Trial Review

Please note the ANZCTR will be unattended from Friday 20 December 2024 for the holidays. The Registry will re-open on Tuesday 7 January 2025. Submissions and updates will not be processed during that time.

Registering a new trial?

To achieve prospective registration, we recommend submitting your trial for registration at the same time as ethics submission.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12620000915954
Ethics application status
Approved
Date submitted
5/05/2020
Date registered
17/09/2020
Date last updated
17/09/2020
Date data sharing statement initially provided
17/09/2020
Type of registration
Retrospectively registered

Titles & IDs
Public title
The Prediction modelling for Risk-Stratified care for women with Gestational Diabetes (PeRSonal GDM) study: Calculating the individualised risk of adverse outcomes for women with gestational diabetes.
Scientific title
The Prediction modelling for Risk-Stratified care for women with Gestational Diabetes (PeRSonal GDM) study: a model development and validation study for adverse outcomes in women with gestational diabetes.
Secondary ID [1] 301116 0
Nil known
Universal Trial Number (UTN)
Trial acronym
PeRSonal GDM
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Gestational diabetes 317210 0
Condition category
Condition code
Metabolic and Endocrine 315348 315348 0 0
Diabetes
Reproductive Health and Childbirth 315349 315349 0 0
Fetal medicine and complications of pregnancy

Intervention/exposure
Study type
Observational
Patient registry
False
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
The condition observed is gestational diabetes. This observational study will use health data routinely collected for clinical care and to meet statutory perinatal data reporting requirements. Participants in this study will not be required to complete any additional questionnaires, interviews or tests. Observation will include the duration of the pregnancy and the immediate postpartum period until discharge from the hospital.
Intervention code [1] 317421 0
Not applicable
Comparator / control treatment
No control group
Control group
Uncontrolled

Outcomes
Primary outcome [1] 323601 0
Composite adverse pregnancy outcome consisting of a combination of the following eight prioritised, objective and serious adverse component pregnancy outcomes. The following is a list of the component outcomes with the definition of each listed in the square brackets.
1. Hypertensive disorders of pregnancy [Pregnancy-induced hypertension, pre-eclampsia or eclampsia]
2. Large-for-gestational-age (LGA) [Birth weight > 90th percentile corrected for gestation and fetal sex using Australian population growth chart]
3. Neonatal hypoglycaemia requiring intravenous treatment [A neonate with a low blood glucose level fulfilling institutional criteria for intravenous treatment consisting of either a dextrose bolus or dextrose infusion]
4. Shoulder dystocia [When, after delivery of the head, the baby's anterior shoulder gets caught above the mother's pubic bone]
5. Fetal death [Death of fetus after 20 weeks gestation]
6. Neonatal death [Death of live-born neonate]
7. Neonatal bone fracture [Neonatal fracture (femur, humerus, clavicle or skull) suffered at birth]
8. Neonatal nerve palsy [Neonatal nerve palsy (brachial plexus injury or facial nerve injury) suffered at birth]

This outcome will be assessed using data from an existing maternity outcomes dataset. This dataset contains routinely and prospectively collected maternal, obstetric and neonatal data for all deliveries at the maternity service.
Timepoint [1] 323601 0
At the time of discharge from hospital following delivery.
Secondary outcome [1] 382331 0
Birth of large-for-gestational age neonate defined as birth weight > 90th percentile corrected for gestation and fetal sex using Australian population growth chart.

This outcome will be assessed using data from an existing maternity outcomes dataset (described above).
Timepoint [1] 382331 0
At the time of discharge from hospital following delivery.
Secondary outcome [2] 382602 0
Hypertensive disorders of pregnancy defined as pregnancy-induced hypertension, pre-eclampsia or eclampsia.

This outcome will be assessed using data from an existing maternity outcomes dataset (described above).
Timepoint [2] 382602 0
At the time of discharge from hospital following delivery.
Secondary outcome [3] 382603 0
Neonatal hypoglycaemia requiring intravenous treatment defined as a neonate with a low blood glucose level fulfilling institutional criteria for intravenous treatment consisting of either a dextrose bolus or dextrose infusion.

This outcome will be assessed using data from an existing maternity outcomes dataset (described above).
Timepoint [3] 382603 0
At the time of discharge from hospital following delivery.
Secondary outcome [4] 382604 0
Shoulder dystocia defined as when, after delivery of the head, the baby's anterior shoulder gets caught above the mother's pubic bone.

This outcome will be assessed using data from an existing maternity outcomes dataset (described above).
Timepoint [4] 382604 0
At the time of discharge from hospital following delivery.
Secondary outcome [5] 382605 0
Fetal death defined as death of fetus after 20 weeks gestation.

This outcome will be assessed using data from an existing maternity outcomes dataset (described above).
Timepoint [5] 382605 0
At the time of discharge from hospital following delivery.
Secondary outcome [6] 382606 0
Neonatal death defined as death of fetus after 20 weeks gestation.

This outcome will be assessed using data from an existing maternity outcomes dataset (described above).
Timepoint [6] 382606 0
At the time of discharge from hospital following delivery.
Secondary outcome [7] 382607 0
Neonatal bone fracture defined as fracture in the neonate at one of four sites (femur, humerus, clavicle or skull) suffered at birth.

This outcome will be assessed using data from an existing maternity outcomes dataset (described above).
Timepoint [7] 382607 0
At the time of discharge from hospital following delivery.
Secondary outcome [8] 382608 0
Neonatal nerve palsy defined as nerve palsy in the neonate at one of two sites (brachial plexus injury or facial nerve injury) suffered at birth.

This outcome will be assessed using data from an existing maternity outcomes dataset (described above).
Timepoint [8] 382608 0
At the time of discharge from hospital following delivery.
Secondary outcome [9] 382614 0
Postpartum glycaemia in the mother assessed using a 75-gram 2-hour oral glucose tolerance test (OGTT).

This outcome will be assessed using data linkage to pathology services.
Timepoint [9] 382614 0
Six to 52 weeks (1 year) postpartum.

This outcome will be assessed using results from any available OGTTs undertaken six to 52 weeks (1 year) following delivery. Where results from more than one OGTT are available the least abnormal test will be used.
Secondary outcome [10] 382615 0
Type 2 diabetes in the mother defined where any single one of the following criteria are met:
HbA1c greater than or equal to 6.5% (48 mmol/mol)
Fasting glucose greater than or equal to 7.0 mmol/L
Random glucose greater than or equal to 11.1 mmol/L
On a 75 g oral glucose tolerance test: fasting glucose greater than or equal to 7.0 mmol/L or 2 hr glucose greater than or equal to 11.1 mmol/L

This outcome will be assessed using data linkage to pathology services.
Timepoint [10] 382615 0
5 years following delivery
Secondary outcome [11] 382616 0
Glycaemia assessed in the mother using glycated haemoglobin (HbA1c). This outcome will be assessed using data linkage to pathology services.
Timepoint [11] 382616 0
5 years following delivery
Secondary outcome [12] 382617 0
Glycaemia in the mother assessed using fasting plasma glucose. This outcome will be assessed using data linkage to pathology services.
Timepoint [12] 382617 0
5 years following delivery
Secondary outcome [13] 382618 0
Glycaemia in the mother assessed using fasting capillary blood glucose. This outcome will be assessed using data linkage to pathology services.
Timepoint [13] 382618 0
5 years following delivery
Secondary outcome [14] 382619 0
Glycaemia in the mother assessed using post-prandial capillary blood glucose. This outcome will be assessed using data linkage to pathology services.
Timepoint [14] 382619 0
5 years following delivery
Secondary outcome [15] 382620 0
Diagnosis of gestational diabetes defined using the International Association of Diabetes in Pregnancy Study Groups diagnostic criteria (2010). This outcome will be assessed using data linkage to pathology services.
Timepoint [15] 382620 0
At any time during pregnancy

Eligibility
Key inclusion criteria
All pregnancies within the existing Monash Health maternity outcomes database with a birth recorded from 1 January 2016 to 31 December 2018.
Minimum age
18 Years
Maximum age
50 Years
Sex
Females
Can healthy volunteers participate?
Yes
Key exclusion criteria
None

Study design
Purpose
Natural history
Duration
Longitudinal
Selection
Defined population
Timing
Both
Statistical methods / analysis
This study will broadly examine prognosis and prediction modelling in gestational diabetes (GDM) across 3 phases:
1. The development and validation of a prognostic prediction model for pregnancy complications in women with gestational diabetes (primary analysis)
2. The updating of an existing diagnostic prediction model for the development of gestational diabetes amongst pregnant women
3. The development and validation of a diagnostic prediction model for the development of dysglycaemia following gestational diabetes

PRIMARY ANALYSIS (PHASE 1)
AIM: The development and validation of a prognostic prediction model for pregnancy complications in women with gestational diabetes.

To make individualised predictions for the binary composite of an adverse pregnancy outcome, we will apply a logistic regression modelling framework with the logit-probability of the composite outcome as the dependent variable.

Handling of predictors
Continuous variables will be kept as continuous in the model (rather than dichotomising), to avoid a loss of prognostic information. Those predictors that are highly correlated with others contribute little information and will be excluded from the statistical analysis.

The functional form of the relationship of continuous predictors with the outcome will be modelled with non-linear functions such as fractional polynomials (FP).

Model-building procedures (including predictor selection)
Candidate predictor variables will be selected a priori based on existing literature and clinical expertise as described above. During modelling, predictors will be selected by using a LASSO (Least Absolute Shrinkage and Selection Operator) method, which simultaneously selects the variables and penalises the model coefficients for over-optimism.

Examination of predictor interactions will be undertaken for the following groups of predictors: weight, gestational weight gain (GWG) and body mass index (BMI), and fasting, 1h and 2h glucose levels from OGTT.

Internal validation and assessment of model performance
The model performance will be assessed in terms of discrimination and calibration. We will use a bootstrap re-sampling technique to adjust for over-optimism in the estimation of model performance due to validation in the same dataset that is used to develop the model itself. We will use the area under the curve (AUC) of the receiver operating characteristic (ROC) curve with 95% confidence interval to assess the overall discriminatory ability of the developed model. We will report the apparent and adjusted for over-optimism model performance. A calibration plot will be created. This plot will facilitate the graphical assessment of calibration by putting affected women into groups ordered by predicted risk and considering the agreement between the mean predicted risk and the observed events in each risk group, usually deciles. The calibration will be summarized using the intercept and slope of the calibration plot. Internal validation, where the model’s predictions are compared to the observed data, should return perfect calibration to the development data (calibration slope = 1).

External validation
External validation of the developed model will be undertaken to assess temporal transportability. We will report the predictive performance in a more recently treated cohort at the same maternity service using the same measures of discrimination and calibration as used in internal validation. Development and validation data are identical in terms of eligibility criteria, outcome and predictors.

Presentation of a simplified model for clinical use
Once a final model is identified, we will simplify and adapt the presentation of the model to facilitate its application to clinical practice. Alternative modes of presentation will be explored with a focus on maximising end-user usability and promoting translation into clinical care. Various presentation formats will be considered, including a simplified scoring system, nomogram and web or app-based electronic risk calculators.

Assessment of clinical utility
To supplement traditional measures of predictive model performance, discrimination and calibration, clinical utility will be formally evaluated. We will use decision curve analysis to explore the net benefit of developed models over the entire range of probability thresholds. We will represent the net benefit as a function of the decision threshold in a decision curve plot. This will explore whether there is an overall net-benefit for using the models to stratify the population into two risk groups as a basis for a risk-stratified model of care:
1. Low-risk where the risk of adverse pregnancy outcomes is less than a pre-specified value—this group may be considered for a less intensive model-of-care;
2. High-risk where the risk is greater than a pre-specified value—this group should receive specialist-led hospital-based care.

Further formative research is planned to ascertain optimal risk thresholds. This will include engagement with stakeholders, including women affected by GDM and clinicians. A combination of focus groups and an electronic survey will be used.

Sensitivity analyses
We will conduct additional analysis to address the confounding effect of insulin treatment on predictor-outcome associations and hence the performance of the prediction model. This will consider four possible approaches with sensitivity analysis used to evaluate the robustness of each:
1. Derivation of a propensity score of being treated with insulin based on women pre-treatment characteristics. We will then weight observations by using the inverse probability of treatment weighting (IPTW). In this way, women with lower propensity to be treated will have more weight in the development of the prognostic model than those who had a higher probability of being treated.
2. Inclusion of insulin treatment as a component of the composite outcome.
3. Exclusion of cases where insulin treatment was used.
4. Exploration of the multinomial regression model framework for combinations of the composite outcome of adverse pregnancy outcome and insulin treatment.

The primary analysis will develop and validate a model based on clinical characteristics. Prognosis may also be influenced by an affected woman’s capacity to implement lifestyle measures such a dietary modification and increased exercise. Therefore, we will undertake a sensitivity analysis to evaluate whether measures of socioeconomic disadvantage can improve the prediction of adverse pregnancy outcomes.

All statistical analysis will be performed using Stata version 16.1 (College Station, TX: StataCorp LLC.).

ADDITIONAL ANALYSES
PHASE 2
AIM: The updating of an existing diagnostic prediction model for development of gestational diabetes (GDM) amongst pregnant women.

Missing data analyses
Exploratory analyses will be first conducted to assess the distributions of specific variables in the new dataset and its overall comparability with the original sample used for model development. This will consist of a comprehensive missingness analysis. Multiple imputation will be considered to account for missing data. We will conduct post-hoc sensitivity analyses to examine missingness handling.

Predicted outcome variable
We will first consider the outcome as specified in the original paper which is the individual GDM risk. Additional models will be run with other outcome variables, namely, individual GDM risk among those who received treatment and did not receive treatment. The three separate outcome variables are:
1. GDM [0== no_GDM; 1== GDM]
2. GDM + treatment [0== no_GDM and GDM no treatment; 1== GDM + treatment]
3. GDM + treatment2 [0=no_GDM; 1== GDM no treatment; 2== GDM + treatment]

Independent variables
Baseline model will consist of the same set of parameters included in the original paper The original risk score contains an “ethnicity” variable. We will model with “ethnicity” and “country of birth” separately to assess the better predictor of the two. . In addition, we will incorporate the following variables to assess their contribution to improving predictive performance: gestational weight gain, fasting blood sugar as numeric variables.

Analysis approach
A standard logistic regression without random effects will be run first so that results can be directly compared with the original risk tool. Mixed effects logistic regression modelling with the individual specified as the random effect will also be run so that all available information in the sample is accounted for. A decision-curve analysis will be performed to evaluate the net-benefit of using the updated diagnostic prediction model for selective screening versus universal screening.

Key analytic steps will comprise a) updating the original model with current dataset and baseline parameters b) updating the model by adding in extra predictors (gestational weight gain, fasting glucose) and c) re-estimating the regression coefficients of the original and additional predictors in the updated dataset. Model performance metrics spanning classification metrics (sensitivity, specificity, negative and positive predictive values, F-score, kappa coefficient), discrimination metrics (ROC curves, AUROC values) and calibration measures (calibration plots and statistics) will be used to evaluate and compare the robustness of updated clinical risk score.

PHASE 3
AIM: Diagnostic prediction model for development of dysglycaemia following gestational diabetes (GDM).

Logistic regression analysis will be used to examine the prognosis of women with GDM. Analysis will (1) examine factors that are associated with the different prognosis in women with GDM and (2) compare outcomes between women with and without GDM. All statistical analysis will be conducted using Stata (Release 15. College Station, TX: StataCorp LLC.).

A multivariable prediction model for postpartum dysglcyaemia including type 2 diabetes will be developed using non-identifiable dataset of routinely collected maternity outcome data. The clinical prediction problem will be framed using key items from the CHARMS (CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies). The methodology for model development and validation will be guided by the Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD) Statement.

Recruitment
Recruitment status
Active, not recruiting
Date of first participant enrolment
Anticipated
Actual
18/11/2019
Date of last participant enrolment
Anticipated
Actual
31/12/2019
Date of last data collection
Anticipated
31/12/2023
Actual
Sample size
Target
27000
Accrual to date
Final
27446
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 16548 0
Monash Medical Centre - Clayton campus - Clayton
Recruitment hospital [2] 16549 0
Dandenong Hospital - Dandenong
Recruitment hospital [3] 16550 0
Casey Hospital - Berwick
Recruitment postcode(s) [1] 30110 0
3168 - Clayton
Recruitment postcode(s) [2] 30111 0
3175 - Dandenong
Recruitment postcode(s) [3] 30112 0
3806 - Berwick

Funding & Sponsors
Funding source category [1] 305560 0
Government body
Name [1] 305560 0
National Health and Medical Research Council
Country [1] 305560 0
Australia
Primary sponsor type
University
Name
Monash Centre for Health Research and Implementation, Monash University
Address
Locked Bag 29
Clayton VIC 3168
Country
Australia
Secondary sponsor category [1] 305966 0
Hospital
Name [1] 305966 0
Diabetes Unit, Monash Health
Address [1] 305966 0
Special Medicine Building
Monash Medical Centre
246 Clayton Road
Clayton VIC 3168
Country [1] 305966 0
Australia
Secondary sponsor category [2] 306041 0
Hospital
Name [2] 306041 0
Monash Women's
Address [2] 306041 0
Monash Medical Centre
246 Clayton Road
Clayton VIC 3168
Country [2] 306041 0
Australia
Other collaborator category [1] 281300 0
Individual
Name [1] 281300 0
Prof Shakila Thangaratinam
Address [1] 281300 0
WHO Collaborating Centre for Women's Health
Institute of Translational Medicine
Heritage Building
Mindelsohn Way
Birmingham B15 2TH
Country [1] 281300 0
United Kingdom

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 305866 0
Monash Health Human Research Ethics Committee
Ethics committee address [1] 305866 0
Ethics committee country [1] 305866 0
Australia
Date submitted for ethics approval [1] 305866 0
17/09/2019
Approval date [1] 305866 0
15/11/2019
Ethics approval number [1] 305866 0
RES-19-0000713L

Summary
Brief summary
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 101862 0
Prof Helena Teede
Address 101862 0
Monash Centre for Health Research and Implementation
Locked Bag 29
Clayton VIC 3168
Country 101862 0
Australia
Phone 101862 0
+61 03 8572 2644
Fax 101862 0
Email 101862 0
Helena.Teede@monash.edu
Contact person for public queries
Name 101863 0
Shamil Cooray
Address 101863 0
Monash Centre for Health Research and Implementation
Locked Bag 29
Clayton VIC 3168
Country 101863 0
Australia
Phone 101863 0
+61 03 8572 2379
Fax 101863 0
Email 101863 0
Shamil.Cooray@monash.edu
Contact person for scientific queries
Name 101864 0
Shamil Cooray
Address 101864 0
Monash Centre for Health Research and Implementation
Locked Bag 29
Clayton VIC 3168
Country 101864 0
Australia
Phone 101864 0
+61 03 8572 2379
Fax 101864 0
Email 101864 0
Shamil.Cooray@monash.edu

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment
We do not have approval from the local human research and ethics committee to share individual participant data.


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
7757Study protocol    The study protocol is currently being peer-reviewe... [More Details]



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
SourceTitleYear of PublicationDOI
EmbaseProtocol for development and validation of a clinical prediction model for adverse pregnancy outcomes in women with gestational diabetes.2020https://dx.doi.org/10.1136/bmjopen-2020-038845
N.B. These documents automatically identified may not have been verified by the study sponsor.