Type 1 and Type 2 Diabetes and COVID-19 Related Mortality in England: A Whole Population Study

This version posted 19th May 2020.
The copyright holder of this pre-print (which has not been certified by peer
review) is NHS England.
Type 1 and Type 2 diabetes and COVID-19 related mortality in England: a whole
population study
Emma Barron 1 MSc, Chirag Bakhai2 MBA, Partha Kar2,3 MD, Andy Weaver2 MSc, Dominique Bradley2 PhD,
Hassan Ismail2 BSc, Peter Knighton4 MPhys, Naomi Holman 2,4,5 PhD, Kamlesh Khunti6 MD, Naveed Sattar5
MD, Nick Wareham7 PhD, Bob Young 8 MD, Jonathan Valabhji2,9,10 MD
1. Public Health England, Station Rise, York, YO1 6GA, UK
2. NHS England & Improvement, Skipton House, 80 London Road, London, SE1 6LH, UK
3. Portsmouth Hospitals NHS Trust, Southwick Hill Road, Portsmouth, Hampshire, PO6 3LY, UK
4. NHS Digital, 1 Trevelyan Square, Board Lane, Leeds, LS1 6AE, UK
5. Institute of Cardiovascular Medical Sciences, University of Glasgow, 126 University Place, Glasgow, G12
8AT, UK
6. Diabetes Research Centre, University of Leicester, Leicester General Hospital, Gwendolen Road, Leicester,
LE5 4PW, UK
7. MRC Epidemiology Unit, Institute of Metabolic Science, University of Cambridge, Cambridge Biomedical
Campus, Cambridge CB2 0QQ
8. Diabetes UK, Wells Lawrence House, 126 Back Church Lane, London, E1 1FH, UK
9. Imperial College Healthcare NHS Trust, The Bays, S Wharf Rd, Paddington, London W2 1NY
10. Imperial College London, South Kensington, London, SW7 2BU , UK
Corresponding author:
Jonathan Valabhji
NHS England & Improvement, Skipton House, London, UK
Tel: 0113 8251692
Email: [email protected]
Word count: Abstract 250; Full text 3,681
Currently undergoing peer review for journal publication

This version posted 19th May 2020. The copyright holder of this pre-print (which has not been certified by peer
Abstract
Background
Although diabetes has been associated with COVID-19 mortality, the absolute and relative risks for Type 1 and
Type 2 diabetes are unknown.
Methods
A population cohort study assessing risks of in-hospital death with COVID-19 between 1st March and 11th May
2020, including individuals registered with a General Practice in England and alive on February 19 th 2020.
Multivariate logistic regression examined diabetes status, by type, and associations with in-hospital death,
adjusting for demographic factors and comorbidities.
Findings
Of the 61,414,470 individuals registered, 263,830 (0∙4%) had a recorded diagnosis of Type 1 and 2,864,670
(4∙7%) of Type 2 diabetes. There were 23,804 COVID-19 related deaths. One third occurred in people with
diabetes: 7,466 (31∙4%) with Type 2 and 365 (1∙5%) with Type 1 diabetes. Crude mortality rates per 100,000
persons over the 72 days for the overall population and for those with Type 1 and Type 2 diabetes were 38∙8
(38∙3-39∙3), 138∙3 (124∙5-153∙3), and 260∙6 (254∙7-266∙6) respectively. Adjusted for age, sex, deprivation,
ethnicity and geographical region, people with Type 1 and Type 2 diabetes had 3∙50 (3∙15-3∙89) and 2∙03 (1∙97-
2∙09) times the odds respectively of dying in hospital with COVID-19 compared to those without diabetes,
attenuated to 2∙86 and 1∙81 respectively when also adjusted for previous hospital admissions with coronary heart
disease, cerebrovascular disease or heart failure.
Interpretation
This nationwide analysis in England demonstrates that all types of diabetes are independently associated with a
significant increased risk of in-hospital death with COVID-19.
Funding
NHS England & Improvement and Public Health England.

Research in context
Evidence before this study
From March 2020, we performed weekly searches of PubMed and MedRxiv using the terms COVID-19, SARS-
CoV-2, coronavirus, SARS virus and diabetes. Studies from China, Italy, the USA and the UK have suggested
that people with diabetes have higher risks of more severe outcomes with COVID-19, including death. One
population-based UK study reported a higher risk of COVID-19 related death in those with diabetes after
adjustment for demographic factors and other comorbidities. However, none of these studies have assessed
differences in risk by type of diabetes.
Added value of this study
This is the largest COVID-19 related study, covering almost the entire population of England, and is the first
study to investigate the relative and absolute risks of death in hospital with COVID-19 by type of diabetes,
adjusting for key confounders. It demonstrates that one third of all deaths in-hospital with COVID-19 occur in
people with diabetes. Adjusted for age, sex, deprivation, ethnicity and geographical region, people with Type 1
and Type 2 diabetes had 3.50 and 2.03 times the odds respectively of dying in hospital with COVID-19
compared to those without diabetes. These relative odds were attenuated to 2.86 and 1.81 respectively when also
adjusted for previous hospital admissions with cardiovascular comorbidities.
Implications of all the available evidence
People with diabetes are at higher risk of COVID-19 related death, and those with Type 1 diabetes are at higher
risk than those with Type 2 diabetes. These insights are important in both understanding the pathophysiological
mechanisms underlying the determinants of more severe outcomes with COVID-19, and in informing potential
clinical and public health responses to the pandemic.

Introduction
By 11th May 2020, 4,252,290 people worldwide, from 213 countries and territories, were known to have had
Coronavirus disease 2019 (COVID-19), caused by SARS-CoV-2 infection, and 287,131 had died.1 Studies using
univariate analyses from China,2 Italy and the USA,3,4 and multivariate analyses from the USA and the UK,5,6
have suggested that people with diabetes have higher risks of more severe outcomes with COVID-19, including
death. The latter study, using data from General Practices in England covering approximately 40% of the
English population, included adjustments for age, ethnicity and socioeconomic deprivation.6 However, none of
these studies differentiated between Type 1 diabetes and Type 2 diabetes, a distinction which is important in
both understanding the pathophysiological mechanisms underlying the increased risk of people with diabetes
and in informing potential clinical and public health responses to that risk.
Data, including type of diabetes, are routinely collected on people diagnosed with diabetes through the National
Diabetes Audit (NDA) and, in 2018/19, 98% of General Practices in England participated in the NDA . Although
Type 1 and 2 diabetes are phenotypically distinct, with differing aetiology and pathophysiology, there can be
overlap in clinical presentation and the classification of type of diabetes relies on some degree of clinical
judgement.7 In some individuals, the recorded type of diabetes may therefore change over time. Since March
2020, the COVID Patient Notification System has recorded information on in-hospital deaths relating to
COVID-19. We used these data, collected as part of routine clinical care, to investigate the relative and absolute
risks of in-hospital death with COVID-19 infection in England for people with Type 1 diabetes and people with
Type 2 diabetes, over the period from 1st March 2020 to 11th May 2020.
Methods
Study design
A population cohort study assessing the risk of in-hospital death relating to COVID-19, covering all individuals
registered with a general practice in England and alive on February 19th 2020, assessing risk in people with
Type 1 diabetes and people with Type 2 diabetes. The observation period for deaths was 1st March 2020 to 11th
May 2020.
Data sources

In response to the increasing demand for analysis into the COVID-19 outbreak, NHS England created a bespoke
hub of relevant datasets in the National Commissioning Data Repository .
This study used de-identified data from the February 2020 Master Person Index (MPI), a reference data set
using the National Health Application and Infrastructure Services (NHAIS) which extracts General Practice
electronic patient records. Patient demographics, birth-month and year, sex, Lower Super Output Area (LSOA)
based on postcode of residence and Clinical Commissioning Group (used for health administrative purposes of
which there are 191 within 7 regions) based on General Practice registration, are included in the extract.
The latest full year extract of the NDA, covering the period 1st January 2018 to 31st March 2019, was used to
identify individuals with known diabetes. The NDA collates routinely recorded data from people with diabetes
in General Practice using the General Practice Extraction Service (GPES) and specialist care electronic clinical
records submitted using the Clinical Audit Platform.8 These data are collected by NHS Digital under a Direction
issued by NHS England under section 254 of the Health and Social Care Act for England 2012. Data are not
extracted if the person has registered their dissent from permission to use their record for secondary analysis.
Individuals are identified for inclusion in the NDA extract if they have a valid code for diabetes (excluding
gestational diabetes) in their electronic health record. Type of diabetes was based on the codes recorded in
clinical records: Type 1 diabetes, Type 2 diabetes or other diabetes (including conditions such as Maturity Onset
Diabetes of the Young (MODY)). The NDA has published reports on care process and treatment targets
annually for the past 17 years and complication and mortality reports on a less frequent basis.8 In the annual care
process and treatment target reports, different codes for the type of diabetes received from either primary or
secondary care were resolved by giving primacy to that received from a specialist care provider. In the
complication and mortality reports, the type of diagnosis most recently assigned by specialist care in any year
was used if available, and otherwise, the most recently used code from primary care was used. The same
approach was adopted for this study.
The Bridges to Health National Population Segmentation dataset was used to identify individuals’ long -term
conditions and ethnicity. The Bridges to Health Segmentation Model was developed in partnership between
Outcomes Based Healthcare,9 NHS England and Improvement, NHS Arden and GEM Commissioning Support
Unit, and Public Health England. The model utilises secondary care data sources, including over 10 years of
Secondary Uses Service, a collection of data from all hospitals in England including Admitted Patient Care
Data, Outpatient data and Emergency Care data, to categorise the England population according to their health

and care needs. The Segmentation dataset includes comorbidity and ethnicity data for individuals, derived using
activity occurring up to 31st March 2019 for comorbidity and 28th February 2020 for ethnicity.
Deaths in hospital with COVID-19 were taken from the COVID Patient Notification System (CPNS), a bespoke
daily data collection set up in March 2020 as part of the response to support COVID-19. Inclusion in this dataset
initially required a positive test for SARS-CoV-2 infection. However, this was subsequently extended on the
28th April 2020 to include those without a positive test but with COVID-19 registered as a cause of death based
on clinical judgement. This study used data extracted from this dataset on the 11th May 2020.
All datasets were pseudonymised in line with NHSE’s Purpose Specific Data Mart (PDSM) which supports
linkage of data whilst mitigating risks associated with re-identification of individuals in record level data.
Outcome
The outcome was death in hospital with COVID-19 ascertained through the CPNS.
Covariates
In addition to diabetes status, age, sex, ethnicity, and deprivation were identified as potential confounding
factors. Diabetes status was categorised as Type 1, Type 2, Other or no diabetes recorded. Sex was recorded as
male, female or missing. Age was calculated as at 1st February 2020 from birth-month and –year and grouped
into 10-year age bands. Ethnicity was classified as white, Asian, black, mixed, other or unknown. Social
deprivation was measured using Quintiles of the Index of Multiple Deprivation 2019 associated with the LSOA
derived from the individual’s postcode.10 Given the geographical variation in population exposure to SARS-
CoV-2 across England, region was also identified as a potential outcome moderator. Individuals were allocated
to one of the seven regions in England used for healthcare administration purposes (East of England; London;
Midlands; North East and Yorkshire; North West; South East; and South West) according to the responsible
Clinical Commissioning Group of the General Practice with which they were reg istered.
We included data on significant cardiovascular comorbidities (coronary heart disease (CHD), cerebrovascular
disease (CBVD) and heart failure (HF)) ascertained through coding in the Bridges to Health Segmentation
Model.9 CBVD was identified for all individuals in the MPI with an ICD10 code for Stroke (cerebral infarction
and SAH), TIA or other CBVD recorded in the Admitted Patient Care data or Outpatient data since April 2008
or a SNOMED code for Stroke or TIA recorded in the Emergency Care data since October 2017. CHD was
identified for all individuals in the MPI with an ICD10 code for Angina, Myocardial Infraction (MI) or other

CHD recorded in the Admitted Patient Care data or Outpatient data since April 2008, OPCS code for MI or
other CHD recorded in the Admitted patient Care data or Outpatient data, or SNOMED code for MI or other
CHD recorded in the Emergency Care data since October 2017. Heart Failure was identified for all individuals
in the MPI with an ICD10 code for Heart Failure recorded in the Admitted patient Care data or Outpatient data
or SNOMED code for Heart Failure recorded in the Emergency Care data since October 2017. All codes were
picked up in any diagnosis position (primary or secondary). All comorbidity data was recorded until 31 st March
2019.
Statistical analysis
The associations between diabetes status, sex, age group, ethnic group, deprivation, region and comorbidities
and mortality in hospital with COVID-19 were determined. Crude mortality rates over the 72-day observation
period per 100,000 people were calculated using the MPI population as the denominator. Mortality rates for a
given subgroup were calculated with respect to the MPI population for the given subgroup.
A multivariate logistic regression analysis was used to examine whether diabetes status was associated with in-
hospital death in England with COVID-19 adjusting for age, sex, ethnicity, socioeconomic deprivation quintile
and region. A second logistic regression model included CHD, CBVD and HF to assess the impact of these
comorbidities on the association between diabetes and in-hospital mortality with COVID-19. The C-statistic was
calculated to assess model fit. Sensitivity analyses were performed excluding unknown ethnicity. In addition, we
repeated the analyses using an alternative method for allocating diabetes type, based on national annual audit
data.
Statistical significance was defined as p-value <0.05 and confidence intervals (CI) were set at 95%. All data
were analysed using Stata version 16. All numbers taken directly from the NDA were rounded to the nearest
five persons to protect confidentiality. Data cells with between 1-4 counts in the CPNS were suppressed due to
data protection regulations.
Results
There were 61,414,470 individuals registered with a GP practice in England and alive on the 19th February
2020. Of those, 263,830 (0∙4%) had a recorded diagnosis of Type 1 diabetes, 2,864,670 (4.7%) had a recorded
diagnosis of Type 2 diabetes and 41,750 had other types of diabetes (0∙1%). The characteristics of the baseline
population in England are provided in Table 1; 49∙9% were male, the mean (SD) age was 40∙9 (23∙2) years and

13∙4% were of black, Asian, mixed and other (BAME) ethnicity (6∙1% Asian, 3∙0% black, 1∙5% mixed and
2∙7% other). Previous CHD was recorded in 3∙5% of the population, CBVD in 1∙5% and HF in 1∙0%. Data were
missing for; sex (<0.01%), ethnic group (21.2%) and deprivation quintile (0.1%). There were no missing data
for age, region and comorbidities.
Table 1 provides the characteristics of the population by type of diabetes. Compared to the general population,
individuals with Type 1 diabetes were slightly older (mean age: 46∙6 (19∙5) years) with a higher proportion of
men (56∙6%) and higher proportions of people with previous CHD, CBVD and HF; 9∙6%, 3∙7% and 3∙2%
respectively. Compared to the general population, individuals with Type 2 diabetes were older (mean age: 67.4
(13∙4) years), with a higher proportion of men (55∙9%), people of BAME ethnicity (19∙7%) (12∙0% Asian, 4∙3%
black, 0∙8% mixed and 2∙6% other) and individuals from the most deprived quintile (24∙3%). A greater
proportion of people with type 2 diabetes had evidence of previous CHD (19∙2%), CBVD (6∙6%) and HF (6∙2%)
than either those with Type 1 diabetes or the general population. The level of missing data on ethnicity was
lower in people with Type 1 diabetes (4%) and Type 2 diabetes (9%) than in the overall population.
There were 23,804 hospital deaths with COVID-19 in England reported up to 11th May 2020. Overall, one third
of these deaths occurred in people with diabetes, with Type 2 diabetes accounting for 7,466 (31∙4%) deaths,
Type 1 diabetes 365 (1∙5%) deaths and other types of diabetes 69 (0∙3%). The characteristics of people who
were recorded as having died in hospital with COVID-19 are provided in Table 2. Overall, 61∙5% were male,
the mean age was 78∙6 (12∙1) years and 16∙1% were from BAME ethnic groups (7∙5% Asian, 5∙7% black, 0∙7%
mixed and 2∙2% other). The highest proportion of deaths were in those from the most deprived quintile of the
population (23∙8%), decreasing to 15∙8% from the least deprived quintile. The highest proportion of deaths was
in London (22∙5%) followed by the Midlands (19∙7%). Only 4∙5% of deaths were from the South West region.
Previous CHD was recorded in 30∙8% of people who died, CBVD was recorded in 19∙8% and HF in 17∙7%.
Table 2 shows the characteristics of those that died in-hospital with COVID-19 by type of diabetes. Individuals
with Type 1 diabetes who died with COVID-19 in hospital were younger (mean age: 72∙2 (13∙0) years) than all
hospital deaths with COVID-19 and a higher proportion of deaths were seen in people from BAME ethnic
groups (12∙1% Asian, 10∙1% black and 3∙0% other (mixed suppressed due to small numbers)). There was a very
marked inverse relationship with deprivation with 29∙6% of deaths in people with Type 1 diabetes seen in the
most deprived quintile and only 10∙4% in the least deprived quintile. There were higher proportions of people
with a history of CHD, CBVD and HF; 47∙9%, 29∙6% and 29∙6% respectively. Individuals with Type 2 diabetes

who died with COVID-19 in hospital were of a similar age to all hospital deaths with COVID-19 (mean age:
77∙9 (11∙0) years), with a higher proportion of deaths in BAME ethnic groups (12∙8% Asian, 9∙3% black, 1∙0%
mixed and 2∙7% other). More deaths occurred in people with Type 2 diabetes in the most deprived quintile
(27∙8%) than in the least deprived quintile (12∙8%). Past CHD was recorded in 38∙8% of people with type 2
diabetes who died in hospital with COVID-19, CBVD was recorded in 22∙0% and HF in 22∙8%.
The crude rate of in-hospital mortality with COVID-19 up to 11th May 2020 was 38∙8 (38∙3-39∙3) per 100,000
persons over the 72 days for the general population. The rate per 100,000 persons in this period was 138∙3
(124∙5-153∙3) for the population with Type 1 diabetes, 260∙6 (254∙7-266∙6) for those with Type 2 diabetes and
165∙3 (128∙6-209∙2) for people with other types of diabetes (Table 2). Mortality rates increased markedly by age
group, from 0∙5 (0∙5-0∙6) per 100,000 persons aged 0-39 years to 415∙8 (408.8-423∙0) per 100,000 persons aged
80+ years (Figure 1). Within each age group, rates were significantly higher for people with Type 1 and Type 2
diabetes than for the general population, and significantly higher for Type 1 diabetes than Type 2 diabetes for
older age groups.
Results of the regression analysis showed that there was a large increase in death in-hospital with COVID-19 by
age. The odds ratio (OR) was 0∙01 (95% CI: 0∙01 to 0∙01) for individuals <40 years and 9∙14 (8∙78 to 9∙52) for
individuals aged 80+ years compared to the 60-69-year reference group. Thus, there is a 700-fold difference in
risk between those aged under 40 compared to those over 80. Odds were higher for men 1∙94 (1∙89 to 1∙99) and
were higher in those living in more deprived areas with an odds ratio of 1∙89 (1∙81 to 1∙98) in the most deprived
compared to the least deprived quintile of the population. There were higher odds for BAME ethnic groups with
ORs of 1∙35 (1∙29 to 1∙43) for Asian groups and 1∙71 (1∙61 to 1∙82) for black groups compared to the white
population. There were significant differences by region (Figure 1).
Adjusted for age, sex, deprivation, ethnicity and region, people with Type 1 diabetes had 3∙50 (3∙15-3∙89) times
the odds of dying in hospital with COVID-19, compared to the population without known diabetes, while people
with Type 2 diabetes had 2∙03 (1∙97-2∙09) times the odds of dying in hospital with COVID-19. The C-statistic
with 0.93.
In the second model, which included history of comorbidities; CHD, CBVD and HF were each significantly
associated with in-hospital death with COVID-19 with ORs of 1∙32 (1∙28 to 1∙36), 2∙23 (2∙16 to 2∙31) and 2∙23
(2∙14 to 2∙31) respectively. The association with comorbidities slightly attenuated the association with age, male
sex and deprivation seen in the model without comorbidity data. A modest attenuation was also seen for the

association with type of diabetes. Adjusted for age, sex, deprivation, ethnicity, region and cardiovascular
comorbidity, the odds ratio for dying in-hospital with COVID-19 in people with Type 1 diabetes compared to
the population without known diabetes was 2.86 and 1.81 for people with Type 2 diabetes.
In a sensitivity analysis excluding individuals with missing ethnicity data the results were unchanged
(Supplementary table S1). Nor was there any material difference in a sensitivity analysis using a definition of
diabetes based on the NDA care processes and treatment targets report (Supplementary table S2).
Discussion
This is the largest COVID-19 related study of its kind, covering almost the entire population of England, and is
the first study to investigate the relative and absolute risk of death in hospital with COVID-19 by type of
diabetes, adjusting for key confounders. It demonstrates increased risk in people with all types of diabetes, with
one third of all deaths in-hospital with COVID-19 occurring in people with diabetes.
In the time frame observed, and after adjusting for age, sex, ethnicity, socioeconomic deprivation and region,
people with Type 1 diabetes were at three and a half times the risk of in -hospital death with COVID-19, while
people with Type 2 diabetes were at twice the risk, compared to people without a diagnosis of diabetes. Further
adjustment for diagnosed cardiovascular comorbidities attenuated these risks slightly but after adjustment, there
was still an additional risk of 186% for people with Type 1 diabetes and 81% for those with Type 2 diabetes
compared to people without diabetes. The risk of all-cause mortality in people with diabetes is increased under
normal circumstances, but the observed excess risk linked to death in hospital with COVID-19 is higher than
that reported for Type 1 diabetes (148%) and Type 2 diabetes (50%) in the most recently published NDA
Complications and Mortality Report.11 People with other diagnoses of diabetes had similar risk to people with
Type 2 diabetes in both models in our analyses, however, as this category of people is relatively small and
represents a highly heterogeneous group, further inferences are limited.
This analysis, adjusting for comorbidities allows an interpretation of the independent effect of diabetes on in -
hospital death with COVID-19 beyond the well-established link between diabetes and cardiovascular
comorbidities which are themselves determinants of COVID-19 mortality risk. In this and previous analyses, HF
and CBVD have been shown to be associated with serious outcomes related to COVID-19.5,6 We demonstrate
an association between previous CHD and mortality, an association seen in some but not all previous studies. 5,6

These co-morbidities were selected for inclusion in analyses in this study as they are recognised cardiovascular
complications of diabetes and are well represented within the Bridges to Health Segmentation Model compared
to recorded prevalence in the Quality and Outcomes Framework, 12 an annual performance-based incentive
programme for General Practices in England. A limitation of this study is that other comorbidities were not
included in analyses. In particular, hypertension and chronic kidney disease were not included due to incomplete
recording in the hospital-derived Bridges to Health Segmentation Model. These comorbidities are better defined
in primary care acquired datasets. A systematic review suggested an association between poor COVID-19
related outcomes and hypertension,13 although this has not been detected in some multivariate analyses which
have shown significant associations with chronic kidney disease, 5,6 a common complication of diabetes that
could partially mediate the higher risks described previously. Further studies are needed to examine this in the
future.
As reported in a recently published multivariate analysis using data from England, 6 our analyses showed an
increased risk of in-hospital death with COVID-19 for older people, men, people of black, Asian or mixed
ethnicity and those who live in areas of high socioeconomic deprivation . While a number of studies have
reported an association between diabetes and severe outcomes of COVID-19, 2-6 the findings here are novel in
suggesting that the influence of diabetes on risk of death with COVID-19 is independent of age, ethnicity,
deprivation and cardiovascular comorbidities, and is seen in people with all sub-types of diabetes, being highest
in those with Type 1 diabetes.
There are many possible reasons for the increased risk of death in hospital from COVID-19 in people with Type
1 diabetes compared to those with Type 2 diabetes. It may be hypothesised that the difference in risk could
relate to the different aetiology and pathophysiology of the types of diabetes, varying patterns of diabetes
complications or iatrogenic harms (such as hypoglycaemia), differing patterns, treatments, intensity an d duration
of glycaemia, or the influence of comorbidities which were either not adjusted for in these analyses or for which
we only imperfectly adjusted. An excess risk of other infectious disease morbidity and mortality has previously
been observed in Type 1 compared to Type 2 diabetes. The risk of developing pneumonia was reported to be
2.98 higher for Type 1 diabetes and 1.58 for Type 2 diabetes compared to the general population. 14
On a relative scale, our analyses show that Type 1 diabetes was associated with more than twice the additional
risk of in-hospital death with COVID-19 compared to the non-diabetic population (186%) than was observed in
people with Type 2 diabetes (81%). However, on an absolute scale, the unadjusted rates of in-hospital death

with COVID-19 over this 72-day period for Type 1 diabetes (138 per 100,000 persons) were almost half that for
Type 2 diabetes (261 per 100,000 person), largely reflecting the different age structure of the two populations.
Age was the dominant risk factor for in-hospital death with COVID-19 and had a much greater influence on risk
than diabetes status, sex, ethnicity or socioeconomic deprivation. Even with the additional risk associated with
Type 1 diabetes or Type 2 diabetes, people under the age of 40 years with either type of diabetes were at very
low absolute risk of in-hospital death with COVID-19 during the observation period of this study in England.
Conclusion
The findings of the study have important implications for people with diabetes, healthcare professionals and
policy makers. We would encourage the use of these findings, along with those from other studies investigating
associations with serious COVID-19 related outcomes, to provide reassurance for people who are at low
absolute risk, despite having diabetes. For those who are at higher risk the results inform public guidance
including recommendations for shielding. Further elucidation of the modifiable risk factors for poorer COVID-
19 outcomes in people with diabetes will be critical in guiding management and providing targeted support.

Funding
NHS England & Improvement and Public Health England provided resources for these analyses.
Author contributions
Jonathan Valabhji, Emma Barron, Chirag Bakhai, Andy Weaver, Naomi Holman, Kamlesh Khunti, Naveed
Sattar, Nick Wareham, and Bob Young conceived the study. Emma Barron, Dominique Bradley, Hassan Ismail,
Naomi Holman and Peter Knighton managed the data and carried out the statistical analysis. All the authors
collaborated in interpretation of the results and drafting of the manuscript.
Declarations of Interest
Jonathan Valabhji is the National Clinical Director for Diabetes and Obesity at NHS England & Improvement.
Partha Kar is National Specialty Advisor for Diabetes and Obesity at NHS England & Improvement. Chirag
Bakhai is the Primary Care Advisor to the NHS Diabetes Programme. Bob Young is Clinical lead for the
National Diabetes Audit and a trustee of Diabetes UK. Kamlesh Khunti has acted as a consultant and speaker for
Novartis, Novo Nordisk, Sanofi-Aventis, Lilly and Merck Sharp & Dohme. Kamlesh Khunti has also received
grants in support of investigator and investigator-initiated trials from Novartis, Novo Nordisk, Sanofi-Aventis,
Lilly, Merck Sharp & Dohme, Pfizer and Boehringer Ingelheim and has served on advisory boards for Novo
Nordisk, Sanofi-Aventis, Lilly and Merck Sharp & Dohme. Naveed Sattar has consulted for Amgen,
Astrazeneca, Boehringer Ingelheim, Eli Lilly, Novo Nordisk, Pfizer and Sanofi and received grant support from
Boehringer Ingelheim.

References
1. Worldometer. COVID-19 coronavirus pandemic. May 11, 2020. https://www.worldometers.info/coronavirus/
2. Zhou F, Yu T, Du R et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in
Wuhan, China: a retrospective cohort study. Lancet 2020; 395: 1054-62.
3. Istituto Superiore di Sanita: Report of characteristics of patients died positive for COVID-19 in Italy. https
://www.epice ntro.iss. it/coron aviru s/bolle ttino /Repor t-COVID -2019_17_marzo -v2. pdf Article in Italian
4. Bode B, Garrett V, Messler J et al. Glycemic Characteristics and Clinical Outcomes of COVID-19 Patients
Hospitalized in the United States. Available at: https://glytecsystems.com/wp-content/uploads/JDST-Glytec-
Covid-Research.pdf
5. Petrilli CM, Jones SA, Yang J et al Factors associated with hospitalization and critical illness among 4,103
patients with COVID-19 disease in New York City. medRxiv 2020.04.08.20057794.
https://www.medrxiv.org/content/10.1101/2020.04.08.20057794v15.
6. The OpenSAFELY Collaborative. OpenSAFELY: factors associated with COVID-19-related hospital death
in the linked electronic health records of 17 million adult NHS patients. medRxiv preprint posted 07-05-2020
available at: https://www.medrxiv.org/content/10.1101/2020.05.06.20092999v1
7. Tuomi T, Santoro N, Caprio S, Cai M, Weng J and Groop L. The many faces of diabetes: a disease with
increasing heterogeneity. Lancet 2014; 383: 1084-94
8. National Diabetes Audit. Available at: https://digital.nhs.uk/data-and-information/clinical-audits-and-

registries/national-diabetes-audit [Accessed: May 2020]
9. Outcomes Based Healthcare. Bridges to Health segmentation model: Person-level clinical segmentation data
model produced by Outcomes Based Healthcare® Ltd (OBH). Version: 1.0, delivered under licence to NHSEI
and AGEM CSU 04.12.19. Copyright © 2019 Outcomes Based Healthcare
10. English Indices of deprivation. https://www.gov.uk/government/statistics/english-indices-of-deprivation-

2019 [Access May 2020]
11. National Diabetes Audit – Report 2 Complications and Mortality, 2017-18. Published December 13, 2019.
Available at: https://digital.nhs.uk/data-and-information/publications/statistical/national-diabetes-audit/report-2-
-complications-and-mortality-2017-18 [Access May 2020]
12. Quality and Outcomes Framework. https://qof.digital.nhs.uk/ [Access May 2020]
13. Yang et al. Prevalence of comorbidities and its effects in patients infects with SARS-CoV-2: a systematic
review and meta-analysis. International Journal of Infectious Diseases 2020; 94; 91-95
14. Carey IM, Critchley JA, DeWilde S, Harris T, Hosking FJ and Cook DG. Risk of infection in Type 1 and
Type 2 diabetes compared with the general population: A matched cohort study. Diabetes Care 2018; 41: 513-
521.

This version posted 19th May 2020. The copyright holder of this pre-print (which has not been certified by peer review) is NHS England.
Tables and Figures
Table 1: Number of individuals registered with a GP practice in England and alive at 19 th February 2020 by diabetes type
N Percentage
Overall Type 1 Type 2 Other No Diabetes Overall Type 1 Type 2 Other No Diabetes
Total 61,414,470 263,830 2,864,670 41,750 58,244,220 100∙0% 100∙0% 100∙0% 100∙0% 100∙0%
0 to 39 years 30,506,055 100,760 67,735 6,815 30,330,745 49∙7% 38∙2% 2∙4% 16∙3% 52∙1%
40 to 49 years 8,073,780 41,680 212,945 5,630 7,813,525 13∙1% 15∙8% 7∙4% 13∙5% 13∙4%
50 to 59 years 8,266,300 49,160 519,825 8,520 7,688,795 13∙5% 18∙6% 18∙1% 20∙4% 13∙2%
Age
60 to 69 years 6,359,460 36,125 723,790 8,510 5,591,035 10∙4% 13∙7% 25∙3% 20∙4% 9∙6%
70 to 79 years 5,057,230 24,180 766,815 7,215 4,259,020 8∙2% 9∙2% 26∙8% 17∙3% 7∙3%
80+ 3,151,645 11,925 573,560 5,060 2,561,095 5∙1% 4∙5% 20∙0% 12∙1% 4∙4%
Male 30,635,515 149,330 1,601,045 22,610 28,862,530 49∙9% 56∙6% 55∙9% 54∙2% 49∙6%
Sex Female 30,778,160 114,495 1,263,615 19,140 29,380,910 50∙1% 43∙4% 44∙1% 45∙8% 50∙4%
Unknown 790 5 10 0 775 0∙0% 0∙0% 0∙0% 0∙0% 0∙0%
Asian 3,769,395 14,030 344,780 4,355 3,406,230 6∙1% 5∙3% 12∙0% 10∙4% 5∙8%
Black 1,867,605 8,570 122,985 2,095 1,733,955 3∙0% 3∙2% 4∙3% 5∙0% 3∙0%
Mixed 937,125 3,025 22,265 465 911,365 1∙5% 1∙1% 0∙8% 1∙1% 1∙6%
Ethnic group
Other 1,671,615 4,880 74,385 1,265 1,591,085 2∙7% 1∙8% 2∙6% 3∙0% 2∙7%
White 40,132,970 222,795 2,042,950 28,370 37,838,855 65∙3% 84∙4% 71∙3% 68∙0% 65∙0%
Unknown 13,035,760 10,530 257,300 5,200 12,762,725 21∙2% 4∙0% 9∙0% 12∙5% 21∙9%
IMD 1 (most
deprived) 12,757,075 55,930 696,675 10,360 11,994,110 20∙8% 21∙2% 24∙3% 24∙8% 20∙6%
IMD 2 12,817,845 53,965 638,925 9,430 12,115,530 20∙9% 20∙5% 22∙3% 22∙6% 20∙8%
Deprivation
IMD 3 12,306,210 53,330 573,660 8,430 11,670,790 20∙0% 20∙2% 20∙0% 20∙2% 20∙0%
quintile
IMD 4 11,876,070 51,425 513,315 7,245 11,304,090 19∙3% 19∙5% 17∙9% 17∙4% 19∙4%
IMD 5 (least deprived) 11,606,695 48,985 440,200 6,250 11,111,265 18∙9% 18∙6% 15∙4% 15∙0% 19∙1%
Unknown 50,570 200 1,900 30 48,435 0∙1% 0∙1% 0∙1% 0∙1% 0∙1%
East 7,053,615 32,420 310,725 5,265 6,705,205 11∙5% 12∙3% 10∙8% 12∙6% 11∙5%
Region London 10,545,135 33,225 463,180 7,145 10,041,585 17∙2% 12∙6% 16∙2% 17∙1% 17∙2%
Midlands 11,397,835 53,140 583,885 8,495 10,752,320 18∙6% 20∙1% 20∙4% 20∙3% 18∙5%
15
N Percentage
Overall Type 1 Type 2 Other No Diabetes Overall Type 1 Type 2 Other No Diabetes
NE & Yorks 9,179,220 43,830 461,840 6,175 8,667,375 14∙9% 16∙6% 16∙1% 14∙8% 14∙9%
North West 7,686,250 32,180 372,540 4,340 7,277,195 12∙5% 12∙2% 13∙0% 10∙4% 12∙5%
South East 9,581,305 41,500 400,965 5,600 9,133,245 15∙6% 15∙7% 14∙0% 13∙4% 15∙7%
South West 5,971,105 27,540 271,545 4,725 5,667,295 9∙7% 10∙4% 9∙5% 11∙3% 9∙7%
Coronary Heart No admission 59,259,570 238,460 2,314,195 36,680 56,670,235 96∙5% 90∙4% 80∙8% 87∙9% 97∙3%
Disease
Admission 2,154,900 25,375 550,475 5,065 1,573,985 3∙5% 9∙6% 19∙2% 12∙1% 2∙7%
Cerebrovascular No admission 60,498,915 254,155 2,674,260 39,735 57,530,765 98∙5% 96∙3% 93∙4% 95∙2% 98∙8%
Disease
Admission 915,555 9,680 190,410 2,010 713,455 1∙5% 3∙7% 6∙6% 4∙8% 1∙2%
No admission 60,783,235 255,350 2,686,460 39,880 57,801,545 99∙0% 96∙8% 93∙8% 95∙5% 99∙2%
Heart failure
Admission 631,235 8,485 178,210 1,865 442,675 1∙0% 3∙2% 6∙2% 4∙5% 0∙8%
16
Table 2: Deaths in-hospital with COVID-19 in England between 1 st March 2020 and 11 th May 2020 by diabetes type
Number of COVID-19 deaths Percentage Rate per 100,000 persons over 72 days
Type Type No No No
Overall 1 2 Other Diabetes Overall Type 1 Type 2 Other Diabetes Overall Type 1 Type 2 Other Diabetes
Total 23,804 365 7,466 69 15,904 100% 100% 100% 100% 100% 38∙8 138∙3 260∙6 165∙3 27∙3
(38∙3- (124∙5- (254∙7- (128∙6- (26∙9-
39∙3) 153∙3) 266∙6) 209∙2) 27∙7)
Age 0 to 39 164 * 20 * * 0.7% * 0.3% * * 0∙5 (0∙5- * 29∙5 * *
years 0∙6) (18∙0-
45∙6)
40 to 49 390 * 91 * 279 1.6% * 1.2% * 1.8% 4∙8 (4∙4- * 42∙7 * 3∙6
years 5∙3) (34∙4- (3∙2-
52∙5) 4)
50 to 59 1,333 49 405 * * 5.6% 13.4% 5.4% * * 16∙1 99∙7 77∙9 * *
years (15∙3- (73∙7- (70∙5-
17) 131∙8) 85∙9)
60 to 69 2,890 73 1,048 7 1,762 12.1% 20.0% 14.0% 10.1% 11.1% 45∙4 202∙1 144∙8 82∙3 31∙5
years (43∙8- (158∙4- (136∙2- (33∙1- (30∙1-
47∙1) 254∙1) 153∙8) 169∙5) 33)
70 to 79 5,921 98 2,103 22 3,698 24.9% 26.8% 28.2% 31.9% 23.3% 117∙1 405∙3 274∙3 305 86∙8
years (114∙1- (329∙1- (262∙7- (191∙1- (84∙1-
120∙1) 493∙9) 286∙2) 461∙8) 89∙7)
80+ 13,106 125 3,799 32 9,150 55.1% 34.2% 50.9% 46.4% 57.5% 415∙8 1048∙1 662∙4 632∙2 357∙3
(408∙8- (872∙5- (641∙5- (432∙4- (350-
423) 1248∙8) 683∙8) 892∙4) 364∙7)
Sex Male 14,636 232 4,815 46 9,543 61.5% 63.6% 64.5% 66.7% 60.0% 47∙8 155∙4 300∙7 203∙5 33∙1
(47- (136-1 (292∙3- (149- (32∙4-
48∙6) 76∙7) 309∙4) 271∙4) 33∙7)
Female 9,168 133 2,651 23 6,361 38.5% 36.4% 35.5% 33.3% 40.0% 29∙8 116∙2 209∙8 120∙2 21∙7
(29∙2- (97∙3- (201∙9- (76∙2- (21∙1-
30∙4) 137∙7) 217∙9) 180∙3) 22∙2)
Unknown 0 0 0 0 0 0% 0% 0% 0% 0% 0 0 0 n/a 0
(0- (0- (0- (0-
466∙63) 61481∙3) 46110∙9) 474∙76)
Ethnic group Asian 1,779 44 959 * * 7.5% 12.1% 12.8% * * 47∙2 313∙6 278∙1 * *
(45- (227∙9- (260∙8-
49∙4) 421) 296∙3)
Black 1,360 37 698 5 620 5.7% 10.1% 9.3% 7.2% 3.9% 72∙8 431∙7 567∙5 238∙7 35∙8 (33-
(69- (304- (526∙2- (77∙5- 38∙7)
76∙8) 595∙1) 611∙3) 557)
Mixed 171 * 75 * 91 0.7% * 1.0% * 0.6% 18∙2 * 336∙8 * 10 (8-
(15∙6- (264∙9- 12∙3)
21∙2) 422∙2)
17
Type Type No No No
Other 517 11 203 0 303 2.2% 3.0% 2.7% 0.0% 1.9% 30∙9 225∙4 272∙9 0 19 (17-
(28∙3- (112∙5- (236∙6- (0- 21∙3)
33∙7) 403∙3) 313∙1) 291∙6)
White 19,042 267 5,351 53 13,371 80.0% 73.2% 71.7% 76.8% 84.1% 47∙4 119∙8 261∙9 186∙8 35∙3
(46∙8- (105∙9- (255-269) (139∙9- (34∙7-
48∙1) 135∙1) 244∙4) 35∙9)
Unknown 935 * 180 * * 3.9% * 2.4% * * 7∙2 (6∙7- * 70 * *
7∙6) (60∙
1-81)
Deprivation IMD 1 5,662 108 2,073 14 3,467 23.8% 29.6% 27.8% 20.3% 21.8% 44∙4 193∙1 297∙6 135∙1 28∙9 (28-
quintile (most (43∙2- (158∙4- (284∙9- (73∙9- 29∙9)
deprived) 45∙6) 233∙1) 310∙6) 226∙7)
IMD 2 5,376 80 1,866 17 3,413 22.6% 21.9% 25.0% 24.6% 21.5% 41∙9 148∙2 292∙1 180∙3 28∙2
(40∙8- (117∙5- (279- (105- (27∙2-
43∙1) 184∙5) 305∙6) 288∙6) 29∙1)
IMD 3 4,650 88 1,411 12 3,139 19.5% 24.1% 18.9% 17.4% 19.7% 37∙8 165 246 142∙3 26∙9 (26-
(36∙7- (132∙3- (233∙3- (73∙5- 27∙9)
38∙9) 203∙3) 259∙1) 248∙6)
IMD 4 4,315 51 1,154 12 3,098 18.1% 14.0% 15.5% 17.4% 19.5% 36∙3 99∙2 224∙8 165∙7 27∙4
(35∙3- (73∙8- (212- (85∙6- (26∙4-
37∙4) 130∙4) 238∙2) 289∙4) 28∙4)
IMD 5 3,771 38 952 14 2,767 15.8% 10.4% 12.8% 20.3% 17.4% 32∙5 77∙6 216∙3 224 24∙9 (24-
(least (31∙5- (54∙9- (202∙7- (122∙5- 25∙8)
deprived) 33∙5) 106∙5) 230∙5) 375∙8)
Unknown 30 0 10 0 20 0.1% 0.0% 0.1% 0.0% 0.1% 59∙3 0 526 0 41∙3
(40- (0- (252∙3- (0- (25∙2-
84∙7) 1844∙4) 967∙4) 11527∙7) 63∙8)
Region East of 2,837 50 766 16 2,005 11.9% 13.7% 10.3% 23.2% 12.6% 40∙2 154∙2 246∙5 303∙8 29∙9
England (38∙8- (114∙5- (229∙4- (173∙7- (28∙6-
41∙7) 203∙3) 264∙6) 493∙4) 31∙2)
London 5,360 82 1,927 14 3,337 22.5% 22.5% 25.8% 20.3% 21.0% 50∙8 246∙8 416 195∙9 33∙2
(49∙5- (196∙3- (397∙7- (107∙1- (32∙1-
52∙2) 306∙3) 435) 328∙7) 34∙4)
Midlands 4,682 96 1,504 9 3,073 19.7% 26.3% 20.1% 13.0% 19.3% 41∙1 180∙7 257∙6 105∙9 28∙6
(39∙9- (146∙3- (244∙7- (48∙4- (27∙6-
42∙3) 220∙6) 270∙9) 201∙1) 29∙6)
North East 3,326 41 993 10 2,282 14.0% 11.2% 13.3% 14.5% 14.3% 36∙2 93∙5 215 161∙9 26∙3
and (35- (67∙1- (201∙8- (77∙7- (25∙3-
Yorkshire 37∙5) 126∙9) 228∙8) 297∙8) 27∙4)
North West 3,603 43 1,061 6 2,493 15.1% 11.8% 14.2% 8.7% 15.7% 46∙9 133∙6 284∙8 138∙2 34∙3
(45∙4- (96∙7- (267∙9- (50∙7- (32∙9-
48∙4) 180) 302∙5) 300∙9) 35∙6)
South East 2,920 33 882 8 1,997 12.3% 9.0% 11.8% 11.6% 12.6% 30∙5 79∙5 220 142∙8 21∙9
(29∙4- (54∙7- (205∙7- (61∙7- (20∙9-
31∙6) 111∙7) 235) 281∙4) 22∙8)
18
Type Type No No No
South West 1,076 20 333 6 717 4.5% 5.5% 4.5% 8.7% 4.5% 18 72∙6 122∙6 127 12∙7
(17- (44∙4- (109∙8- (46∙6- (11∙7-
19∙1) 112∙2) 136∙5) 276∙5) 13∙6)
Coronary Heart No 16,463 190 4,572 46 11,655 69.2% 52.1% 61.2% 66.7% 73.3% 27∙8 79∙7 197∙6 125∙4 20∙6
Disease admission (27∙4- (68∙8- (191∙9- (91∙8- (20∙2-
28∙2) 91∙8) 203∙4) 167∙3) 20∙9)
Admission 7,341 175 2,894 23 4,249 30.8% 47.9% 38.8% 33.3% 26.7% 340∙7 689∙7 525∙7 453∙9 270
(332∙9- (591∙3- (506∙7- (287∙7- (261∙9-
348∙5) 799∙8) 545∙2) 681∙1) 278∙2)
Cerebrovascular No 19,090 257 5,825 53 12,955 80.2% 70.4% 78.0% 76.8% 81.5% 31∙6 101∙1 217∙8 133∙4 22∙5
Disease admission (31∙1- (89∙1- (212∙3- (99∙9- (22∙1-
32) 114∙3) 223∙5) 174∙5) 22∙9)
Admission 4,714 108 1,641 16 2,949 19.8% 29.6% 22.0% 23.2% 18.5% 514∙9 1115∙8 861∙8 795∙6 413∙3
(500∙3- (915∙3- (820∙6- (454∙8- (398∙6-
529∙8) 1347∙2) 904∙6) 1292) 428∙5)
Heart failure No 19,582 257 5,765 55 13,505 82.3% 70.4% 77.2% 79.7% 84.9% 32∙2 100∙6 214∙6 137∙9 23∙4
admission (31∙8- (88∙7- (209∙1- (103∙9- (23-
32∙7) 113∙7) 220∙2) 179∙5) 23∙8)
Admission 4,222 108 1,701 14 2,399 17.7% 29.6% 22.8% 20.3% 15.1% 668∙8 1273∙1 954∙5 750∙3 541∙9
(648∙8- (1044∙4- (909∙7- (410∙2- (520∙5-
689∙3) 1537∙1) 1001) 1258∙8) 564∙1)
*Suppressed due to small numbers
19
Figure 1: Unadjusted in-hospital COVID-19 mortality rate per 100,000 persons between 1 st March 2020
to 11 th May 2020 by type of diabetes
*Age groups for 0-39 Type 1 and 40-49 for Type 1 have been suppressed due to small numbers of events to
comply with data protection regulations.
20
Figure 2: Adjusted odds ratios for in-hospital deaths with COVID-19 in England (number of
deaths=23,804) between 1 st March 2020 and 11 th May 2020 by different risk factors
*Data shown are the results of a multivariable logistic regression which included the explanatory variables
shown, plus region, in a population of 61,414,470 people.
21
Figure 3: Adjusted odds ratios for in-hospital death with COVID-19 in England (number of
deaths=23,804) between 1 st March 2020 and 11 th May 2020 by different risk factors including
cardiovascular co-morbidities
*Data shown are the results of a multivariable logistic regression which included the explanatory variables
shown, plus region, in a population of 61,414,470 people
22
Supplementary materials
Table S1: COVID-19 related hospital deaths in England between 1 st March 2020 and 11 th May 2020;
excluding individuals with unknown ethnicity, multivariate logistic regression
95% CI
Death Odds ratio 95% CI lower upper P value
Age < 40 0∙01 0∙01 0∙01 <0∙001
40-49 0∙11 0∙10 0∙12 <0∙001
50-59 0∙35 0∙33 0∙38 <0∙001
60-69 1∙00
70-79 2∙61 2∙50 2∙74 <0∙001
80+ 9∙11 8∙73 9∙49 <0∙001
Sex Female 1∙00
Male 1∙93 1∙88 1∙98 <0∙001
Deprivation quintile IMD 1 (most deprived) 1∙90 1∙82 1∙98 <0∙001
IMD 2 1∙54 1∙48 1∙61 <0∙001
IMD 3 1∙27 1∙21 1∙32 <0∙001
IMD 4 1∙14 1∙09 1∙19 <0∙001
IMD 5 (least deprived) 1∙00
Unknown 2∙22 1∙54 3∙20 <0∙001
Region London 1∙00
South West 0∙24 0∙22 0∙25 <0∙001
South East 0∙49 0∙46 0∙51 <0∙001
Midlands 0∙57 0∙55 0∙60 <0∙001
East of England 0∙61 0∙58 0∙64 <0∙001
North West 0∙66 0∙63 0∙69 <0∙001
North East and Yorkshire 0∙50 0∙48 0∙52 <0∙001
Diabetes status No diabetes 1∙00
Type 1 3∙50 3∙15 3∙88 <0∙001
Type 2 2∙01 1∙96 2∙07 <0∙001
Other type 2∙16 1∙70 2∙74 <0∙001
Ethnic group White 1∙00
Asian 1∙36 1∙29 1∙44 <0∙001
Mixed 1∙43 1∙23 1∙67 <0∙001
Black 1∙73 1∙63 1∙83 <0∙001
Other Ethnic Groups 1∙12 1∙03 1∙23 0∙012
23
Table S2: COVID-19 related hospital deaths in England between 1 st March 2020 and 11 th May 2020; an
alternative method of defining type of diabetes based on the NDA care processes and treatment targets
report, multivariate logistic regression
95% CI 95% CI
Death Odds ratio lower upper P value
Age < 40 0∙01 0∙01 0∙01 <0∙001
40-49 0∙11 0∙10 0∙12 <0∙001
50-59 0∙36 0∙34 0∙39 <0∙001
60-69 1∙00
70-79 2∙63 2∙51 2∙75 <0∙001
80+ 9∙14 8∙77 9∙52 <0∙001
Sex Female 1∙00
Male 1∙94 1∙89 1∙99 <0∙001
Deprivation quintile IMD 1 (most deprived) 1∙89 1∙81 1∙98 <0∙001
IMD 2 1∙54 1∙48 1∙61 <0∙001
IMD 3 1∙26 1∙21 1∙32 <0∙001
IMD 4 1∙14 1∙09 1∙19 <0∙001
IMD 5 (least deprived) 1∙00
Unknown 2∙20 1∙54 3∙16 <0∙001
Region London 1∙00
South West 0∙23 0∙22 0∙25 <0∙001
South East 0∙48 0∙46 0∙51 <0∙001
Midlands 0∙57 0∙54 0∙59 <0∙001
East of England 0∙60 0∙58 0∙63 <0∙001
North West 0∙65 0∙62 0∙68 <0∙001
North East and Yorkshire 0∙49 0∙47 0∙51 <0∙001
Diabetes status No diabetes 1∙00
Type 1 3∙30 2∙91 3∙74 <0∙001
Type 2 2∙04 1∙99 2∙10 <0∙001
Other type 2∙14 1∙69 2∙71 <0∙001
Ethnic group White 1∙00
Asian 1∙36 1∙29 1∙43 <0∙001
Mixed 1∙43 1∙23 1∙66 <0∙001
Black 1∙72 1∙62 1∙82 <0∙001
Other Ethnic Groups 1∙11 1∙02 1∙22 0∙02
Unknown 0∙33 0∙31 0∙35 <0∙001
24

Type 1 and Type 2 Diabetes and COVID-19 Related Mortality in England: A Whole Population Study

Uploaded by

Copyright:

Available Formats

Type 1 and Type 2 Diabetes and COVID-19 Related Mortality in England: A Whole Population Study

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Type 1 and Type 2 Diabetes and COVID-19 Related Mortality in England: A Whole Population Study

Uploaded by

Copyright:

Available Formats

This version posted 19th May 2020.

MD, Nick Wareham7 PhD, Bob Young 8 MD, Jonathan Valabhji2,9,10 MD

1. Public Health England, Station Rise, York, YO1 6GA, UK

4. NHS Digital, 1 Trevelyan Square, Board Lane, Leeds, LS1 6AE, UK

Campus, Cambridge CB2 0QQ

10. Imperial College London, South Kensington, London, SW7 2BU , UK

NHS England & Improvement, Skipton House, London, UK

Tel: 0113 8251692

Word count: Abstract 250; Full text 3,681

Currently undergoing peer review for journal publication

Type 2 diabetes are unknown.

adjusting for demographic factors and comorbidities.

disease, cerebrovascular disease or heart failure.

significant increased risk of in-hospital death with COVID-19.

NHS England & Improvement and Public Health England.

Currently undergoing peer review for journal publication

Evidence before this study

differences in risk by type of diabetes.

Added value of this study

adjusted for previous hospital admissions with cardiovascular comorbidities.

Implications of all the available evidence

clinical and public health responses to the pandemic.

Currently undergoing peer review for journal publication

Currently undergoing peer review for journal publication

hub of relevant datasets in the National Commissioning Data Repository .

approach was adopted for this study.

Currently undergoing peer review for journal publication

Currently undergoing peer review for journal publication

data protection regulations.

Currently undergoing peer review for journal publication

for age, region and comorbidities.

Currently undergoing peer review for journal publication

older age groups.

population. There were significant differences by region (Figure 1).

Currently undergoing peer review for journal publication

represents a highly heterogeneous group, further inferences are limited.

Currently undergoing peer review for journal publication

in those with Type 1 diabetes.

Currently undergoing peer review for journal publication

Currently undergoing peer review for journal publication

collaborated in interpretation of the results and drafting of the manuscript.

Currently undergoing peer review for journal publication

1. Worldometer. COVID-19 coronavirus pandemic. May 11, 2020. https://www.worldometers.info/coronavirus/

8. National Diabetes Audit. Available at: https://digital.nhs.uk/data-and-information/clinical-audits-and-

10. English Indices of deprivation. https://www.gov.uk/government/statistics/english-indices-of-deprivation-

12. Quality and Outcomes Framework. https://qof.digital.nhs.uk/ [Access May 2020]

Currently undergoing peer review for journal publication

Tables and Figures

*Suppressed due to small numbers

You might also like