Diabetes Diagnosis During Acute Admissions (FIND-IT) (FIND-IT)

Study aims

Primary research question What is the prevalence of glucose intolerance (diabetes and IGT) in patients presenting to an acute medical emergency department?

Objectives

The aim of the project is to see whether this is a simple method of identifying the many patients who have undiagnosed glucose intolerance and ensuring those who are identified are commenced on the right treatment path for them to reduce both individual risk and burden to the NHS.

Patients presenting to A&E have routine bloods taken for condition which are they are being investigated and treated for. Therefore the aim is to identify the prevalence of undiagnosed diabetes (HbA1c >48mmol/mol) or impaired glucose tolerance/pre-diabetes (HbA1c >39mmol/mol) in patients attending the accident and emergency department or acute medical unit and to see if this is a good screening measure for diagnosis of diabetes. This project will help identify those undiagnosed with glucose intolerance (T2D and IGT) and instigate appropriate treatment and improve outcomes for this group of patients. This will in the long term reduce the burden to the NHS. This project will help in the development of guidance for diagnosis of T2D in an acute setting and treatment for hospital admission and continued care.

This project will include 10,000 consecutive patients over the age of 30 years attending the A&E or AMU departments of Tameside and Glossop Integrated Care NHS FT. All patients will be screened for glucose intolerance with a blood test in which patients' blood would be taken anyway for clinical reasons and we will perform an HbA1c investigation on the sample collected.

In addition to measuring HbA1c the investigators will look for other risk factors, such as socio economic status, obesity, ethnicity, weekly exercise patterns and other metabolic factors (e.g. lipids).

All results will be forwarded to the GP with advice on management if HbA1c is raised and to follow-up the patient.

This will be the largest UK project to include patients presenting to A&E and assessed for underlying glucose intolerance. The project will run for 12 months.

Study design

Due to the observational nature of this study, no blinding is required to the investigators as the primary question is laboratory diagnosis and not treatment received

This will be an observational cohort study of patients hospitalised at Tameside General Hospital,

Patient demographic details will be extracted from their clinical notes and electronic patient records, then placed on an anonymised database for analysis. Data will be analysed to determine findings, including diagnosis of T2D/IGT, reason for admission, presence of diabetic complications (microvascular / macrovascular). Projected recruitment is minimum of 10,000 patients to be opportunistically recruited from A&E and AMU and acute medical wards of TGH.

Study groups

Patients will be opportunistically recruited from in-patient admissions with a medical illness and presenting either to the A&E, AMU or acute medical wards.

Inclusion criteria:

• Aged ≥30 years.
• Males and female sex

Exclusion criteria:

• Aged <30 years.
• Type 1 diabetes

Data will be collected for patients based on whether they have been recorded as having a clinical diagnosis of diabetes or not by history and GP records.

A secure, password-encrypted database will be held on a secure server at the hospital, detailing the patient's identifiable details, as well as an assigned study number. Once patient study demographics have been obtained, these will be held on a separate, anonymised database with only the study number as the identifier.

FINDRISC score will be assessed for all patients.

Data Data to be collected Please see Appendix 1 for the data extraction pro forma A detailed list of data to be extracted is laid out in Tables 1 and 2. Data will be collected by co-investigators/research nurses.

Table 1. Data to be collected at enrolment, if available as part of clinical care

Variable Source Form of data Patient identifiable data Electronic patient record (EPR) String Diagnosis on admission Patient medical records String Co-morbidities Patient medical records Binary i.e. has/does not have condition Weight and height EPR/patient medical records Continuous Diabetes Patient medical records Binary i.e. yes/no for each type Treatment history Patient medical records Binary i.e. received/did not receive Vitamin D level (if measured) EPR/laboratory Continuous - could be converted to binary e.g. replete/deplete Vitamin D treatment Patient medical records Binary i.e. yes/no Ethnicity EPR/patient medical records Categorical Creatinine on admission EPR Continuous Foot examination Patient history / examination Binary i.e. yes/no Cardiovascular disease (see proforma) Patient medical records Binary i.e. yes/no Death EPR/patient medical records Binary i.e. yes/no Discharge EPR/patient medical records Binary i.e. yes/no

Data handling and record keeping Data will be manually extracted from the sources detailed above and stored on a password-encrypted database held on Microsoft Excel software. This will be stored on a secure server with standard NHS encryption and firewalls at Tameside General Hospital. Data will be cross-validated by two independent investigators to ensure validity of data and quality of data entry. Records will be retained for ten years after the study concludes for the purposes of potential secondary analysis, then it will be destroyed. However all personal identifiable data will be destroyed at 5 years. All data collection will adhere to the Data Protection Act 1998 and the Caldicott Principles.

Statistical considerations

Sample size calculations Since this is an observational study sample size calculation is not required.

Analysis All analysis will be carried out by principal investigator. Subgroup analysis will also be carried out, stratified by age group, gender and ethnicity. Approximately 20% of the general population of Tameside and Glossop are Black, Asian and Minority Ethnic (BAME), so it is anticipated that subanalysis will be able to be conducted by ethnicity.

Primary analysis will also include logistic regression of ethnic groups against the above treatment outcome variables, with adjustment for age, gender and diabetes status.

Compliance

Subject compliance Compliance will not be an issue, as all data will be gathered from hospital in-patient admission data.

Withdrawal of subjects As this is an observational study, there is no cause for the investigators to withdraw a subject from the study.

Ethical considerations

Appropriate ethical approvals will be sought through the current Health Research Authority (HRA) approval system. Once an outcome has been established, we will update the protocol with the IRAS number.

As the patient data required will already be available as part of clinical care, there will be no need to obtain informed consent from patients. As patients will already be having blood taken for routine care no extra blood will be necessary.

Personal identifiable data will not leave Tameside and Glossop Integrated Care NHS Foundation Trust and will adhere to the GDPR. Confidential patient information will be processed solely for research, in accordance with the Department of Health and Social care Regulation of the Health Service Control of Patient Information Regulations 2002'. Only anonymised data will be included in the results.

Patients who are diagnosed with diabetes will have their results sent to the GP. Appropriate treatment will be instituted if necessary prior to hospital discharge. Patients will receive a copy of the discharge summary and information about diagnosis, blood results and treatments.

Reporting and dissemination

Results of this study will be published in peer-reviewed scientific journals, aiming for those of high impact factor. All data will be anonymised before publication. In addition, results will also be presented at relevant medical conferences, both nationally and internationally. Furthermore, a plain English summary of results will be provided for the Trust and the NIHR websites. Media engagement will be managed via the Trust press office. The plain English summary and overall findings will also be disseminated via the Trust's social media outlets.