This research has the following specific objectives:

1. To elucidate how within-person variation in lifestyle factors affect fluctuations in blood glucose concentrations in individuals at high risk of diabetes.This study will elucidate how variation in food intakes, physical activity, and psychological stress affect variation in blood glucose concentrations throughout the day. These results can identify potential targets for interventions to reduce excessive fluctuations in blood glucose concentrations.
2. To describe to what extent the response of individuals to a standardized meal tolerance test can predict real-life variation in blood glucose concentrations. This study will evaluate how much variation in glucose concentrations under real-life conditions can be explained by an individual's response to a standardized mixed meal tolerance test. This will provide insight into the relative importance of variation in dietary and other lifestyle behaviours on an individual's predisposition to higher blood glucose responses.
3. To elucidate the role of oral processing behaviour and saliva properties on blood glucose concentrations. This study will elucidate whether variation in oral processing behaviours (e.g. number of chews taken, oro-sensory exposure time) and saliva properties (a-amylase activity, flow rate) predicts variation in blood glucose concentrations across individuals.
4. To assess whether research collecting multiple repeated measures of food intake, activity, and stress is feasible in large-scale epidemiological studies.This study will provide important insights into the feasibility over the long-run to collect multiple repeated data points on lifestyle behaviours through mobile phone applications and 24-hour glucose and physical activity monitoring in large scale studies in the Singapore population.

This is an observational study in free-living participants over up to 14 consecutive days. The rationale is that this period will provide sufficient intra-individual data for exploring behavioural and psychological determinants of blood glucose fluctuation. Multiple repeated data points on lifestyle behaviours will be collected through mobile phone applications, and 24-hour glucose concentrations and physical activity will be monitored objectively over the study period.

STUDY DESIGN:

As this is a pilot study, 36 participants will be recruited with anticipated 15% dropout resulting in 30 remaining participants. The recruitment target is determined based on what is financially feasible and sufficient to meet the study aims.Hierarchical clustering will be used to identify subjects with similar continuous blood glucose trajectories over 1 week by classifying them into groups. Followed by assessing whether the group labels from the cluster analysis are associated with the variation in dietary intakes, physical activity and psychological stress using Chi-square test or Fisher Exact test where appropriate. To quantify the effect of specific events (e.g., dietary intake, physical activity and stress affect) on continuous blood glucose trajectory within the two hour window after an event, linear models with generalized estimating equations will be used, where the characteristics of the event, socio-demographic variables of the subject and time since event are predictors in the model while accounting for the correlations from repeated continuous blood glucose in the same individual from multiple events. This study will also assess whether the characteristics of the event and sociodemographic variables are potential effect modifiers on the association between time and continuous blood glucose. To allow for flexibility in modelling time and continuous blood glucose, smoothing spline will also be used.

RECRUITMENT:

Participants of the Singapore Population Health Studies cohort who have agreed to be contacted for other studies, will be contacted by telephone and invited to participate in the present study (Telephone script and screening questionnaire; and this will be done in compliance with Personal Data Protection Act (PDPA) 2012. Eligible participants will be invited to attend NUH Investigational Medicine Unit.

Participants will be instructed to refrain from strenuous exercise for 48 hours before the procedure, and to undergo overnight fasting (10-12 hours) before the start of the procedure. No alcohol may be consumed for 24 hours before the study procedure. Participants will be asked to refrain from taking Vitamin C, multivitamins, and aspirin during the eleven day study period.

TEST SESSION 1 (DAY 1):

The first study visit will take approximately 3 hours. At the clinic visit, informed consent will be obtained in a private space, a mobile phone application will be downloaded onto participants phones to ensure compatibility, and blood pressure, height, weight, and waist circumference measured. A continuous glucose monitor (Freestyle Libre; Abbott Diabetes Care, Witney, Oxon, UK) will be fitted on the upper part of their non-dominant arm. If the participant has a lot of arm hair, the area will be shaved. A waterproof plaster will be placed over the device.

A mixed meal tolerance test (MMTT) will be conducted. The duration of MMTT will last for about 2 out of the 3 hours allocated for test session 1. Participants will be asked to sit on a bed and a cannula will be fitted in one of the antecubital veins in the pit of the elbow on the participants arm with a 3-way stopper. A blood sample will be taken via the cannula. Participants will then be asked to ingest a standardized meal (fried rice). While eating the standardised meal, participants will be video-recorded using a laptop webcam at face-level. Post-meal satiety responses using a visual analogue scale will be collected on a laptop every 15 minutes. Blood samples will be collected via an antecubital catheter before eating (11ml) and then at 5 min, 10 min, 15 min, 30 min, 45 min, 60 min, 90 min and 120 min (8ml each) after completion of the meal. The total blood volume taken will be 65-75ml. Blood will be sent immediately to NUH Referral Lab for analyses.

Questionnaires will be administered on demographics and lifestyle, including usual diet (FFQ with frequency rating from "never or rarely" to "6+ a day") and physical activity (GPAQ Questionnaire-Yes/No and input of actual activity duration), eating behaviours (TFEQ Questionnaire on 4-point scale from "definitely true" to "definitely false"), alcohol consumption (Yes/No), smoking (Yes/No), stress (K10 on likert scale from "All of the time" to "None of the time", where higher score indicate greater stress), and sleep (Pittsburgh sleep quality index-input of sleep time and duration and questions on 4-point scale from "Not during the past month" to "Three or more times a week"). Glucose and level of daily physical activity will be monitored using a continuous glucose monitor sensor and a wrist worn accelerometer. Food intake, physical activity and sleep time will be recorded by participants through an ecological momentary assessment (EMA) survey pushed out via mobile phone application.

Oral processing behaviours and saliva properties of test meal bolus: Bolus particle size (oral processing behaviour measure) and saliva uptake (saliva properties) will be assessed by having participants will be asked to chew a fixed mouthful of the test meal until ready to swallow, then expectorating it in a cup. The duration of oral processing behaviours and saliva properties measures will last for about 10 minutes out of the 3 hours allocated for test session 1.

Once the MMTT is complete, vital signs (blood pressure and pulse rate) will be measured. The cannula will be removed, and pressure applied locally on the site of bleeding. Researchers will ensure the participant is feeling well and is not giddy before continuing with the protocol.

DAY 2 to 10 (OFF-SITE):

Participants will be instructed to wear the continuous glucose monitor sensor continuously for 10 days. Participants will be asked to scan the sensor with a reader device at least every 6 hours during waking hours, and immediately before and after sleeping. The reader display will be masked with a sticker so participants will not be able to see their glucose readings. Participants will also be fitted with a wrist-worn accelerometer (Actigraph - Model wGT3x+ BT), a small tamper-proof electronic device that can be worn on a non-removable strap on the wrist of the non-dominant hand. Although the accelerometer may be worn during showering and swimming, participants will be given one additional strap in case removal is required.Participants will be guided on use of the ecological momentary assessment (EMA) mobile phone application during the baseline visit, and be asked to record within the app, their food intake, physical activity, and sleep times. A trained interviewer will demonstrate how the use the app and the participant will have a chance to ask questions. Participants will receive a reminder via the app, and will be asked to start recording activities prospectively from day 4 of the study.

From day 4 onwards, participants will receive one-minute time-triggered surveys via the app up to six times per day in which participants will be asked about hunger levels, stress, and psychological states. A phone number and email address will be provided to participants for them to contact the study team if participants face problems during the study. Participants will also be issued with a paper handout with reminders about using the app and sensor. Participants with phone models that are discovered to be incompatible with the app during study visit 1 or later in the study period will be issued with a phone handset on which participants can complete lifestyle tracking and EMA surveys. At the end of day 4, participants will be telephoned and asked if participants experienced any technical issues. For issues that cannot be resolved over the telephone, participants will be asked to meet researchers at Saw Swee Hock School of Public Health, or a place of their convenience. If the issue relates to the glucose sensor, if necessary participants will be asked to restart the monitoring period, and will be fitted with a new glucose sensor.

TEST SESSION 2 (DAY 11):

The final study visit will occur on day 11 and will take approximately 2 hours. Participants' body composition will be measured using a BOD POD Gold Standard Body Composition Tracking System (COSMED, Concord, CA, USA) is an air displacement plethysmograph which uses whole-body densitometry to determine body composition (fat and fat-free mass). Participants will be asked to change into swimwear and sit inside the machine for 5 minutes. Participants will be provided with results on their body composition.

Oral processing behaviours and saliva properties: Saliva measures will be taken from the participant. Resting saliva will be taken by asking the participant to expectorate into a tube (approximately 5ml) while stimulated whole saliva samples will be collected by asking the participant to chew on a clean, non-toxic inert Parafilm square and expectorating over a period of 5 minutes (approximately 5-10ml). The duration for saliva measures in about 30 minutes. Mouth volume will be measured by asking the participants to take in the maximum volume of filtered water participants can hold and then expectorating it into a volumetric cylinder. Participants will also be asked to chew on 3 test foods (white rice, rice cake, raw carrot) using a fixed chew protocol (i.e. 15 chews, 30 chews and chew until point of first swallow) and then expectorating the bolus to assess their oral processing behavior based on bolus properties including saliva uptake and particle size distribution. The duration is approximately 1 hour.

Participants will return the accelerometer, glucose monitor reader, and study phone (if issued). A research assistant will remove the glucose sensor from the participant's arm.

Before leaving the unit, participants will be asked to take part in a 15 minute open-ended interview about taking part in this research, focusing on their experiences using the glucose monitor and research app. Participants will receive a letter reporting the following measurements and results collected during the study: blood pressure, height, weight, BMI, HbA1C concentration, and fasting glucose concentration. The report will include information on interpreting the results, and occasions when it is advised to see a doctor

Inclusion Criteria:

• Aged 21-60 years.
• One of the following : pre-diabetes (as determined by previous HbA1C test within the range 42-47mmol/mol), family history of diabetes (mother, father, brother, sister, son, or daughter), or BMI 25-30 kg/m2.
• Ownership of a compatible smartphone.
• Access to a data plan.
• Able to provide informed consent.
• Healthy dentition with ability to bite, chew and swallow normally

Exclusion Criteria:

• Known sensitivity to medical-grade adhesives
• Bleeding disorders
• Allergies to any ingredients of the standardised meal or test foods
• Pregnant or lactating women
• Participants who did not agree to be contacted for future research
• Participants with conditions: diabetes mellitus, hypertension or thyroid disease, myocardial infarction in the past year, cancer history, or psychosocial impairment.

Participants taking aspirin for long-term medical conditions will be excluded.

Adults in Singapore with higher risk of type 2 diabetes