Durban Diabetes Study:A Study of the Epidemiology of Diabetes Mellitus in Urban South Africans of African Descent (DDS)

STUDY OBJECTIVES

• To estimate the prevalence of diabetes mellitus and associated cardiometabolic and infectious risk factors in this population
• To provide aetiological insights into the variation in cardiometabolic and infectious risk factors in adults using both population genetic and epidemiological approaches
• To compare diagnostic tests for diabetes in this population
• To create a unique framework for building a large scale cross sectional study in an African population to examine a wide range of health indices-and lay the groundwork for additional long-term studies
• to inform health policy and public health programmes aimed at addressing the rise in non-communicable diseases (NCDs) in South Africa, which may also shape public health strategies in other African countries

STUDY DESIGN

The DDS involves a number of stages:

Stage 1: Geographical sampling frame The eThekwini Municipality (city of Durban) is selected as the initial sampling frame and broken down into nine planning unit clusters (PUCs), representing nine townships (Umlazi, Inanda, KwaMashu, Ntuzuma, Mpumalanga Complex, Cato Manor, Clermont, Lamontville and Chesterville) totalling a population of approximately 1,378,750 individuals. Streets are randomly selected in each PUC proportional to the PUC size and the proportion of formal and informal housing in each PUC. Households are selected by systematic cluster sampling of homes located on the randomly selected streets.

Stage 2: Community awareness Once local ethics approval is obtained from the BioMedical Research Ethics Committee at the University of KwaZulu-Natal (UKZN) and the support of the KwaZulu-Natal Provincial Department of Health is obtained, the local health authorities and community leaders from each of the nine PUCs, will be contacted for approval.

Stage 3: Participant enrolment Once the selected PUC areas have been sensitised to the study (end of Stage 2), they are visited again by the DDS team. During this stage the DDS team visits households in the selected neighbourhoods and all family members who are of African descent, not pregnant and aged 18 years or older are invited to participate in the study. Consent is obtained from residents agreeing to participate. All residents are invited to attend a central venue (i.e. local community centre/school) the following week where the DDS team collects health and lifestyle questionnaire information, performs simple biophysical measures and collects blood and urine samples. While recruiting participants in the community, the DDS team uses geographical information system (GIS) mapping techniques to gather accurate geographical and boundary data on the selected PUC areas.

Stage 4: Data and sample collection, testing and data analysis Data and sample collection are undertaken on specific mornings at a central venue. This is important as participants are required to fast overnight for an oral glucose tolerance test (OGTT). Questionnaire and biophysical data from the study are collected, stored electronically, and analysed. Blood and urine samples are transported to the University of KwaZulu-Natal in Durban for storage. Laboratory tests are done at a private laboratory. Some samples are sent to specialist centres in the UK, Europe, the USA and other countries for genomic analyses.

Stage 5: Feedback of results to study participants Starting eight weeks after sample collection, participant results will be available at the DDS feedback office. All OGTT and lipid results will be posted back to participants, unless they choose not to receive them. To receive all other results (e.g. HIV, hepatitis C virus [HCV], liver function tests) participants must make an appointment at the DDS feedback office to collect their results. This is optional, and participants who do not wish to collect their results can choose not to contact the DDS feedback office. If appropriate, participants collecting results that are abnormal or potentially harmful will be advised to visit their local clinic or hospital for further investigation and management as part of their standard clinical care.

DATA AND SAMPLE COLLECTION A. Collection of samples

1. Blood sample overview Only fasted participants are eligible for complete blood sample collection. Non-fasted participants are only eligible for the collection of the 10 ml plain serum and the four ethylenediaminetetraacetic acid (EDTA) blood samples. Aseptic venepuncture is performed by a registered research nurse via a closed evacuated blood collection system, using standardised procedures to minimise the risk of infection..
2. Blood samples required Eleven blood tubes are collected from each fasted participant: a 10 ml plain serum tube, two 10 ml EDTA whole blood tubes, two 2 ml EDTA whole blood tubes and, for the OGTT, three 2 ml sodium fluoride (NaF) tubes and three 2 ml plain serum blood tubes. Thus, a maximum of 46 ml venous blood (approximately three tablespoons of blood) is taken from each participant. For non-fasted participants, only a 10 ml plain serum sample and the four EDTA whole blood samples are collected. Since linked anonymous HIV testing is undertaken as part of this study, the study nurse is trained in HIV counselling and follows relevant national guidelines. For fasted participants, the 75 g OGTT will be performed as recommended by WHO. Categories of glucose tolerance will be according to the WHO 1998 Classification. The 10 ml plain serum tube is used to test blood for infectious diseases and cardiometabolic biomarkers. The first 10 ml EDTA whole blood tube is shipped to the UK for genetic analysis. The second 10 ml EDTA whole blood tube undergoes plasma extraction for HCV viral load testing and RNA extraction. The two 2 ml EDTA whole blood tubes are used for HbA1c testing and full blood count, respectively. Linked anonymous HIV testing As part of this study, consenting participants will be tested for HIV via linked anonymous testing. This will be done in accordance with national guidelines.
3. Urine sample collection A spot (untimed) urine sample will be taken for measuring albumin-to-creatinine ratio and urinary sodium and potassium.
4. Collection of questionnaire data Once appropriate consent is obtained and blood sample collection has begun, participants move to the questionnaire station. The questionnaire used in this study is an electronic questionnaire (EQ) and is an adaptation of the standardised and validated WHO STEPwise approach to Surveillance (STEPS) tool designed for the collection of chronic disease risk factors, including socio-demographic indices, lifestyle, infectious disease risk factors and cardiometabolic history.
5. Collection of biophysical measurements

Study participants move to the biophysical measurements station which are done using calibrated and validated equipment. The following measures are taken using standardised protocols:

• Height and weight (used to calculate body mass index)
• Waist circumference and hip circumference (used to calculate waist-hip ratio)
• Mid upper-arm circumference
• Blood pressure and heart rate

BLOOD SAMPLE ANALYSES

All blood samples are assayed for infectious and cardiometabolic biomarkers at a private laboratory in Durban. Confirmatory HIV tests will be done. Below is a list of laboratory methods used for serological and biochemical tests:

• Plasma glucose: Enzymatic method, based on hexokinase and glucose-6-phosphate dehydrogenase enzymes
• Serum creatinine: Based on the reaction of picric acid with creatinine in an alkaline medium
• Serum lipids:
• Cholesterol: Enzymatic method, using cholesterol esterase and cholesterol oxidase conversion, followed by a Trinder endpoint
• Triglycerides: Based on the Fossati three step enzymatic reaction with a Trinder endpoint. The single reagent procedure quantifies the total triglycerides including the mono- and di-glycerides and the free glycerol fractions
• HDL: Enzymatic reaction based on the procedures developed by Izawa, Okada and Matsui
• LDL: Calculated using the Friedewald formula or enzymatic method
• Lipoprotein (a): Turbidometric (Beckmann)
• HbA1c: High performance liquid chromatography. Instrument: BioRad Variant II
• HIV: 4th generation ELISA (Abbott Architect)
• Hepatitis B and C: ELISA

DNA and RNA extraction will take place at the University of KwaZulu-Natal, the University of Cambridge, the Wellcome Trust Sanger Institute, or an authorised third party facility in the UK. The study coordinator is responsible for shipping samples to the UK for DNA and RNA extraction. Purified DNA and RNA is then sent to the Wellcome Trust Sanger Institute, UK, for genetic analysis.

GENETIC ANALYSIS

1. Overview Given the need for large scale automated pipelines and next-generation Illumina sequencing and genotyping technology, all genotyping and sequencing is done at the Wellcome Trust Sanger Institute, UK (http://www.sanger.ac.uk/). The Wellcome Trust Sanger Institute is an international genome centre with extensive experience and expertise in handling DNA and RNA samples from global genetic research initiatives, including blood and samples from several countries in SSA. A material transfer agreement has been devised for the shipment of biological samples to the UK.
2. Extraction and quantification of DNA and RNA DNA and RNA are extracted at the University of KwaZulu-Natal, the University of Cambridge, the Wellcome Trust Sanger Institute or by an authorised third party service provider in the UK, e.g. GenProbe, Manchester. Purified DNA or RNA is then sent to the Wellcome Trust Sanger Institute. All extracted nucleic acid is quantified by either gel electrophoresis or picogreen (for DNA samples) or ribogreen (for RNA samples) and quantified relative to standards of known concentration. All nucleic acid samples are stored at the Wellcome Trust Sanger Institute in secure -80 ˚C freezers.

3 Genotyping and sequencing Current genome-wide human diversity arrays do not sufficiently capture the high levels of genetic diversity and population structure found in African populations, limiting their utility. Therefore, we will use next-generation sequencing, genotyping arrays and imputation to help determine the association between human genetic variation in the study population and quantitative risk factors and diseases and viral diversity. All genotyping and sequencing is done at the Wellcome Trust Sanger Institute. In addition, because of the massive storage and curation requirements for the genomic data from next-generation sequencing method, all raw genomic data is stored at Wellcome Trust Sanger Institute, UK.

4 Data transfer, storage and management All raw genomic data will be held at the Wellcome Trust Sanger Institute, UK and the European Genome-phenome Archive (EGA) at the European Molecular Biology Laboratory-European Bioinformatics Institute (EBI), UK.