbA1c level into:<30 mmol/ mol, 30-41 mmol/mol, 42-47 mmol/mol, 48-53 mmol/ mol, 54-58 mmol/mol, 59-75 mmol/mol, 76-86 mmol/mol and>86 mmol/mol.

'Monitoring' tests were defined as those in people who had >1 test during the study period, and had at least one HbA1c value of ≥48 mmol/mol (figure 1). Most of these individuals are likely to be those with existing diabetes, even if some of their HbA1c levels dropped below 48 mmol/mol. To calculate whether a 'monitoring' test would have been due during the CIP, we used an expected interval of up to 3 months for those with suboptimal control (≥59 mmol/mol for the test immediately prior to the CIP), and up to 6 months for those with good control (<59 mmol/mol). 8 If a test was due during the CIP, we were able to determine whether this test was actually performed or not.

We assumed that most individuals with multiple HbA1c tests during the study period, but in whom all results were <48 mmol/ mol, were most likely to be those people being regularly reviewed as part of screening in high-risk groups, such as those with prediabetes, those in whom health checks were performed, and women with a history of gestational diabetes or polycystic ovaries. As guidance for these individuals generally recommends annual screening, 18-21 we used a 12-month interval to assess if a test was due during the CIP.

'Diagnostic' tests were defined as either: (a) those with only one test during the study period or (b) those whose first test was after the first 18 months (figure 1). Within this latter group, we assumed that those who had no tests during the first 18 months were unlikely to be people in either the monitoring or screening groups, at least initially. In these cases, we assumed that the first test was a 'screening' test and all subsequent tests either 'screening' (if all subsequent tests were <48 mmol/mol) or 'monitoring' (figure 1). Missed 'diagnostic' tests were based on the proportion of all 'diagnostic' tests performed (as a percentage of total tests) during the 12-month period immediately prior to the CIP, using the following formula:

where: Dx=diagnostic tests, non-Dx=management+screening tests, Total=diagnostic+management+screening tests, P=Performed tests, M=missed tests.

The rationale for using this approach was, of necessity, different than those used to calculate missed screening and diabetes monitoring tests as they do not have a defined follow-up testing interval. Hence, we used reference to the previous year's data and the data on the monitoring and screening tests to estimate the number of missed diagnostic tests.

To estimate the impact of missed 'monitoring' tests on change in HbA1c level, we used our previous approach, which assessed the relationship between testing interval and change in HbA1c, 8 for each of the six sites (online supplemental table 1). We then calculated, on a patient-by-patient basis for each HbA1c category, (a) the expected change in HbA1c if the test was done at the recommended interval and (b) the expected change in HbA1c if missed tests were done after the end of the study period (ie, 1 October 2020). We then subtracted (b) from (a) to give the impact of the delay on the change in HbA1c (online supplemental table 2). This represents the minimum impact as it is not likely that all missed tests would be done on 1 October 2020. Table 1 shows that the overall mean HbA1c (43.9-45.2 mmol/ mol) and number of tests per patient (2.09-2.30) were similar across the sites. These were also similar when 'monitoring' tests alone were examined (60.2-61.2 mmol/mol and 4.10-4.78, respectively). Site-specific age and gender distributions of general practices covered were similar. Median Index of Multiple Deprivation rank associated with the practices illustrated that the sites encompassed areas of relatively high (eg, STHK) and low deprivation (eg, CUH).

We observed an overall 8.3% increase in HbA1c testing between the first (October 17 to September 18) and second (October 18 to September 19) 12-month periods (range: +2.9% to +12.6%). This is consistent with the year-on-year trend we observed for HbA1c across UK laboratories (mean annual change from 2017 to 2020: 8.1%), through data collected routinely by The Benchmarking Partnership and available to one of the coauthors (DH). The Benchmarking Partnership 27 collects data on a range of laboratory parameters, including test volumes, as part of a UK-wide pathology benchmarking service. An 8% increase would predict 122 371 tests/month (1 468 456 tests) during October 19 to September 20. However, the actual tests performed during this period averaged 87 806 tests/month (−28.4%); most of the reduction was observed during the 6.28-month CIP. Figure 2 shows that, across all sites, there was a sharp decline in HbA1c requests immediately after the initial lockdown (from 23 March 2020), dropping by 81.3%-87.8% in April compared with the 2019 mean. This was followed by a gradual rise, reaching 70.6%-96.3% of expected volumes by September 20. HbA1c requests from general practice fell by 84.9%-90.9%, while those from other sources (mostly acute hospitals) fell by 51.8%-75.6%.

Overall, there were 32 280 'monitoring' tests/month performed prior to the CIP (table 2) . However, the actual number performed during the CIP was 19 138/month, a reduction of 40.7%. When we calculated the missed tests due during the CIP, this equated to 79 131 tests (table 2); 39.7% of those expected (similar to the above 40.7% reduction in testing compared with the pre-CIP period).

Of concern was the observation that 36.1% of the missed tests (4548 tests/month) were in those with suboptimal control (≥59 mmol/mol); 7.3% (920 tests/month) in those with a previous HbA1c of >86 mmol/mol.

Extrapolating this to the UK population we estimated that, during the CIP, 1.41 m 'monitoring' tests would have been missed (>225 000 tests/month). Of these, 0.51 m (81 228/month) would be in those with suboptimal control.

Overall, there were 49 993 'screening' tests/month performed during the pre-CIP period (table 2) . However, the number performed during the CIP was 32 212 /month; a 35.6% reduction. The tests due during the CIP equated to 149 455 missed screening tests (table 2) , representing 42.5% of the expected tests. This was higher than the 35.6% reduction in 'screening' testing compared with the pre-CIP period, perhaps suggesting that a proportion of expected screening tests are not generally carried out (even prior to the pandemic).

As expected, most 'screening' tests (78.4%) performed in the pre-CIP period were within the reference range (<42 mmol/ mol). This was similar for those performed during the CIP (79.6%). On average, 4264 /month missed 'screening' tests were in the prediabetes range (table 2) .

Extrapolating these data to the UK, during the CIP, 2.67 m 'screening' tests would have been missed (>0.42 m/month), including 0.48 m tests (>76 000/month) within the prediabetes range.

During the CIP, an estimated 22 498 test/month were missed, a 60.1% reduction compared with the pre-CIP period (table 2) . When missed test were expressed as a percentage of total expected diagnostic tests, this suggested that missed tests represented 64.8% of all 'diagnostic' tests; similar to the 60.1% compared with the pre-CIP period.

Of the 22 498 missed 'diagnostic' tests/month, 8.4% and 2.6% were estimated to be within the prediabetes and diabetes ranges, respectively. An estimated 613 missed tests were in people with an HbA1c>86 mmol/mol (table 2) . Across the UK, these equate to 2.52 m missed 'diagnostic' tests during the CIP (0.40 m/month), including ~212 000 in the prediabetes range and 68 800 delayed new diabetes diagnoses (~33 800 and~11 000 /month, respectively).

We then examined the impact of missed 'monitoring' tests in terms of change in HbA1c caused by the delay in testing. Figure 3A shows the mean testing interval up until the end of the study period, compared with recommended intervals. This showed a mean delay of 2.2-3.4 months in those with good control and ~3.4 months in those with suboptimal control. Figure 3B shows the difference between the expected HbA1c change based on testing according to recommended interval and that expected if all missed tests were done on 1 October 2020. Our data identified four HbA1c groups; (1)<48 mmol/mol: minimal impact of testing delay (<0.3 mmol/mol), (2) HbA1c 48-58 mmol/mol: small impact (~1 mmol/mol), (3) HbA1c 59-86 mmol/mol: moderate impact (~2 mmol/mol) and (4) HbA1c>86 mmol/mol: large impact (>3 mmol/mol).

UK wide, this suggests that ~565 000 people (40.0% of patients) will see an additional HbA1c increase of ~1 mmol/mol, ~407 000 (28.8%) of ~2 mmol/mol, and ~103 000 (7.3%) of >3 mmol/mol. In contrast, ~338 000 (24.0% of patients) may warrant less frequent testing (9-12 months).

Using routine clinical laboratory data, we estimate that ~6.6 million UK HbA1c tests were missed during the 6 month CIP: 1.4 m 'monitoring' tests (0.5 million with suboptimal control), 2.7 million 'screening' tests (0.5 million with prediabetes) and 2.5 million 'diagnostic' tests (0.2 million with prediabetes,~69 000 within the diabetes range). In those with diabetes, we calculated that 76% of patients (~1.1 million) would see their HbA1c rise by >1 mmol/mol more than expected; over 100 000 of these with the worst control (>86 mmol/mol) would see an additional rise of >3 mmol/mol due to missed tests. These are broadly in keeping with our preliminary analysis. 28

Across all sites, the volume of HbA1c test dropped markedly in April 2020 and rose slowly thereafter. This is consistent with Carr et al who showed HbA1c testing in UK people with type 2 diabetes reduced by 77%. 7 Similarly, using South African laboratory data, Kruger et al 6 showed that HbA1c testing overall reduced by ~64% during March to June 2020 compared with the same period the previous year, with the most marked reduction (81%) in April.

In a UK study of the indirect effects of the COVID-19 pandemic on physical and mental health using the UK Clinical Research Practice Datalink, Mansfield et al 29 showed that 'Primary care contacts for almost all conditions dropped considerably after the introduction of population wide restrictions.' A WHO survey conducted in June 2020 showed that treatment for diabetes and associated complications were disrupted in 49% of countries. 30 Similar results have been observed in an international survey of healthcare professionals. 31

We showed that HbA1c tests in people with diabetes fell by an average of 40%/month during the CIP. While the majority of missed monitoring tests were in those with lower HbA1c values, 36.1% were in those with suboptimally controlled diabetes, including 7.3% with previous values of >86 mmol/mol. This is particularly concerning as people with diabetes who have higher HbA1c values have increased COVID-19-related mortality 16 and disease severity. 15 Holman et al 16 , showed that mortality, and diabetes-related complications, rose with increasing HbA1c. Both HbA1c and blood glucose are risk factors for higher COVID-19-associated mortality. 12 23 Furthermore, studies show that SARS-CoV-2 infection exacerbates the underlying Table 2 Number of tests performed/missed, by HbA1c level, compared with those performed in the 12 months prior to the COVID-19 impact period (CIP); (A) 'monitoring', (B) 'screening' and (C) 'diagnostic' 

pathophysiology of hyperglycaemic in people with diabetes, leading to further rises in HbA1c. 13 32 In a meta-analysis of 25 studies, Pan et al 33 demonstrated that HbA1c is a risk factor for poorer outcome in patients with acute coronary syndrome, illustrating the importance of maintaining stricter glycaemic control in people with diabetes, both outside and within the context of SARS-CoV-2 infection. Hence, lack of engagement with this group of people not only raises their pre-COVID-19-associated risk, but increases the probability of poor outcomes in the event of SARS-CoV-2 infection. Our data demonstrate that HbA1c testing in over 100 000 people in the UK with HbA1c values of >86 mmol/mol was missed or delayed. In this group, we estimated that this would result, in an average increase of >3 mmol/mol. Smaller delayassociated increases in HbA1c were observed in other groups of patients with HbA1c values of 48-86 mmol/mol who missed tests (estimated at around one third of all people with diabetes). 34 While these increases appear modest, at a population level this represents a significant retrograde step in the management of the most at-risk patients.

We assumed that the change in HbA1c due to the interval between testing would follow a similar pattern to that observed prior to the CIP, and using our previously described methodology. 8 35 However, this may be an underestimate of the impact of the CIP for two reasons. First, the calculated impact of the delay assumes all missed tests are then performed on the first day following the CIP (ie, 1/10/20). This is highly unlikely to be the case, so the length of the delay in testing, and hence the impact on the HbA1c level, will therefore be greater than our calculations. Second, both discussions with local diabetes patient groups (Diabetes UK local branch) and indications from the literature [36] [37] [38] suggest that the significant restrictions imposed throughout the CIP would adversely affect lifestyle factors important for diabetes control.

We estimated that, on average, 23 799 'screening' tests/month (42.5%) were missed during the CIP. While this group may appear less important than the 'monitoring' group, 'screening' tests account for over 40% of missed tests. This equates to 0.4 m tests/month of the 6-month CIP, ~18% of which were in the prediabetes range.

UK NICE guidance recommend use of a risk assessment tool and appropriate tailored lifestyle advice for those with an HbA1c of 42-47 mmol/mol. 18 It suggests that such individuals should be offered a referral to a 'local, evidence-based, qualityassured intensive lifestyle-change programme'. However, in the absence of the test, such advice/referral may have been delayed and hence some may prematurely convert to full diabetes. It is estimated that 5%-10% of patients with prediabetes convert to diabetes each year. 39 40 In this context, during the CIP, we expect that 2.6%-5.2% of the expected 478 196 UK patients with prediabetes in the 'screening' group to convert to diabetes; 12 400-24 800 new diabetes patients in whom treatment would have been delayed. This is particularly important as an estimated 1 in 3 of newly diagnosed patients have complications. 41

The 'diagnostic' group comprised 141 285 missed tests of whom 11 880 would have been in the prediabetes range. From a national perspective, this equates to a further 212 143 patients with prediabetes of whom 5500-11 000 would be expected to convert to diabetes during the CIP. These people, in addition to the expected 68 521 UK-wide missed tests in the diabetes range, would also have delayed diagnoses and treatment.

Laboratory data generally does not include the reason for HbA1c testing nor the type/duration of diabetes. Hence, it was not possible to definitively differentiate monitoring from diagnostic/screening tests. However, the distribution of tests appears consistent with expectations and the main message of the study remains valid. For example, National Diabetes Audit data indicates that around 34% of patients with type 2 diabetes have an HbA1c≥59 mmol/mol. 34 This is a comparable with our estimates of around 40% for both type 1 and 2 diabetes. Similarly, our estimates for pre-diabetes in the screening and diagnostic tests (~30%) is in keeping with overall estimates for prediabetes prevalence (~35%) in the US 39 and UK. 40 It is true that HbA1c alone is not the only marker of glucose dysregulation; blood glucose measurement is also important in this regard. Hence, our figures may underestimate the magnitude of the impact of the pandemic on diabetes detection and monitoring. Certainly, derangement in blood glucose has similar negative implications; higher blood glucose levels result in poorer COVID-19 outcomes, irrespective of whether they had Original research diabetes or not, 12 and severe COVID-19 was associated with higher blood glucose levels. 32 We recognise that targets for HbA1c monitoring for primary care, as reflected in the Quality and Outcome Framework (QOF) indicators listed within the contract for general practitioners, 42 refer to longer intervals for measuring HbA1c in people with diabetes (12-15 months) than those defined in NICE clinical guidelines. 10 However, QOF, being linked to primary care funding, has a somewhat different remit than NICE clinical guidelines, and we believe that these should be considered a minimum standard rather than a clinically ideal one.

We also acknowledge that, in the data extrapolation to the UK population, there were some differences in the degree of lockdown restrictions between the devolved nations of the UK. However, these differences were more limited during the initial lockdown in March 2020 and hence we do not believe that this would have significant impact on the overall findings of the study.

This study highlights the impact of the pandemic on day-to-day management of people with, and at risk of, diabetes. This will have consequences for their future health that need to be taken account of in the coming years.

► We estimate that ~6.6 million UK glycated haemoglobin (HbA1c) tests were missed during April to September 2020: 1.4million tests in people with diabetes (0.5 million with suboptimal control) and 5.2 million 'screening'/'diagnostic' tests (0.7 million with prediabetes, ~70 000 within the diabetes range). ► In those with diabetes, >100 000 with HbA1c>86 mmol/mol would see a rise of >3 mmol/mol due to missed tests. ► This highlights the potential impact of the COVID-19 pandemic on day-to-day management of those people with, and at risk of, diabetes. This has major clinical outcome implications.

Author affiliations 1 The Benchmarking Partnership, Alsager, UK

Lessons on the COVID-19 pandemic, for and by primary care professionals worldwide

Redesigning primary care to address the COVID-19 pandemic in the midst of the pandemic

Diagnosis of physical and mental health conditions in primary care during the COVID-19 pandemic: a retrospective cohort study

COVID-19: generate and apply local modelled transmission and morbidity effects to provide an estimate of the variation in overall relative healthcare resource impact at general practice granularity

Pandemic threatens primary care for long term conditions

The impact of COVID-19 on routine patient care from a laboratory perspective

Impact of COVID-19 on the diagnoses, HbA1c monitoring and mortality in people with type 2 diabetes: a UK-wide cohort study involving 13 million people in primary care

Reduced testing frequency for glycated hemoglobin, HbA1c, is associated with deteriorating diabetes control

The frequency of testing for glycated haemoglobin, HbA1c, is linked to the probability of achieving target levels in patients with suboptimally controlled diabetes mellitus

National Institute for Health and Care Excellence (NICE)

Glycemic Targets: Standards of Medical Care in Diabetes-2021

Elevation of blood glucose level predicts worse outcomes in hospitalized patients with COVID-19: a retrospective cohort study

Covid-19 and diabetes: a complex bidirectional relationship

Glycemic characteristics and clinical outcomes of COVID-19 patients hospitalized in the United States

COVID-19 severity is Tripled in the diabetes community: a prospective analysis of the pandemic's impact in type 1 and type 2 diabetes

Risk factors for COVID-19-related mortality in people with type 1 and type 2 diabetes in England: a population-based cohort study

18 National Institute for Health and Care Excellence (NICE). Public Health guidance [PH38]. Type 2 diabetes: prevention in people at high risk

Clinical knowledge summary. Management of polycystic ovary syndrome in adults

Predicting dysglycaemia in patients under investigation for acute coronary syndrome

National Institute for Health and Care Excellence (NICE. Nhs health check programme

What is the role of admission HbA1c in managing COVID-19 patients?

Worldometers mid-year UK population

Patients registered at a GP practice

United Kingdom Accreditation Service

The Benchmarking Partnership. National pathology benchmarking programme

Impact of the UK COVID-19 pandemic on HbA1c testing and its implications for diabetes diagnosis and management

Indirect acute effects of the COVID-19 pandemic on physical and mental health in the UK: a population-based study

WHO survey World Health Organization. COVID-19 significantly impacts health research for noncommunicable diseases

Impact of COVID-19 on routine care for chronic diseases: a global survey of views from healthcare professionals

The impact of COVID-19 on blood glucose: a systematic review and meta-analysis

Prognostic value of HbA1c for in-hospital and short-term mortality in patients with acute coronary syndrome: a systematic review and metaanalysis

National diabetes audit -report 1 care processes and treatment targets 2018-19, full report

Inappropriate requesting of glycated hemoglobin (Hb A1c) is widespread: assessment of prevalence, impact of national guidance, and practice-to-practice variability

Effect of coronavirus disease 2019 pandemic on the lifestyle and glycemic control in patients with type 2 diabetes: a cross-section and retrospective cohort study

Impact of lockdown COVID-19 on metabolic control in type 2 diabetes mellitus and healthy people

Behavioral changes in patients with diabetes during the COVID-19 pandemic

Prediabetes: a high-risk state for diabetes development

Prevalence of prediabetes in England from 2003 to 2011: population-based, cross-sectional study

general medical services (GMS) contract quality and outcomes framework (QOF)

This article is made freely available for use in accordance with BMJ's website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.

Adrian H Heald http:// orcid. org/ 0000-0002-9537-4050 Mike Stedman http:// orcid. org/ 0000-0002-0491-7823 Pensee Wu http:// orcid. org/ 0000-0003-0011-5636 Christopher Duff http:// orcid. org/ 0000-0002-3753-0043 Lewis Green http:// orcid. org/ 0000-0001-5792-5408 Anthony A Fryer http:// orcid. org/ 0000-0001-8678-0404

Competing interests None declared.

Ethics approval According to the decision tool provided by the UK Health Research Authority (http://www. hra-decisiontools. org. uk/ research/), this study was not considered to be research and did not require NHS Research Ethics Committee review. It therefore did not require ethical committee approval or informed consent.Provenance and peer review Not commissioned; externally peer reviewed.Data availability statement Data are available upon reasonable request. Data are available on reasonable request from the authors.Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.