Lifestyle intervention to prevent postpartum type 2 diabetes in Asian women with recent gestational diabetes and normal post-pregnancy glucose tolerance: study protocol for a randomized controlled trial with a targeted postpartum lifestyle programme

Abstract

Background

The primary aim of the Integrated Hyperglycaemia Incentivized Postnatal Surveillance (IHIPS) study is to reduce the incidence of type 2 diabetes mellitus (T2DM) among women with a history of gestational diabetes mellitus (GDM) in their index pregnancy who demonstrate normal glucose tolerance in the early postpartum period. This is being evaluated through a single-centre randomized controlled trial (RCT) conducted at KK Women’s and Children’s Hospital (KKH) in Singapore.

Methods

The study aims to recruit 200 women between 8 and 12 weeks postpartum. The primary objective is to assess the cumulative incidence of T2DM over a 36-month follow-up period. Secondary objectives include evaluating the incidence of dysglycaemia (prediabetes and T2DM) at 12-, 24-, and 36-month follow-ups, as well as monitoring longitudinal changes in cardiometabolic risk factors, physical activity, and dietary intake from 6 to 36 months. Cost-effectiveness will also be assessed. Participants in the intervention group will receive a multi-component programme that includes personalized goal-setting workshops supported by wearable technologies, such as continuous glucose monitors and Fitbit trackers. The control group will receive standard postnatal care. All participants will provide blood samples and undergo anthropometric and body composition measurements. Participants will also complete self-administered questionnaires evaluating dietary intake, physical activity, health-related quality of life, and their experience with the clinical trial.

Discussion

The IHIPS study is expected to provide insights to guide the development of early postpartum, hospital-based lifestyle interventions, with the potential to improve long-term health outcomes for mothers and their infants.

Trial registration

ClinicalTrials.gov NCT05081037. Registered on 06 September 2021, https://clinicaltrials.gov/study/NCT05081037.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13063-026-09516-7.

Introduction

Background and rationale {9a}

The Singapore Ministry of Health recommends universal GDM screening at 24–28 weeks for all pregnant women without pre-existing diabetes [1]. GDM affects 20–25% of local pregnancies [2], far exceeding the 2019 global prevalence of 13.4% [3], and is associated with an increased risk of developing T2DM [4]. Asian women, including those in Singapore, face disproportionately higher risks of GDM [5] and subsequent T2DM [6], with 44% of women diagnosed with GDM developing postpartum dysglycaemia, or T2DM within 4–7 years [5].

Women with a history of GDM are an ideal population for diabetes prevention, as glucose management during pregnancy does not eliminate long-term cardiometabolic risk [6]. Lifestyle interventions, even years after pregnancy, can reduce T2DM risk [7, 8], but many women progress to diabetes within the first few years postpartum [5]—when risk is highest [9, 10]—and are underrepresented in existing studies [10]. Preventive efforts should therefore begin early, ideally within 6 months postpartum, to optimize health behaviours and maximize reductions in diabetes incidence [11–13].

Current postpartum care for women with prior GDM focuses mainly on those with impaired fasting glucose (IFG) or impaired glucose tolerance (IGT), considered at higher cardiovascular risk [11, 14], while overlooking women with normal glucose tolerance (NGT) who also face elevated long-term metabolic risk. Although WHO recommends a 75-g OGTT to reassess glycaemic status postpartum and repeat testing every 1–3 years [15–17], adherence is poor due to limited clinician emphasis, insufficient patient education, and fragmented follow-up pathways [18, 19]. Postpartum care pathways for these women are frequently fragmented and inconsistent, highlighting a critical gap in efforts to mitigate long-term metabolic risk. Postpartum IGT typically progresses to T2DM at 5–10% per year [20, 21], but 17% of women with NGT at 3 months postpartum develop prediabetes or diabetes within a year [22], with evidence also showing increased long-term cardiovascular risk in this group [23].

Lifestyle interventions have been shown to reduce T2DM risk in women with a history of gestational diabetes (GDM), often using behavioural strategies such as counselling, motivational interviewing, and wearable fitness trackers [7, 24, 25]. However, evidence is limited in high-income Asian populations, particularly for women with prior GDM who have normal glucose tolerance postpartum. [12]. No studies have yet incorporated wearable technologies like continuous glucose monitoring (CGM) into postpartum interventions, despite their proven benefits for glycaemic control during pregnancy [26].

The IHIPS study is a novel, multi-component lifestyle intervention targeting women with normal glucose tolerance in the early postpartum period following gestational diabetes mellitus (GDM). It focuses on Singapore’s three major ethnic groups—Chinese, Malay, and Indian—who are broadly representative of Southeast Asian populations at elevated risk of GDM and T2DM [27]. The intervention combines personalized care from healthcare professionals with wearable technologies, including continuous glucose monitoring (CGM) and fitness trackers, to support early postpartum lifestyle modification. We hypothesize that real-time feedback and reinforcement will enhance self-awareness, motivation, and adoption of healthier behaviours in diet, physical activity, and weight management, ultimately reducing T2DM risk.

Explanation for the choice of comparator {9b}

Women randomized to the Control Group will receive standard postnatal care, supplemented with the “Win Against Diabetes” booklet by SingHealth, which provides evidence-based guidance on lifestyle behaviours to prevent and manage T2DM. This comparator was chosen to reflect current usual care in Singapore and to ensure that all participants receive at least minimal education on diabetes prevention, allowing the study to evaluate the additional effect of the multi-component intervention beyond standard care.

Objectives {10}

The primary objective is to compare the cumulative incidence of T2DM at 36 months between the intervention and standard care groups, starting from the baseline visit at 8–12 weeks postpartum. Secondary objectives, assessed at 6, 12, 24, and 36 months, include comparing the groups for incidence of dysglycaemia, changes in cardiometabolic and anthropometric measures, diet quality, physical activity, and total energy intake. The study will also evaluate the cost-effectiveness of the intervention.

Methods: patient and public involvement and trial design

Patient and public involvement {11}

This study did not involve patients or members of the public in the design, conduct, or reporting of the trial. The intervention and study procedures were developed by the research team based on existing evidence and clinical guidelines, and all study materials and protocols were reviewed and approved by the institutional ethics committee.

Trial design {12}

The IHIPS study is a single-centre randomized controlled trial (parallel two-arm, single-centre, nonblinded) evaluating a multi-component postpartum lifestyle intervention for women with NGT post-delivery following a recent history of GDM. Participants are randomized in a 1:1 allocation ratio to either the intervention or control group. The study design includes a 6-month intervention programme and a follow-up period of up to 36 months from the baseline visit (8–12 weeks postpartum) (Fig. 1). The flow diagram illustrates the enrolment process of the study (Fig. 2).

Fig. 1.

Study protocol for a randomized controlled trial: Integrated Hyperglycaemia Incentivized Postnatal Surveillance Study. Abbreviations: OGTT, oral glucose tolerance test; HDL, high-density lipoprotein; LDL, low-density lipoprotein; GDM, gestational diabetes mellitus. * All questionnaires were self-administered, except for the 24-h dietary recall, which was conducted through interviewer administration by the research coordinators

Fig. 2.

A flow diagram showing the flow of participants through the I-HIPS intervention in women with normal glucose tolerance postnatal at 6-10 weeks after GDM in index pregnancy randomized controlled trial. GDM: gestational diabetes mellitus; OGTT: oral glucose tolerance test: IADSPG: The International Association of Diabetes and Pregnancy Study Groups; NGT: normal glucose tolerance; T2DM: type 2 diabetes mellitus; CGM: continuous glucose monitor; BMI: body mass index

Methods: participants, interventions, and outcomes

Trial setting {13}

Participants of the study will be recruited from a single centre, which is the KK Women’s and Children’s Hospital, Singapore.

Characteristics of the people who are needed for the trial are shown below.

Characteristic	The people we would expect to see included
Age	Mean (33.9) SD (4.2) years
Sex	Female
Gender	Woman
Race, ethnicity, and ancestry	Chinese
	Malay
	Indian
	Others of South East Asian descent
Socioeconomic status	68% have at least university degree or higher
	24% have a post-secondary education or diploma
	8% have at least a primary school or secondary school level of education
Geographic location	Singapore
Other characteristics relevant to the trial	NA

Eligibility criteria for participants {14a}

To be eligible for the study, participants must be women diagnosed with GDM during their index pregnancy, based on a three-timepoint 75-g OGTT using the International Association of Diabetes and Pregnancy Study Groups (IADPSG) criteria—with a fasting plasma glucose ≥ 5.1 mmol/L, 1-h ≥ 10.0 mmol/L, or 2-h ≥ 8.5 mmol/L [3]. At 6 to 10 weeks postpartum, they must demonstrate normal glucose tolerance (NGT) on a two-timepoint 75-g OGTT, defined as a fasting glucose < 6.1 mmol/L and 2-h glucose < 7.8 mmol/L. Additional inclusion criteria include having a body mass index (BMI) between 20 and 40 kg/m² and being physically fit to engage in light to moderate physical activity. Participants are excluded if they are unable to speak or read English, as study assessments are only available in English. Other exclusion criteria include not self-identifying as Chinese, Malay, or Indian ethnicity; having serious skin conditions that prevent the use of a continuous glucose monitoring (CGM) sensor; having any serious chronic medical condition (e.g. cancer, unstable heart disease, severe kidney disease, uncontrolled mental illness, dementia, or cognitive impairment); taking medications known to affect glucose metabolism; not owning a compatible smartphone for the study apps; or participating in concurrent clinical trial, lifestyle intervention, or fitness/wellness programme.

Eligibility criteria for sites and those delivering interventions {14b}

All interventions will be delivered by trained staff of KK Women’s and Children’s Hospital, including dietitians and clinical exercise physiologists. Staff delivering the intervention receive standardized training by the study Principal Investigator to ensure consistent and protocol-adherent delivery.

Who will take informed consent? {32a}

Screening procedure

Before informed consent is taken for eligible participants, the study employs a three-step screening process:

1. Initial contact: Women diagnosed with GDM during their index pregnancy are approached by the study team either shortly after delivery, or following direct referrals and recruitment via posters. Those who express interest provide verbal consent to be contacted after their routine postpartum OGTT at 6–10 weeks postpartum.

2. Postpartum OGTT screening: As part of standard postnatal care, women with prior GDM undergo a 75-g OGTT. Women who completed the OGTT between 6 and 10 weeks postpartum and are found to have NGT—defined as fasting plasma glucose (FPG) < 6.0 mmol/L (108 mg/dL) and 2-h glucose < 7.8 mmol/L (140.4 mg/dL)—are invited for an in-clinic screening visit approximately 2 weeks later, between 8 and 12 weeks postpartum.

3. Final eligibility screening: During the in-clinic visit, participants undergo anthropometric assessments by trained research coordinators. Height (SECA-213) and weight (Tanita SC-240) are measured to calculate body mass index (BMI), which must fall within the eligible range of 20–40 kg/m². Participants are also asked to self-report their ability to engage in light to moderate physical activity over the study duration. Those meeting all eligibility criteria are provided with detailed written study information, and written informed consent is obtained by trained research coordinators.

Additional consent provisions for collection and use of participant data and biological specimens {32b}

Consent for the collection and use of participant data and biological specimens is covered under a single written informed consent document; no additional consent is required.

Intervention and comparator

Intervention and comparator description {15a}

Study groups

Control group

Women randomized to the Control Group will receive standard postnatal care along with the “Win Against Diabetes” booklet by SingHealth, which provides information on lifestyle behaviours to manage and prevent T2DM.

Intervention group

Women in the Intervention Group will receive the same standard postnatal care and booklet as the Control Group, plus additional intervention components as described below:

1. Three personalized goal-setting workshops in the first 6 months of the RCT

2. Wearable devices: A CGM sensor was inserted on the back of either the right or left upper arm, without any over-bandage, for up to 14 days. Fitbit trackers were encouraged to be worn at all times when possible.

3. Delivery of personalized health nudges via SMS

Intervention

Review of baseline lifestyle behaviours

Participants are assessed by dietitians and Clinical Exercise Physiologists at KKH, who evaluate their dietary and physical activity behaviours using the 6P Lifestyle Behaviour Tool (Portion, Proportion, Pleasure, Phase, Physicality, Psychology) [12]. They will also receive education on the use of CGM sensors and Fitbit trackers to support their health goals. The personalized health coaching intervention spans 6 months and consists of three sessions; each spaced approximately 2 months apart. Individualized goals are set and continuously adjusted throughout the intervention to align with each participant’s evolving lifestyle needs during the postpartum period. All dietitians and Clinical Exercise Physiologists were trained in Motivational Interviewing techniques [28] to support and promote behavioural change among participants before the study recruitment began.

Personalized health nudges via SMS

Personalized health nudges are delivered to support adherence to one or more of the 6P domains—Portion, Proportion, Pleasure, Phase, Physicality, and Psychology—based on each participant’s individualized goals. Participants also receive guidance on how to use their wearable devices (as described in Sect. 3.5.1) to help achieve these goals during the intervention period and sustain them beyond its completion.

Wearable devices

A CGM wearable device (FreeStyle Libre, Abbott Laboratories Ltd.) is used to monitor participants’ subcutaneous glucose variability over a 14-day period. Participants are able to view their glucose readings in real time via a dedicated reader or the FreeStyle LibreLink smartphone application, which not only enables seamless data sharing with healthcare providers for monitoring and feedback during the personalized health coaching intervention, but also serves as a powerful motivational tool. By making the impact of dietary choices, physical activity, and lifestyle habits immediately visible, the continuous glucose monitor (CGM) fosters greater awareness and accountability, encouraging users to adopt and sustain healthier behaviours [29]. In addition, each participant will receive a commercially available Fitbit Charge 4 or 6 (Fitbit Inc., San Francisco, CA, USA) to track daily step count and the intensity of physical activity, with the emphasis of reducing sedentary behaviour and encouraging moderate-to-vigorous intensity activity. The Fitbit is a wrist-worn fitness tracker equipped with a large OLED display and a 3-axis accelerometer, which uses a proprietary algorithm to detect and record steps based on movement data. Participants are instructed to wear the Fitbit continuously, removing it only for bathing, and to synchronize it with the Fitbit app daily to ensure accurate data collection.

Criteria for discontinuing or modifying allocated intervention/comparator {15b}

Participants may be withdrawn from the study if they fail to comply with the study protocol or if the Principal Investigator determines that continued participation is no longer appropriate.

Strategies to improve adherence to intervention/comparator {15c}

Independent audits of the trial will be conducted by the Office of Research Integrity and Compliance (ORIC) to ensure adherence to the study protocol and regulatory standards.

Concomitant care permitted or prohibited during the trial {15d}

Concomitant care: All participants will continue to receive standard postnatal care throughout the trial. Prohibited interventions: Participation in any other lifestyle intervention studies during the trial period is not allowed. No other treatments or standard care interventions are restricted.

Ancillary and post-trial care {34}

Ancillary and post-trial care: No specific post-trial care is planned, as this is a low-risk study. Compensation for harm: The Principal Investigator will be responsible for managing any harm that may occur as a result of participation in the trial.

Outcomes {16}

Data collection for all study participants from the intervention and control groups

Figure 1 provides an overview of the data collection procedures. The instruments and data collection forms used in this study are available from the authors upon reasonable request.

Blood collection

A fasting blood test is performed at 8–12 weeks postpartum at the baseline visit for all participants and repeated at subsequent timepoints throughout the study. This test includes HbA1c and a lipid profile, comprising total cholesterol, triglycerides, HDL cholesterol, and LDL cholesterol. HbA1c offers a longer-term view of glycaemic control and may reveal early dysglycaemia not detected by OGTT alone, while the lipid profile helps assess cardiometabolic risk, which is often elevated in this population [30].

At all follow-up timepoints—6, 12, 24, and 36 months after the baseline visit—participants undergo a 2-h 75-g oral glucose tolerance test (OGTT), with blood samples collected at fasting (FG), 1 h (1hPG), and 2 h (2hPG) post-glucose ingestion. The two-timepoint OGTT (fasting and 2-h values) is used to formally diagnose dysglycaemia, including prediabetes and type 2 diabetes mellitus (T2DM), based on standard diagnostic thresholds recommended by the 2006 WHO criteria [31].

T2DM is defined as a fasting plasma glucose ≥ 7.0 mmol/L (126 mg/dL) or 2-h plasma glucose ≥ 11.1 mmol/L (200 mg/dL). Impaired glucose tolerance (IGT) is diagnosed when fasting glucose is < 7.0 mmol/L (126 mg/dL) with a 2-h plasma glucose between 7.8 and < 11.1 mmol/L (140–199 mg/dL), while impaired fasting glucose (IFG) is defined as fasting glucose between 6.1 and 6.9 mmol/L (110–125 mg/dL) and a 2-h plasma glucose < 7.8 mmol/L (< 140 mg/dL). Longitudinal collection of this data enables the detection of subclinical metabolic changes over time and the identification of early predictors of glycaemic deterioration in this high-risk cohort.

All biochemical analyses are conducted using an automated biochemical analyzer (Abbott Alinity) at KK Women’s and Children’s Hospital Laboratories.

Clinical examination

Blood pressure (BP), height (centimetres), weight (kilograms), body fat mass (kg), total body water (kg), and waist circumference (centimetres) are measured at the baseline visit and at each follow-up visit throughout the study. BP is measured twice using an automated blood pressure monitor (Dinamap Carescape V100). A third measurement is obtained if the difference between the first and second measures exceeds 10 mmHg, with the average taken for analysis. Height is recorded to the nearest 0.5 cm using a calibrated wall-mounted stadiometer (SECA 2130), while weight, body fat mass, and total body water are measured using a calibrated portable bioimpedance analysis instrument (Tanita SC-240). Height and weight are each measured twice, to the nearest 0.1 cm and 0.1 kg, respectively. A third measurement is taken if the difference between the first two readings exceeds 1.0 cm for height or 0.2 kg for weight. The average of the measures is reported. BMI is calculated as weight in kilograms divided by height in metres squared (kg/m²). Waist circumference is measured in centimetres (cm), with a measuring tape placed horizontally at the midpoint between the iliac crest and the lowest lateral rib, directly on the skin. Two measurements are taken to the nearest 0.1 cm, and a third measurement will be taken if the difference between the first two readings exceeds 1.0 cm. To ensure data quality, all research coordinators underwent training in standardized protocols for anthropometric and body composition measurements.

Anthropometric and body composition measures, including weight, waist circumference, and fat distribution, provide insight into adiposity patterns and insulin resistance risk, particularly relevant in Asian populations [32].

Self-reported questionnaire

The questionnaires used to collect data on sociodemographic and lifestyle characteristics, family history of diabetes, breastfeeding practices, and study feedback were custom-designed for this study. This intends to elucidate the contextual and behavioural factors that may influence postpartum metabolic health and engagement with the intervention.

Dietary intake at 26–28 weeks’ gestation will be assessed using a single 24-h dietary recall. This recall is conducted by trained clinical research staff on either a weekday or weekend, employing a five-step, multiple-pass interviewing method. Standardized household measuring utensils and food portion photographs are used to help participants accurately quantify their intake from the previous day [33].

Nutrient intake will be estimated using Dietplan (Forestfield Software, UK), a nutrient analysis software program based on a food composition database containing local foods, with slight modifications to correct inaccuracies. For local dishes not found in the database, nutrient compositions will be calculated based on ingredient-level nutrition values from generic recipes. For individual food items not included in the database, nutrient data will be sourced from food packaging labels or the U.S. Department of Agriculture nutrient database, particularly for commonly imported Western food products [34].

Physical activity is assessed using the International Physical Activity Questionnaire (IPAQ) long form, which captures a 7-day recall of activity lasting ≥ 10 min per bout. The questionnaire categorizes activity by intensity (vigorous, moderate, and walking) and duration (days per week and minutes per day) [35]. Total weekly minutes of vigorous, moderate, and walking activity will be calculated and capped at a maximum of 3 h per day or 21 h per week, following IPAQ coding guidelines [36].

Collecting dietary intake and physical activity data longitudinally will enable the identification of behavioural patterns and changes over time, providing insight into how these behaviours may be influenced by the intervention.

Health-related quality of life (HrQoL) is evaluated using the EuroQol-5 Dimensions-5 Levels (EQ-5D-5L) instrument. This tool is widely used for assessing general HrQoL, conducting cost-utility analyses of healthcare interventions, and calculating quality-adjusted life years (QALYs). It comprises two parts: a descriptive component covering five health dimensions—Mobility, Self-Care, Usual Activities, Pain/Discomfort, and Anxiety/Depression—each with three response levels (no problems, some problems, extreme problems); and a visual analogue scale (VAS), where respondents rate their overall health on the day of the survey using a 0–100 vertical scale [37].

The 6P tool—comprising Portion, Proportion, Pleasure, Phase, Physicality, and Psychology—was developed as a lifestyle intervention tool and has been validated for use in overweight and obese women [38]. It is administered via a digital platform that enables the generation of individualized feedback reports and real-time monitoring charts. The feedback report provides personalized insights into the health implications of each participant’s lifestyle patterns, and includes recommendations tailored to their specific dietary challenges. The monitoring chart allows participants to track their 6P scores over time, supporting ongoing self-monitoring and engagement. Based on the identified lifestyle issues, participants are coached by dietitians and Clinical Exercise Physiologists to help them improve their health behaviours. Coaching includes goal setting that aligns with each participant’s unique challenges and targets for lifestyle improvement. All data from the 6P database are de-identified, and extracted data are stored on an encrypted hard drive accessible only to authorized study team members, ensuring the protection of participant privacy and data security.

Pregnancy and delivery outcome data

To comprehensively assess the risk of progression to T2DM in women with a history of GDM, we are collecting detailed maternal, delivery, and neonatal data. These variables were selected based on their known, or potential associations with postpartum metabolic outcomes and long-term cardiometabolic risk.

Maternal data: parity; pre-pregnancy weight; gestational week of GDM diagnosis; results of the 75-g OGTT during pregnancy; HbA1c values; treatment of GDM during pregnancy; history of antenatal risk factors such as subfertility, uterine surgery, preterm delivery, preeclampsia in previous pregnancy; pregnancy complications such as preeclampsia (de novo BP _ 140/90 mmHg > 20 weeks with proteinuria or signs of end-organ dysfunction), eclampsia, gestational hypertension (de novo BP _ 140/90 mmHg > 20 weeks), preexisting hypertension, renal disease, cardiac disease and autoimmune disease.

Delivery data: type of delivery (spontaneous vaginal, forceps or vacuum, caesarean section during labour or planned caesarean section) and complications during delivery such as uterine rupture, manual placental removal, third-degree tears, or obstetric hysterectomy.

Neonatal data: birth weight, 1-min and 5-min Apgar score, and shoulder dystocia.

Outcomes

Primary outcomes

The primary outcome is the cumulative incidence of type 2 diabetes mellitus (T2DM) at 36 months following the baseline visit (conducted at 8–12 weeks postpartum), which will be determined by a 75 g oral glucose tolerance test (OGTT). T2DM will be diagnosed based on the WHO criteria of fasting plasma glucose ≥ 7.0 mmol/L (126 mg/dL) or 2-h plasma glucose ≥ 11.1 mmol/L (200 mg/dL).

Secondary outcomes

• To evaluate the incidence of dysglycaemia at 6, 12, 24, and 36 months, where prediabetes will be classified according to the 2006 WHO guidelines as either impaired glucose tolerance (IGT), defined by fasting plasma glucose < 7.0 mmol/L (126 mg/dL) accompanied by a 2-h plasma glucose ≥ 7.8 and < 11.1 mmol/L (140–199 mg/dL), or impaired fasting glucose (IFG), defined by fasting plasma glucose between 6.1 and 6.9 mmol/L (110–125 mg/dL) accompanied by a 2-h plasma glucose < 7.8 mmol/L (< 140 mg/dL) [31].

• To evaluate changes from baseline to 6 months post-intervention, and longitudinally across all measurement timepoints in:

  • i.
    Cardiovascular risk factors associated with metabolic syndrome, including blood pressure (BP), body mass index (BMI), HbA1c, fasting and 2-h blood glucose levels, and fasting lipid profile (triglycerides (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL)) between the control and intervention groups.
  • ii.
    Self-reported total physical activity, measured in MET-minutes/week, between the control and intervention groups.
  • iii.
    Dietary intake, including total energy intake (kilocalories) over 24 h and daily servings from key food groups, between the control and intervention groups.

Tertiary outcomes

An incremental cost-effectiveness analysis will be conducted from the health system perspective, in line with the priorities of Singapore’s Agency for Care Effectiveness [39]. This analysis will incorporate data from the study’s clinical outcomes, as well as local population data and relevant external sources. Costs will be captured using Activity-Based Costing and standardized cost collection tools previously validated [2]. A lifetime horizon will be applied, with costs and outcomes discounted at 3% annually. Long-term cost-effectiveness will be modelled using adapted UKPDS and CDC diabetes models to estimate quality-adjusted life years (QALYs) gained based on changes in HbA1c and other clinical outcomes at trial end.

Harms {17}

• Anticipated harms: Minimal risks are expected from the use of the continuous glucose monitoring (CGM) sensor and FitBit exercise tracker, including minor skin irritation or bruising (< 5%) at the sensor site, discomfort or mild bleeding from blood draws, and temporary pain or swelling at the needle site. These are considered expected and low-risk based on prior studies.

• Unexpected harms: Any other adverse events not described above (e.g. unexpected allergic reactions, device malfunction, or data breaches leading to privacy issues) will be considered unexpected.

• Assessment of harms: All harms will be monitored throughout the study by the research team and reported to the Principal Investigator. Participants are instructed to report any adverse effects, and study staff will document these in case report forms. Harms will be assessed systematically during scheduled visits and non-systematically if reported by participants between visits.

• Standardized terminology: Harms will be grouped and described using standard clinical terms, and all harms—anticipated and unexpected—will be included in study reports and publications, with severity and frequency summarized appropriately.

• Management: The Principal Investigator will be responsible for ensuring appropriate medical care is provided in the event of any harm.

Participant timeline {18}

The participant timeline is shown in Fig. 2.

Sample size {19}

Power calculations and statistical analysis

Sample size calculations

The study is a two-arm randomized controlled trial with equal allocation (1:1). Based on current evidence, the cumulative incidence of T2DM in women with prior GDM over 3 to 5 years is approximately 40% [10, 40–43]. Clinical trials have shown that lifestyle interventions can reduce this risk by up to 50%. Notably, the U.S. Diabetes Prevention Program (DPP) demonstrated a 50% reduction in T2DM incidence among women with a history of GDM following lifestyle modification [7]. The sample size was calculated to detect a 20% absolute risk reduction (from 40 to 20%) in the intervention group compared to the control group, with a two-sided alpha of 0.05 and 80% power. To account for potential attrition, we plan to recruit between 200 participants, allowing for a dropout rate of 10%, in order to retain a final sample size of 180 (90 participants per study arm).

Recruitment {20}

Eligible participants will be identified and approached through multiple recruitment strategies, including direct referrals from Obstetricians and Gynaecologists at KK Women’s and Children’s Hospital (KKH), display of study posters in outpatient clinics, and direct approach of patients in the labour ward. These combined strategies are intended to ensure timely enrolment to reach the target sample size.

Assignment of interventions: randomization

Sequence generation: who will generate the sequence {21a}

The allocation sequence is generated by a study team member not involved in recruitment or intervention delivery.

Sequence generation: type of randomization {21b}

Following consent, participants are enrolled by trained research coordinators and randomly assigned to one of two comparison groups in a 1:1 ratio using block randomization (block size of 2).

Allocation concealment mechanism {22}

To minimize allocation bias and ensure balanced group assignment, the randomization was implemented using sequentially numbered, opaque sealed envelopes to ensure allocation concealment.

Implementation {23}

The personnel who will enrol the participants and assign participants to the interventions are the research coordinators, and those who will assign participants to the interventions will not have access to the random allocation sequence.

Assignment of interventions: blinding

Who will be blinded {24a}

Not applicable—this is an unblinded randomized controlled trial.

How will be blinding be achieved {24b}

Not applicable—this is an unblinded randomized controlled trial.

Procedure for unblinding if needed {24c}

Not applicable—this is an unblinded randomized controlled trial.

Data collection and management

Plans for assessment and collection of outcomes {25a}

Data collection for all study participants from the intervention and control groups

Figure 1 provides an overview of the data collection procedures. The instruments and data collection forms used in this study are available from the authors upon reasonable request.

Blood collection

A fasting blood test is performed at 8–12 weeks postpartum at the baseline visit for all participants and repeated at subsequent timepoints throughout the study. This test includes HbA1c and a lipid profile, comprising total cholesterol, triglycerides, HDL cholesterol, and LDL cholesterol. HbA1c offers a longer-term view of glycaemic control and may reveal early dysglycaemia not detected by OGTT alone, while the lipid profile helps assess cardiometabolic risk, which is often elevated in this population [30].

At all follow-up timepoints—6, 12, 24, and 36 months after the baseline visit—participants undergo a 2-h 75-g oral glucose tolerance test (OGTT), with blood samples collected at fasting (FG), 1 h (1hPG), and 2 h (2hPG) post-glucose ingestion. The two-timepoint OGTT (fasting and 2-h values) is used to formally diagnose dysglycaemia, including prediabetes and type 2 diabetes mellitus (T2DM), based on standard diagnostic thresholds recommended by the 2006 WHO criteria [31].

T2DM is defined as a fasting plasma glucose ≥ 7.0 mmol/L (126 mg/dL) or 2-h plasma glucose ≥ 11.1 mmol/L (200 mg/dL). Impaired glucose tolerance (IGT) is diagnosed when fasting glucose is < 7.0 mmol/L (126 mg/dL) with a 2-h plasma glucose between 7.8 and < 11.1 mmol/L (140–199 mg/dL), while impaired fasting glucose (IFG) is defined as fasting glucose between 6.1 and 6.9 mmol/L (110–125 mg/dL) and a 2-h plasma glucose < 7.8 mmol/L (< 140 mg/dL). Longitudinal collection of this data enables the detection of subclinical metabolic changes over time and the identification of early predictors of glycaemic deterioration in this high-risk cohort.

All biochemical analyses are conducted using an automated biochemical analyzer (Abbott Alinity) at KK Women’s and Children’s Hospital Laboratories.

Clinical examination

Blood pressure (BP), height (centimetres), weight (kilograms), body fat mass (kg), total body water (kg), and waist circumference (centimetres) are measured at the baseline visit and at each follow-up visit throughout the study. BP is measured twice using an automated blood pressure monitor (Dinamap Carescape V100). A third measurement is obtained if the difference between the first and second measures exceeds 10 mmHg, with the average taken for analysis. Height is recorded to the nearest 0.5 cm using a calibrated wall-mounted stadiometer (SECA 2130), while weight, body fat mass, and total body water are measured using a calibrated portable bioimpedance analysis instrument (Tanita SC-240). Height and weight are each measured twice, to the nearest 0.1 cm and 0.1 kg, respectively. A third measurement is taken if the difference between the first two readings exceeds 1.0 cm for height or 0.2 kg for weight. The average of the measures is reported. BMI is calculated as weight in kilograms divided by height in meters squared (kg/m²). Waist circumference is measured in centimetres (cm), with a measuring tape placed horizontally at the midpoint between the iliac crest and the lowest lateral rib, directly on the skin. Two measurements are taken to the nearest 0.1 cm, and a third measurement will be taken if the difference between the first two readings exceeds 1.0 cm. To ensure data quality, all research coordinators underwent training in standardized protocols for anthropometric and body composition measurements.

Anthropometric and body composition measures, including weight, waist circumference, and fat distribution, provide insight into adiposity patterns and insulin resistance risk, particularly relevant in Asian populations [32].

Self-reported questionnaire

The questionnaires used to collect data on sociodemographic and lifestyle characteristics, family history of diabetes, breastfeeding practices, and study feedback were custom-designed for this study. This intends to elucidate the contextual and behavioural factors that may influence postpartum metabolic health and engagement with the intervention.

Dietary intake at 26–28 weeks’ gestation will be assessed using a single 24-h dietary recall. This recall is conducted by trained clinical research staff on either a weekday or weekend, employing a five-step, multiple-pass interviewing method. Standardized household measuring utensils and food portion photographs are used to help participants accurately quantify their intake from the previous day [33].

Nutrient intake will be estimated using Dietplan (Forestfield Software, UK), a nutrient analysis software program based on a food composition database containing local foods, with slight modifications to correct inaccuracies. For local dishes not found in the database, nutrient compositions will be calculated based on ingredient-level nutrition values from generic recipes. For individual food items not included in the database, nutrient data will be sourced from food packaging labels or the U.S. Department of Agriculture nutrient database, particularly for commonly imported Western food products [34].

Physical activity is assessed using the International Physical Activity Questionnaire (IPAQ) long form, which captures a 7-day recall of activity lasting ≥ 10 min per bout. The questionnaire categorizes activity by intensity (vigorous, moderate, and walking) and duration (days per week and minutes per day) [35]. Total weekly minutes of vigorous, moderate, and walking activity will be calculated and capped at a maximum of 3 h per day or 21 h per week, following IPAQ coding guidelines [36].

Collecting dietary intake and physical activity data longitudinally will enable the identification of behavioural patterns and changes over time, providing insight into how these behaviours may be influenced by the intervention.

Health-related quality of life (HrQoL) is evaluated using the EuroQol-5 Dimensions-5 Levels (EQ-5D-5L) instrument. This tool is widely used for assessing general HrQoL, conducting cost-utility analyses of healthcare interventions, and calculating quality-adjusted life years (QALYs). It comprises two parts: a descriptive component covering five health dimensions—Mobility, Self-Care, Usual Activities, Pain/Discomfort, and Anxiety/Depression—each with three response levels (no problems, some problems, extreme problems); and a visual analogue scale (VAS), where respondents rate their overall health on the day of the survey using a 0–100 vertical scale [37].

The 6P tool—comprising Portion, Proportion, Pleasure, Phase, Physicality, and Psychology—was developed as a lifestyle intervention tool and has been validated for use in overweight and obese women [38]. It is administered via a digital platform that enables the generation of individualized feedback reports and real-time monitoring charts. The feedback report provides personalized insights into the health implications of each participant’s lifestyle patterns, and includes recommendations tailored to their specific dietary challenges. The monitoring chart allows participants to track their 6P scores over time, supporting ongoing self-monitoring and engagement. Based on the identified lifestyle issues, participants are coached by dietitians and Clinical Exercise Physiologists to help them improve their health behaviours. Coaching includes goal setting that aligns with each participant’s unique challenges and targets for lifestyle improvement. All data from the 6P database are de-identified, and extracted data are stored on an encrypted hard drive accessible only to authorized study team members, ensuring the protection of participant privacy and data security.

Pregnancy and delivery outcome data

To comprehensively assess the risk of progression to T2DM in women with a history of GDM, we are collecting detailed maternal, delivery, and neonatal data. These variables were selected based on their known, or potential associations with postpartum metabolic outcomes and long-term cardiometabolic risk.

Maternal data: parity; pre-pregnancy weight; gestational week of GDM diagnosis; results of the 75-g OGTT during pregnancy; HbA1c values; treatment of GDM during pregnancy; history of antenatal risk factors such as subfertility, uterine surgery, preterm delivery, preeclampsia in previous pregnancy; pregnancy complications such as preeclampsia (de novo BP _ 140/90 mmHg > 20 weeks with proteinuria or signs of end-organ dysfunction), eclampsia, gestational hypertension (de novo BP _ 140/90 mmHg > 20 weeks), preexisting hypertension, renal disease, cardiac disease and autoimmune disease.

Delivery data: type of delivery (spontaneous vaginal, forceps or vacuum, caesarean section during labour or planned caesarean section) and complications during delivery such as uterine rupture, manual placental removal, third-degree tears, or obstetric hysterectomy.

Neonatal data: birth weight, 1-min and 5-min Apgar score, and shoulder dystocia.

Outcomes

Primary outcomes

The primary outcome is the cumulative incidence of type 2 diabetes mellitus (T2DM) at 36 months following the baseline visit (conducted at 8–12 weeks postpartum), which will be determined by a 75 g oral glucose tolerance test (OGTT). T2DM will be diagnosed based on the WHO criteria of fasting plasma glucose ≥ 7.0 mmol/L (126 mg/dL) or 2-h plasma glucose ≥ 11.1 mmol/L (200 mg/dL).

Secondary outcomes

• To evaluate the incidence of dysglycaemia at 6, 12, 24, and 36 months, where prediabetes will be classified according to the 2006 WHO guidelines as either impaired glucose tolerance (IGT), defined by fasting plasma glucose < 7.0 mmol/L (126 mg/dL) accompanied by a 2-h plasma glucose ≥ 7.8 and < 11.1 mmol/L (140–199 mg/dL), or impaired fasting glucose (IFG), defined by fasting plasma glucose between 6.1 and 6.9 mmol/L (110–125 mg/dL) accompanied by a 2-h plasma glucose < 7.8 mmol/L (< 140 mg/dL) [31].

• To evaluate changes from baseline to 6 months post-intervention, and longitudinally across all measurement timepoints in:

  • i.
    Cardiovascular risk factors associated with metabolic syndrome, including blood pressure (BP), body mass index (BMI), HbA1c, fasting and 2-h blood glucose levels, and fasting lipid profile (triglycerides (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL)) between the control and intervention groups.
  • ii.
    Self-reported total physical activity, measured in MET-minutes/week, between the control and intervention groups.
  • iii.
    Dietary intake, including total energy intake (kilocalories) over 24 h and daily servings from key food groups, between the control and intervention groups.

Tertiary outcomes

An incremental cost-effectiveness analysis will be conducted from the health system perspective, in line with the priorities of Singapore’s Agency for Care Effectiveness [39]. This analysis will incorporate data from the study’s clinical outcomes, as well as local population data and relevant external sources. Costs will be captured using Activity-Based Costing and standardized cost collection tools previously validated [2]. A lifetime horizon will be applied, with costs and outcomes discounted at 3% annually. Long-term cost-effectiveness will be modelled using adapted UKPDS and CDC diabetes models to estimate quality-adjusted life years (QALYs) gained based on changes in HbA1c and other clinical outcomes at trial end.

Plans to promote participant retention and complete follow-up {25b}

Drawing on our experience with recruiting and retaining pregnant and postpartum women, we have implemented several strategies to support high retention rates. These include financial reimbursements: participants receive SGD 50 following each clinic visit (baseline, and at 6-month, 1-, 2-, and 3-year follow-ups) and SGD 25 after each home visit (at 1-, 2-, and 4-month follow-ups). In addition, study visit consultation fees are waived, and home visits are offered at selected timepoints to enhance convenience. As participants are encouraged to view the programme as an extension of their postpartum care, we anticipate a low risk of loss to follow-up.

Data management {26}

Data management, monitoring and safety

Assessment data are entered electronically into REDCap (Research Electronic Data Capture), a secure, web-based application designed for data collection and management. To ensure data quality and protocol adherence, weekly meetings are conducted with all staff involved in data collection, entry, and management to address ongoing and emerging issues. Participant confidentiality is maintained before, during, and after the trial through measures such as pseudonymization, secure storage of identifying information separate from anonymized research data, and password-protected database systems. The principal investigator will have full access to the final dataset and will oversee the intrastudy data-sharing process. Access to participant identifying information is strictly limited to an independent trustee. All data shared with study collaborators will be de-identified to ensure participant confidentiality is preserved. Participants are also advised that they can withdraw from the study at any time without consequence from the research team and medical or home visiting services, and if this occurs, the research team may use any data collected before withdrawal.

Confidentiality {33}

Participant confidentiality

Participants’ personal data will be collected and stored securely in Singapore in accordance with SingHealth’s Data Protection Policy. All personal identifiers will be replaced with unique codes, and the linking file will be stored separately with restricted access. Only the study team will have access to identifiable data. Study monitors, auditors, the Institutional Review Board, and regulatory authorities may access de-identified study records for verification purposes. Data will be used solely for research purposes and will not be publicly disclosed.

Statistical methods

Statistical methods for primary and secondary outcomes {27a}

Sample size calculations

The study is a two-arm randomized controlled trial with equal allocation (1:1). Based on current evidence, the cumulative incidence of T2DM in women with prior GDM over 3 to 5 years is approximately 40% [10, 40–43]. Clinical trials have shown that lifestyle interventions can reduce this risk by up to 50%. Notably, the U.S. Diabetes Prevention Program (DPP) demonstrated a 50% reduction in T2DM incidence among women with a history of GDM following lifestyle modification [7]. The sample size was calculated to detect a 20% absolute risk reduction (from 40 to 20%) in the intervention group compared to the control group, with a two-sided alpha of 0.05 and 80% power. To account for potential attrition, we plan to recruit between 200 participants, allowing for a dropout rate of 10%, in order to retain a final sample size of 180 (90 participants per study arm).

Data analysis

Descriptive statistics will be presented as frequencies and percentages for categorical variables, and mean with standard deviations or median with interquartile range for continuous variables. Statistical tests were performed. Chi-squared or Fisher’s exact test for categorical variables, and independent T-tests. Nonparametric tests were applied when appropriate. Comparisons of the cumulative incidence of T2DM at 36 months from baseline between the control and intervention groups will be assessed using multivariable Poisson regression models with robust standard errors to estimate incidence rate ratios (IRRs). Changes in cardiometabolic markers from baseline to 6 months post-intervention will be analyzed using multivariable linear regression models, also adjusting for potential confounders. Longitudinal changes in these markers from baseline through 6, 12, 24, and 36 months will be evaluated using multivariable linear mixed-effects models to account for repeated measures over time. The principle of intention-to-treat will be adopted for all outcomes. To address missing data, multiple imputation by chained equations will be employed. Sensitivity analyses using complete-case data will also be performed to assess the robustness of the findings. Statistical significance will be defined as a two-tailed p-value < 0.05.

Who will be included in each analysis {27b}

All participants.

How missing data will be handled in the analysis {27c}

To address missing data, multiple imputation by chained equations will be employed.

Methods for additional analyses (e.g. subgroup analyses) {27d}

SPIRIT guidance: Methods for any additional analyses (e.g. subgroup and adjusted analyses).

Interim analyses {28b}

There are no interim analyses plans for this study.

Protocol and statistical analysis plan {5}

The trial protocol and statistical analysis plan are directly integrated into the study protocol being submitted for publication.

Oversight and monitoring

Composition of the coordinating centre and trial steering committee {3d}

The trial coordinating centre comprises the study Principal Investigator (PI), Co-Investigators (Co-Is), research coordinators, a dietitian, and exercise physiologists.

The PI and Co-Is are responsible for ensuring that the study is conducted in accordance with the study protocol and that data collection is rigorous and of high quality. Research coordinators are responsible for coordinating study participants’ clinic visits and follow-up activities. The dietitian and exercise physiologists are involved in delivering the intervention workshops to participants allocated to the intervention group.

To ensure data quality and adherence to the study protocol, weekly meetings are held with all staff involved in data collection, data entry, and data management to address ongoing and emerging issues.

Composition of the data monitoring committee, its role and reporting structure {28a}

A Data Monitoring Committee was not established for this single-centre randomized controlled trial as no interim analyses are planned and the intervention is considered low risk. Study oversight will be provided by the Principal Investigator, with the study team meeting weekly to review study conduct and address any emerging issues. In addition, audit checks will be conducted by the Office of Research Integrity and Compliance (ORIC), as required, to ensure ongoing study integrity and participant safety.

Frequency and plans for auditing trial conduct {29}

Internal monitoring of the trial will be conducted every 6 months. In addition, audit checks will be performed by the Office of Research Integrity and Compliance (ORIC), as required, to ensure ongoing study integrity and participant safety.

Protocol amendments {31}

Protocol amendments are communicated to SingHealth’s CIRB and reflected in updates to the trial registration on ClinicalTrials.gov.

Dissemination policy {8}

Study findings will be shared through publications in international peer-reviewed journals and presentations at scientific conferences. Individuals who make significant contributions to the trial’s design, implementation, analysis, or reporting will be acknowledged with authorship on the final trial report. Data collection to assess the intervention’s primary aim after the 3-year follow-up from baseline visit (10–12 weeks postpartum) is expected to end in September 2028.

Discussion

Anticipated findings

The IHIPS programme is the first hospital-based, multicomponent lifestyle intervention designed to integrate individualized goal-setting workshops with the use of wearable technology. The primary aim of this randomized controlled trial is to reduce the incidence of T2DM among women with a history of GDM in their index pregnancy who have normal glucose tolerance in the early postpartum period. We hypothesize that the combined use of CGM and a fitness tracker will enhance self-awareness and motivation by providing real-time feedback, positive reinforcement, and targeted support for behavioural changes in diet, physical activity, and weight management—ultimately reducing the risk of progression to T2DM. The findings from this study will contribute to the limited body of evidence on the effectiveness of hospital-based lifestyle intervention programmes, particularly in addressing the current gap in follow-up care for postpartum women, especially those with normal glucose tolerance postpartum, a group often overlooked in follow-up care despite their elevated long-term T2DM [22], and CVD risk [23]. Furthermore, a systematic review and meta-analysis found that behavioural strategies such as goal setting, self-monitoring, problem-solving, behavioural substitution, and credible sources are effective components of postpartum lifestyle interventions [44]; however, few studies have examined their implementation in hospital-based programmes targeting early postpartum women [11].

Strengths and limitations

Our study addresses a significant gap in the literature—the lack of long-term, structured lifestyle interventions initiated in hospital settings. The early postpartum period offers a unique window of opportunity for behavioural change, and initiating an intervention during hospitalization allows for immediate engagement at a time when motivation may be higher. This study demonstrates the feasibility of embedding lifestyle interventions into routine hospital workflows, potentially enhancing scalability and sustainability if proven effective.

By initiating the intervention in hospital and continuing follow-up after discharge, we also support greater continuity of care—an important factor for promoting long-term behavioural change and improved health outcomes. Furthermore, our study provides insights into how hospital-initiated programmes with structured follow-up can be designed, and evaluated in real-world settings. To enhance reach and retention, we employed multiple recruitment strategies, including universal participant incentives (e.g. cash reimbursements), home visits for select sessions, and fee waiver for consultations.

Despite these strengths, the study has several limitations. First, although we used diverse recruitment strategies, enrolling participants in the early postpartum period may have limited programme reach. The competing demands of new motherhood—such as infant care, fatigue, and time constraints—can significantly hinder engagement in lifestyle programmes. This period is particularly challenging, as women often prioritize childcare and family responsibilities over their own health and well-being. However, existing evidence suggests that interventions delivered more than 10 years after delivery are even more difficult to implement in routine care [7, 24], and meaningful improvements in metabolic outcomes are most likely when interventions are offered within the first 6 months postpartum [11].

Second, participation in hospital-based programmes may attract women who are already more motivated to improve their health, potentially limiting the generalizability of findings. Additionally, as this was a single-site study conducted within a specific cultural and healthcare context (multi-ethnic Asian Singapore population), external validity may be limited.

Finally, this was a non-blinded randomized controlled trial. The lack of blinding may have introduced performance bias: for example, intervention participants may have been more likely to modify their lifestyle due to increased attention and perceived support, while control participants may have disengaged due to lack of structured guidance. There is also potential information spillover, as participants in the control group may have been indirectly exposed to intervention content through interactions with study staff or other participants. However, our study is designed to minimize such risks, such as training research coordinators to use a consistent script with control participants, and avoid disclosing any intervention content.

Trial status

Participant recruitment began on the 1st of September 2021, is ongoing, and is expected to end on the 31st of July 2025. The final follow-up for the last recruited participant is expected on 15 September 2028.

Abbreviations

BMI

Body mass index

BP

Blood pressure

CDC

Centers for Disease Control and Prevention

CGM

Continuous glucose monitoring

CIRB

Centralised Institutional Review Board

CVD

Cardiovascular disease

DPP

Diabetes Prevention Program

EQ-5D-5L

EuroQol-5 Dimensions-5 Levels

FG

Fasting glucose

Fitbit

Fitbit fitness tracker (commercial wearable device)

FPG

Fasting plasma glucose

GDM

Gestational diabetes mellitus

HbA1c

Glycated haemoglobin

HDL

High-density lipoprotein

IHIPS

Integrated Hyperglycaemia Incentivized Postnatal Surveillance

IADPSG

International Association of Diabetes and Pregnancy Study Groups

IFG

Impaired fasting glucose

IGT

Impaired glucose tolerance

IPAQ

International Physical Activity Questionnaire

IRR

Incidence rate ratio

KKH

KK Women’s and Children’s Hospital

LDL

Low-density lipoprotein

MET

Metabolic equivalent of task

NGT

Normal glucose tolerance

OGTT

Oral glucose tolerance test

QALY

Quality-adjusted life year

RCT

Randomized controlled trial

REDCap

Research Electronic Data Capture

SECA

SECA anthropometric measurement equipment (brand name)

SGD

Singapore dollar

SMS

Short message service

T2DM

Type 2 diabetes mellitus

TG

Triglycerides

UKPDS

United Kingdom Prospective Diabetes Study

USDA

United States Department of Agriculture

VAS

Visual analogue scale

WHO

World Health Organization

Authors’ contributions {3a}

PLQ, KHT, EF contributed to the conception and design of the study. PLQ, WMH, EJKL, FF, and MAAB contributed to the development of the study protocol and the workflow. PLQ wrote the first draft of the manuscript. KHT obtained the research funding for this study. JGE, SYC, HS, LKT, and SS co-authors provided supervision and reviewed the final version of the manuscript.

Funding {7a}

Funding for this investigator-initiated study was provided by the National Medical Research Council (NMRC) Open Fund Large Collaborative Grant Metabolic Health in Asian Women and their Children (OFLCG19May-0033).

Data availability {6}

De-identified individual participant data, including the data dictionary and statistical analysis code, will be made available upon reasonable request following publication of the primary trial results. Requests should be submitted to the corresponding author and will be reviewed by the study investigators. Data will be shared in a secure format, subject to approval by the Institutional Review Board and in accordance with institutional data governance policies. Access to the final trial dataset will be restricted to the Principal Investigator and designated study investigators. There are no contractual agreements that limit investigators’ access to the final trial dataset. No baseline or pilot data are included in this study protocol.

Declarations

Ethics approval and consent to participate {30}

All participants provided informed consent, which will be obtained by the study team’s research coordinators. The study was approved by the SingHealth’s Centralised Institutional Review Board (CIRB) (CIRB Ref No: 202010-00067). Informed consent form and study protocol have been provided as Supplementary Material.

Consent for publication

Not applicable.

Competing interests {7b}

The authors have declared that no competing interests exist.