Prognostic models in obstetrics: available, but far from applicable

Risk of caesarean delivery in labour induction: a systematic review and external validation of predictive models

BACKGROUND

Despite the existence of numerous published models predicting the risk of caesarean delivery in women undergoing induction of labour (IOL), validated models are scarce.

OBJECTIVES

To systematically review and externally assess the predictive capacity of caesarean delivery risk models in women undergoing IOL.

SEARCH STRATEGY

Studies published up to 15 January 2021 were identified through PubMed, CINAHL, Scopus and ClinicalTrials.gov, without temporal or language restrictions.

SELECTION CRITERIA

Studies describing the derivation of new models for predicting the risk of caesarean delivery in labour induction.

DATA COLLECTION AND ANALYSIS

Three authors independently screened the articles and assessed the risk of bias (ROB) according to the prediction model risk of bias assessment tool (PROBAST). External validation was performed in a prospective cohort of 468 pregnancies undergoing IOL from February 2019 to August 2020. The predictive capacity of the models was assessed by creating areas under the receiver operating characteristic curve (AUCs), calibration plots and decision curve analysis (DCA).

MAIN RESULTS

Fifteen studies met the eligibility criteria; 12 predictive models were validated. The quality of most of the included studies was not adequate. The AUC of the models varied from 0.520 to 0.773. The three models with the best discriminative capacity were those of Levine et al. (AUC 0.773, 95% CI 0.720-0.827), Hernández et al. (AUC 0.762, 95% CI 0.715-0.809) and Rossi et al. (AUC 0.752, 95% CI 0.707-0.797).

CONCLUSIONS

Predictive capacity and methodological quality were limited; therefore, we cannot currently recommend the use of any of the models for decision making in clinical practice.

TWEETABLE ABSTRACT

Predictive models that predict the risk of cesarean section in labor inductions are currently not applicable.

A prognostic model, including quantitative fetal fibronectin, to predict preterm labour: the QUIDS meta-analysis and prospective cohort study

BACKGROUND

The diagnosis of preterm labour is challenging. False-positive diagnoses are common and result in unnecessary, potentially harmful treatments (e.g. tocolytics, antenatal corticosteroids and magnesium sulphate) and costly hospital admissions. Measurement of fetal fibronectin in vaginal fluid is a biochemical test that can indicate impending preterm birth.

OBJECTIVES

To develop an externally validated prognostic model using quantitative fetal fibronectin concentration, in combination with clinical risk factors, for the prediction of spontaneous preterm birth and to assess its cost-effectiveness.

DESIGN

The study comprised (1) a qualitative study to establish the decisional needs of pregnant women and their caregivers, (2) an individual participant data meta-analysis of existing studies to develop a prognostic model for spontaneous preterm birth within 7 days in women with symptoms of preterm labour based on quantitative fetal fibronectin and clinical risk factors, (3) external validation of the prognostic model in a prospective cohort study across 26 UK centres, (4) a model-based economic evaluation comparing the prognostic model with qualitative fetal fibronectin, and quantitative fetal fibronectin with cervical length measurement, in terms of cost per QALY gained and (5) a qualitative assessment of the acceptability of quantitative fetal fibronectin.

DATA SOURCES/SETTING

The model was developed using data from five European prospective cohort studies of quantitative fetal fibronectin. The UK prospective cohort study was carried out across 26 UK centres.

PARTICIPANTS

Pregnant women at 22⁺⁰-34⁺⁶ weeks' gestation with signs and symptoms of preterm labour.

HEALTH TECHNOLOGY BEING ASSESSED

Quantitative fetal fibronectin.

MAIN OUTCOME MEASURES

Spontaneous preterm birth within 7 days.

RESULTS

The individual participant data meta-analysis included 1783 women and 139 events of spontaneous preterm birth within 7 days (event rate 7.8%). The prognostic model that was developed included quantitative fetal fibronectin, smoking, ethnicity, nulliparity and multiple pregnancy. The model was externally validated in a cohort of 2837 women, with 83 events of spontaneous preterm birth within 7 days (event rate 2.93%), an area under the curve of 0.89 (95% confidence interval 0.84 to 0.93), a calibration slope of 1.22 and a Nagelkerke R ² of 0.34. The economic analysis found that the prognostic model was cost-effective compared with using qualitative fetal fibronectin at a threshold for hospital admission and treatment of ≥ 2% risk of preterm birth within 7 days.

LIMITATIONS

The outcome proportion (spontaneous preterm birth within 7 days of test) was 2.9% in the validation study. This is in line with other studies, but having slightly fewer than 100 events is a limitation in model validation.

CONCLUSIONS

A prognostic model that included quantitative fetal fibronectin and clinical risk factors showed excellent performance in the prediction of spontaneous preterm birth within 7 days of test, was cost-effective and can be used to inform a decision support tool to help guide management decisions for women with threatened preterm labour.

FUTURE WORK

The prognostic model will be embedded in electronic maternity records and a mobile telephone application, enabling ongoing data collection for further refinement and validation of the model.

STUDY REGISTRATION

This study is registered as PROSPERO CRD42015027590 and Current Controlled Trials ISRCTN41598423.

FUNDING

This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 25, No. 52. See the NIHR Journals Library website for further project information.

Discriminant Analysis of Main Prognostic Factors Associated with Hemodynamically Significant PDA: Apgar Score, Silverman-Anderson Score, and NT-Pro-BNP Level

Hemodynamically significant patent ductus arteriosus (hsPDA) in premature newborns is associated with a risk of PDA-related morbidities. Classification into risk groups may have a clinical utility in cases of suspected hsPDA to decrease the need for echocardiograms and unnecessary treatment. This prospective observational study included 99 premature newborns with extremely low body weight, who had an echocardiogram performed within the first three days of life. Discriminant analysis was utilized to find the best combination of prognostic factors for evaluation of hsPDA. We used binary logistic regression analysis to predict the relationship between parameters and hsPDA. The cohort's mean and standard deviation gestational age was 27.6 ± 2.55 weeks, the mean birth weight was 1015 ± 274 g. Forty-six (46.4%) infants had a PDA with a mean diameter of 2.78 mm. Median NT-pro-BNP levels were 17,600 pg/mL for infants with a PDA and 2773 pg/mL in the non-hsPDA group. The combination of prognostic factors of hsPDA in newborns of extremely low body weight on the third day of life was determined: NT-pro-BNP, Apgar score, Silverman-Anderson score (Se = 82%, Sp = 88%). A cut-off value of NT-pro-BNP of more than 8500 pg/mL can predict hsPDA (Se = 84%, Sp = 86%).

The relation between second-trimester placental elasticity and poor obstetric outcomes in low-risk pregnancies

OBJECTIVES

Increased placental stiffness is associated with various pathological conditions. Our objective was to evaluate the relation between the second-trimester placental elasticity value in low-risk pregnant women and poor obstetric outcomes.

METHODS

A total of 143 pregnant women were enrolled. Placental elasticity values were measured using the transabdominal point shear wave elastography method. 10 random measurements were obtained from different areas of the placenta. The mean was accepted as the mean placental elasticity value. Logistic regression analyses were performed to identify independent variables associated with obstetric outcomes.

RESULTS

Second-trimester placental elasticity value was significantly and positively associated with the poor obstetric outcomes (p=0.038). We could predict a poor outcome with 69.2% sensitivity and 60.7% specificity if we defined the placental elasticity cut-off as 3.19 kPa. Furthermore, in the multiple regression model, the placental elasticity value added significantly to the prediction of birth weight (p=0.043).

CONCLUSIONS

Our results showed that the pregnancies with a stiffer placenta in the second trimester were associated with an increased likelihood of exhibiting poor obstetric outcomes. Also, placental elasticity was independently associated with birth weight.

Predictive models of individual risk of elective caesarean section complications: a systematic review

INTRODUCTION

With increasing caesarean section (c-section) rates, personalized communication of risk has become paramount. A reliable tool to predict complications would support evidence-based discussions around planned mode of birth. This systematic review aimed to identify, synthesize and quality appraise prognostic models of maternal complications of elective c-section.

METHODS

MEDLINE, Embase, Web of Science, CINAHL and the Cochrane Library were searched on 27 January using terms relating to 'c-section', 'prognostic models' and complications such as 'infection'. Any study developing and/or validating a prognostic model for a maternal complication of elective c-section in the English language after January 1995 was selected for analysis. Data were extracted using a predetermined checklist: source of data; participants; outcome to be predicted; candidate predictors; sample size; missing data; model development; model performance; model evaluation; results; and interpretation. Quality was assessed using the Prediction model Risk Of Bias ASsessment Tool (PROBAST) tool.

RESULTS

In total, 7752 studies were identified; of these, 16 full papers were reviewed and three eligible studies were identified, containing three prognostic models derived from hospitals in Japan, South Africa and the UK. The models predicted risk of blood transfusion, spinal hypotension and postpartum haemorrhage. The study authors deemed their studies to be exploratory, exploratory and confirmatory, respectively. From the three studies, a total of 29 unique candidate predictors were identified, with 15 predictors in the final models. Maternal age (n = 3), previous c-section (n = 2), placenta praevia (n = 2) and pre-operative haemoglobin (n = 2) were found to be common predictors amongst the included studies. None of the studies were externally validated and all had a high risk of bias due to the analysis technique used.

CONCLUSION

Few models have been developed to predict complications of elective c-section. Existing models predicting blood transfusion, spinal hypotension and postpartum haemorrhage cannot be recommended for clinical practice. Future research should focus on identifying predictors known before surgery and validating the resulting models.

Prognostic Models Predicting Mortality in Preterm Infants: Systematic Review and Meta-analysis

CONTEXT

Prediction models can be a valuable tool in performing risk assessment of mortality in preterm infants.

OBJECTIVE

Summarizing prognostic models for predicting mortality in very preterm infants and assessing their quality.

DATA SOURCES

Medline was searched for all articles (up to June 2020).

STUDY SELECTION

All developed or externally validated prognostic models for mortality prediction in liveborn infants born <32 weeks' gestation and/or <1500 g birth weight were included.

DATA EXTRACTION

Data were extracted by 2 independent authors. Risk of bias (ROB) and applicability assessment was performed by 2 independent authors using Prediction model Risk of Bias Assessment Tool.

RESULTS

One hundred forty-two models from 35 studies reporting on model development and 112 models from 33 studies reporting on external validation were included. ROB assessment revealed high ROB in the majority of the models, most often because of inadequate (reporting of) analysis. Internal and external validation was lacking in 41% and 96% of these models. Meta-analyses revealed an average C-statistic of 0.88 (95% confidence interval [CI]: 0.83-0.91) for the Clinical Risk Index for Babies score, 0.87 (95% CI: 0.81-0.92) for the Clinical Risk Index for Babies II score, and 0.86 (95% CI: 0.78-0.92) for the Score for Neonatal Acute Physiology Perinatal Extension II score.

LIMITATIONS

Occasionally, an external validation study was included, but not the development study, because studies developed in the presurfactant era or general NICU population were excluded.

CONCLUSIONS

Instead of developing additional mortality prediction models for preterm infants, the emphasis should be shifted toward external validation and consecutive adaption of the existing prediction models.

Bayesian networks in healthcare: What is preventing their adoption?

There has been much research effort expended toward the use of Bayesian networks (BNs) in medical decision-making. However, because of the gap between developing an accurate BN and demonstrating its clinical usefulness, this has not resulted in any widespread BN adoption in clinical practice. This paper investigates this problem with the aim of finding an explanation and ways to address the problem through a comprehensive literature review of articles describing BNs in healthcare. Based on the literature collection that has been systematically narrowed down from 3810 to 116 most relevant articles, this paper analyses the benefits, barriers and facilitating factors (BBF) for implementing BN-based systems in healthcare using the ITPOSMO-BBF framework. A key finding is that works in the literature rarely consider barriers and even when these were identified they were not connected to facilitating factors. The main finding is that the barriers can be grouped into: (1) data inadequacies; (2) clinicians' resistance to new technologies; (3) lack of clinical credibility; (4) failure to demonstrate clinical impact; (5) absence of an acceptable predictive performance; and (6) absence of evidence for model's generalisability. The facilitating factors can be grouped into: (1) data collection improvements; (2) software and technological improvements; (3) having interpretable and easy to use BN-based systems; (4) clinical involvement in the development or review of the model; (5) investigation of model's clinical impact; (6) internal validation of the model's performance; and (7) external validation of the model. These groupings form a strong basis for a generic framework that could be used for formulating strategies for ensuring BN-based clinical decision-support system adoption in frontline care settings. The output of this review is expected to enhance the dialogue among researchers by providing a deeper understanding for the neglected issue of BN adoption in practice and promoting efforts for implementing BN-based systems.

Implementation of a first-trimester prognostic model to improve screening for gestational diabetes mellitus

Background

Improvement in the accuracy of identifying women who are at risk to develop gestational diabetes mellitus (GDM) is warranted, since timely diagnosis and treatment improves the outcomes of this common pregnancy disorder. Although prognostic models for GDM are externally validated and outperform current risk factor based selective approaches, there is little known about the impact of such models in day-to-day obstetric care.

Methods

A prognostic model was implemented as a directive clinical prediction rule, classifying women as low- or high-risk for GDM, with subsequent distinctive care pathways including selective midpregnancy testing for GDM in high-risk women in a prospective multicenter birth cohort comprising 1073 pregnant women without pre-existing diabetes and 60 obstetric healthcare professionals included in nine independent midwifery practices and three hospitals in the Netherlands (effectiveness-implementation hybrid type 2 study). Model performance (c-statistic) and implementation outcomes (acceptability, adoption, appropriateness, feasibility, fidelity, penetration, sustainability) were evaluated after 6 months by indicators and implementation instruments (NoMAD; MIDI).

Results

The adherence to the prognostic model (c-statistic 0.85 (95%CI 0.81–0.90)) was 95% (n = 1021). Healthcare professionals scored 3.7 (IQR 3.3–4.0) on implementation instruments on a 5-point Likert scale. Important facilitators were knowledge, willingness and confidence to use the model, client cooperation and opportunities for reconfiguration. Identified barriers mostly related to operational and organizational issues. Regardless of risk-status, pregnant women appreciated first-trimester information on GDM risk-status and lifestyle advice to achieve risk reduction, respectively 89% (n = 556) and 90% (n = 564)).

Conclusions

The prognostic model was successfully implemented and well received by healthcare professionals and pregnant women. Prognostic models should be recommended for adoption in guidelines.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-021-03749-x.

Adaptive sample size determination for the development of clinical prediction models

BACKGROUND

We suggest an adaptive sample size calculation method for developing clinical prediction models, in which model performance is monitored sequentially as new data comes in.

METHODS

We illustrate the approach using data for the diagnosis of ovarian cancer (n = 5914, 33% event fraction) and obstructive coronary artery disease (CAD; n = 4888, 44% event fraction). We used logistic regression to develop a prediction model consisting only of a priori selected predictors and assumed linear relations for continuous predictors. We mimicked prospective patient recruitment by developing the model on 100 randomly selected patients, and we used bootstrapping to internally validate the model. We sequentially added 50 random new patients until we reached a sample size of 3000 and re-estimated model performance at each step. We examined the required sample size for satisfying the following stopping rule: obtaining a calibration slope ≥ 0.9 and optimism in the c-statistic (or AUC) < = 0.02 at two consecutive sample sizes. This procedure was repeated 500 times. We also investigated the impact of alternative modeling strategies: modeling nonlinear relations for continuous predictors and correcting for bias on the model estimates (Firth's correction).

RESULTS

Better discrimination was achieved in the ovarian cancer data (c-statistic 0.9 with 7 predictors) than in the CAD data (c-statistic 0.7 with 11 predictors). Adequate calibration and limited optimism in discrimination was achieved after a median of 450 patients (interquartile range 450-500) for the ovarian cancer data (22 events per parameter (EPP), 20-24) and 850 patients (750-900) for the CAD data (33 EPP, 30-35). A stricter criterion, requiring AUC optimism < = 0.01, was met with a median of 500 (23 EPP) and 1500 (59 EPP) patients, respectively. These sample sizes were much higher than the well-known 10 EPP rule of thumb and slightly higher than a recently published fixed sample size calculation method by Riley et al. Higher sample sizes were required when nonlinear relationships were modeled, and lower sample sizes when Firth's correction was used.

CONCLUSIONS

Adaptive sample size determination can be a useful supplement to fixed a priori sample size calculations, because it allows to tailor the sample size to the specific prediction modeling context in a dynamic fashion.

Accurate prediction of birth implementing a statistical model through the determination of steroid hormones in saliva

Steroidal hormone interaction in pregnancy is crucial for adequate fetal evolution and preparation for childbirth and extrauterine life. Estrone sulphate, estriol, progesterone and cortisol play important roles in the initiation of labour mechanism at the start of contractions and cervical effacement. However, their interaction remains uncertain. Although several studies regarding the hormonal mechanism of labour have been reported, the prediction of date of birth remains a challenge. In this study, we present for the first time machine learning algorithms for the prediction of whether spontaneous labour will occur from week 37 onwards. Estrone sulphate, estriol, progesterone and cortisol were analysed in saliva samples collected from 106 pregnant women since week 34 by enzyme-immunoassay (EIA) techniques. We compared a random forest model with a traditional logistic regression over a dataset constructed with the values observed of these measures. We observed that the results, evaluated in terms of accuracy and area under the curve (AUC) metrics, are sensibly better in the random forest model. For this reason, we consider that machine learning methods contribute in an important way to the obstetric practice.

A Simple Approach to Adjust for Case-Mix When Comparing Institutional Cesarean Birth Rates

OBJECTIVE

This study aimed to develop a validated model to predict intrapartum cesarean in nulliparous women and to use it to adjust for case-mix when comparing institutional laboring cesarean birth (CB) rates.

STUDY DESIGN

This multicenter retrospective study used chart-abstracted data on nulliparous, singleton, term births over a 7-year period. Prelabor cesareans were excluded. Logistic regression was used to predict the probability of CB for individual pregnancies. Thirty-five potential predictive variables were evaluated including maternal demographics, prepregnancy health, pregnancy characteristics, and newborn weight and gender. Models were trained on 21,017 births during 2011 to 2015 (training cohort), and accuracy assessed by prediction on 15,045 births during 2016 to 2017 (test cohort).

RESULTS

Six variables delivered predictive success equivalent to the full set of 35 variables: maternal weight, height, and age, gestation at birth, medically-indicated induction, and birth weight. Internal validation within the training cohort gave a receiver operator curve with area under the curve (ROC-AUC) of 0.722. External validation using the test cohort gave ROC-AUC of 0.722 (0.713-0.731 confidence interval). When comparing observed and predicted CB rates at 16 institutions in the test cohort, five had significantly lower than predicted rates and three had significantly higher than predicted rates.

CONCLUSION

Six routine clinical variables used to adjust for case-mix can identify outliers when comparing institutional CB rates.

Protocol for the development and validation of a risk prediction model for stillbirths from 35 weeks gestation in Australia

BACKGROUND

Despite advances in the care of women and their babies in the past century, an estimated 1.7 million babies are born still each year throughout the world. A robust method to estimate a pregnant woman's individualized risk of late-pregnancy stillbirth is needed to inform decision-making around the timing of birth to reduce the risk of stillbirth from 35 weeks of gestation in Australia, a high-resource setting.

METHODS

This is a protocol for a cross-sectional study of all late-pregnancy births in Australia (2005-2015) from 35 weeks of gestation including 5188 stillbirths among 3.1 million births at an estimated rate of 1.7 stillbirths per 1000 births. A multivariable logistic regression model will be developed in line with current Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD) guidelines to estimate the gestation-specific probability of stillbirth with prediction intervals. Candidate predictors were identified from systematic reviews and clinical consultation and will be described through univariable regression analysis. To generate a final model, elimination by backward stepwise multivariable logistic regression will be performed. The model will be internally validated using bootstrapping with 1000 repetitions and externally validated using a temporally unique dataset. Overall model performance will be assessed with R², calibration, and discrimination. Calibration will be reported using a calibration plot with 95% confidence intervals (α = 0.05). Discrimination will be measured by the C-statistic and area underneath the receiver-operator curves. Clinical usefulness will be reported as positive and negative predictive values, and a decision curve analysis will be considered.

DISCUSSION

A robust method to predict a pregnant woman's individualized risk of late-pregnancy stillbirth is needed to inform timely, appropriate care to reduce stillbirth. Among existing prediction models designed for obstetric use, few have been subject to internal and external validation and many fail to meet recommended reporting standards. In developing a risk prediction model for late-gestation stillbirth with both providers and pregnant women in mind, we endeavor to develop a validated model for clinical use in Australia that meets current reporting standards.

Predicting neonatal respiratory distress syndrome and hypoglycaemia prior to discharge: Leveraging health administrative data and machine learning

OBJECTIVES

A major challenge for hospitals and clinicians is the early identification of neonates at risk of developing adverse conditions. We develop a model based on routinely collected administrative data, which accurately predicts two common disorders among early term and preterm (<39 weeks) neonates prior to discharge.

STUDY DESIGN

The data included all inpatient live births born prior to 39 weeks (n = 154,755) occurring in the Australian state of Queensland between January 2009 and December 2015. Predictor variables included all maternal data captured in administrative records from the beginning of gestation up to, and including, the delivery, as well as neonatal data recorded at the delivery. Gradient boosted trees were used to predict neonatal respiratory distress syndrome and hypoglycaemia prior to discharge, with model performance benchmarked against a logistic regression models.

RESULTS

The gradient boosted trees model achieved very high discrimination for respiratory distress syndrome [AUC = 0.923, 95% CI (0.917, 0.928)] and good discrimination for hypoglycaemia [AUC = 0.832, 95% CI (0.827, 0.837)] in the validation data, as well as outperforming the logistic regression models.

CONCLUSION

Our study suggests that routinely collected health data have the potential to play an important role in assisting clinicians to identify neonates at risk of developing selected disorders shortly after birth. Despite achieving high levels of discrimination, many issues remain before such models can be implemented in practice, which we discuss in relation to our findings.

Risk prediction model of gestational diabetes mellitus based on nomogram in a Chinese population cohort study

To build a risk prediction model of gestational diabetes mellitus using nomogram to provide a simple-to-use clinical basis for the early prediction of gestational diabetes mellitus (GDM). This study is a prospective cohort study including 1385 pregnant women. (1) It is showed that the risk of GDM in women aged ≥ 35 years was 5.5 times higher than that in women aged < 25 years (95% CI: 1.27–23.73, p < 0.05). In the first trimester, the risk of GDM in women with abnormal triglyceride who were in their first trimester was 2.1 times higher than that of lipid normal women (95% CI: 1.12–3.91, p < 0.05). The area under the ROC curve of the nomogram of was 0.728 (95% CI: 0.683–0.772), the sensitivity and specificity of the model were 0.716 and 0.652, respectively. This study provides a simple and economic nomogram for the early prediction of GDM risk in the first trimester, and it has certain accuracy.

Protocol for development and validation of a clinical prediction model for adverse pregnancy outcomes in women with gestational diabetes

INTRODUCTION

Gestational diabetes (GDM) is a common yet highly heterogeneous condition. The ability to calculate the absolute risk of adverse pregnancy outcomes for an individual woman with GDM would allow preventative and therapeutic interventions to be delivered to women at high-risk, sparing women at low-risk from unnecessary care. The Prediction for Risk-Stratified care for women with GDM (PeRSonal GDM) study will develop, validate and evaluate the clinical utility of a prediction model for adverse pregnancy outcomes in women with GDM.

METHODS AND ANALYSIS

We undertook formative research to conceptualise and design the prediction model. Informed by these findings, we will conduct a model development and validation study using a retrospective cohort design with participant data collected as part of routine clinical care across three hospitals. The study will include all pregnancies resulting in births from 1 July 2017 to 31 December 2018 coded for a diagnosis of GDM (estimated sample size 2430 pregnancies). We will use a temporal split-sample development and validation strategy. A multivariable logistic regression model will be fitted. The performance of this model will be assessed, and the validated model will also be evaluated using decision curve analysis. Finally, we will explore modes of model presentation suited to clinical use, including electronic risk calculators.

ETHICS AND DISSEMINATION

This study was approved by the Human Research Ethics Committee of Monash Health (RES-19-0000713 L). We will disseminate results via presentations at scientific meetings and publication in peer-reviewed journals.

TRIAL REGISTRATION DETAILS

Systematic review proceeding this work was registered on PROSPERO (CRD42019115223) and the study was registered on the Australian and New Zealand Clinical Trials Registry (ACTRN12620000915954); Pre-results.

External validation of prognostic models predicting pre-eclampsia: individual participant data meta-analysis

BACKGROUND

Pre-eclampsia is a leading cause of maternal and perinatal mortality and morbidity. Early identification of women at risk during pregnancy is required to plan management. Although there are many published prediction models for pre-eclampsia, few have been validated in external data. Our objective was to externally validate published prediction models for pre-eclampsia using individual participant data (IPD) from UK studies, to evaluate whether any of the models can accurately predict the condition when used within the UK healthcare setting.

METHODS

IPD from 11 UK cohort studies (217,415 pregnant women) within the International Prediction of Pregnancy Complications (IPPIC) pre-eclampsia network contributed to external validation of published prediction models, identified by systematic review. Cohorts that measured all predictor variables in at least one of the identified models and reported pre-eclampsia as an outcome were included for validation. We reported the model predictive performance as discrimination (C-statistic), calibration (calibration plots, calibration slope, calibration-in-the-large), and net benefit. Performance measures were estimated separately in each available study and then, where possible, combined across studies in a random-effects meta-analysis.

RESULTS

Of 131 published models, 67 provided the full model equation and 24 could be validated in 11 UK cohorts. Most of the models showed modest discrimination with summary C-statistics between 0.6 and 0.7. The calibration of the predicted compared to observed risk was generally poor for most models with observed calibration slopes less than 1, indicating that predictions were generally too extreme, although confidence intervals were wide. There was large between-study heterogeneity in each model's calibration-in-the-large, suggesting poor calibration of the predicted overall risk across populations. In a subset of models, the net benefit of using the models to inform clinical decisions appeared small and limited to probability thresholds between 5 and 7%.

CONCLUSIONS

The evaluated models had modest predictive performance, with key limitations such as poor calibration (likely due to overfitting in the original development datasets), substantial heterogeneity, and small net benefit across settings. The evidence to support the use of these prediction models for pre-eclampsia in clinical decision-making is limited. Any models that we could not validate should be examined in terms of their predictive performance, net benefit, and heterogeneity across multiple UK settings before consideration for use in practice.

TRIAL REGISTRATION

PROSPERO ID: CRD42015029349 .

Predicting postpartum psychiatric admission using a machine learning approach

AIMS

The accurate identification of mothers at risk of postpartum psychiatric admission would allow for preventive intervention or more timely admission. We developed a prediction model to identify women at risk of postpartum psychiatric admission.

METHODS

Data included administrative health data of all inpatient live births in the Australian state of Queensland between January 2009 and October 2014. Analyses were restricted to mothers with one or more indicator of mental health problems during pregnancy (n = 75,054 births). The predictors included all maternal data up to and including the delivery, and neonatal data recorded at delivery. We used multiple machine learning methods to predict hospital admission in the 12 months following delivery in which the primary diagnosis was recorded as an ICD-10 psychotic, bipolar or depressive disorders.

RESULTS

The boosted trees algorithm produced the best performing model, predicting postpartum psychiatric admission in the validation data with good discrimination [AUC = 0.80; 95% CI = (0.76, 0.83)] and achieving good calibration. This model outperformed benchmark logistic regression model and an elastic net model. In addition to indicators of maternal metal health history, maternal and neonatal anthropometric measures and social/lifestyle factors were strong predictors.

CONCLUSION

Our results indicate the potential of a big data approach when aiming to identify mothers at risk of postpartum psychiatric admission. Mothers at risk could be followed-up and supported after neonatal discharge to either remove the need for admission or facilitate more timely admission.

Prediction of stillbirth: an umbrella review of evaluation of prognostic variables

BACKGROUND

Stillbirth accounts for over 2 million deaths a year worldwide and rates remains stubbornly high. Multivariable prediction models may be key to individualised monitoring, intervention or early birth in pregnancy to prevent stillbirth.

OBJECTIVES

To collate and evaluate systematic reviews of factors associated with stillbirth in order to identify variables relevant to prediction model development.

SEARCH STRATEGY

MEDLINE, Embase, DARE and Cochrane Library databases and reference lists were searched up to November 2019.

SELECTION CRITERIA

We included systematic reviews of association of individual variables with stillbirth without language restriction.

DATA COLLECTION AND ANALYSIS

Abstract screening and data extraction were conducted in duplicate. Methodological quality was assessed using AMSTAR and QUIPS criteria. The evidence supporting association with each variable was graded.

RESULTS

The search identified 1198 citations. Sixty-nine systematic reviews reporting 64 variables were included. The most frequently reported were maternal age (n = 5), body mass index (n = 6) and maternal diabetes (n = 5). Uterine artery Doppler appeared to have the best performance of any single test for stillbirth. The strongest evidence of association was for nulliparity and pre-existing hypertension.

CONCLUSION

We have identified variables relevant to the development of prediction models for stillbirth. Age, parity and prior adverse pregnancy outcomes had a more convincing association than the best performing tests, which were PAPP-A, PlGF and UtAD. The evidence was limited by high heterogeneity and lack of data on intervention bias.

TWEETABLE ABSTRACT

Review shows key predictors for use in developing models predicting stillbirth include age, prior pregnancy outcome and PAPP-A, PLGF and Uterine artery Doppler.

Can risk prediction models help us individualise stillbirth prevention? A systematic review and critical appraisal of published risk models

BACKGROUND

Stillbirth prevention is an international priority - risk prediction models could individualise care and reduce unnecessary intervention, but their use requires evaluation.

OBJECTIVES

To identify risk prediction models for stillbirth, and assess their potential accuracy and clinical benefit in practice.

SEARCH STRATEGY

MEDLINE, Embase, DH-DATA and AMED databases were searched from inception to June 2019 using terms relevant to stillbirth, perinatal mortality and prediction models. The search was compliant with Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.

SELECTION CRITERIA

Studies developing and/or validating prediction models for risk of stillbirth developed for application during pregnancy.

DATA COLLECTION AND ANALYSIS

Study screening and data extraction were conducted in duplicate, using the CHARMS checklist. Risk of bias was appraised using the PROBAST tool.

RESULTS

The search identified 2751 citations. Fourteen studies reporting development of 69 models were included. Variables consistently included were: ethnicity, body mass index, uterine artery Doppler, pregnancy-associated plasma protein and placental growth factor. For almost all models there were significant concerns about risk of bias. Apparent model performance (i.e. in the development dataset) was highest in models developed for use later in pregnancy and including maternal characteristics, and ultrasound and biochemical variables, but few were internally validated and none were externally validated.

CONCLUSIONS

Almost all models identified were at high risk of bias. There are first-trimester models of possible clinical benefit in early risk stratification; these require validation and clinical evaluation. There were few later pregnancy models but, if validated, these could be most relevant to individualised discussions around timing of birth.

TWEETABLE ABSTRACT

Prediction models using maternal factors, blood tests and ultrasound could individualise stillbirth prevention, but existing models are at high risk of bias.

Reducing emergency cesarean delivery and improving the primiparous experience: Findings of the RECIPE study

OBJECTIVE

The ability to predict the need for emergency Cesarean delivery holds the potential to facilitate birth choices. The objective of the RECIPE study (Reducing Emergency Cesarean delivery and Improving the Primiparous Experience) was to externally validate a Cesarean delivery risk prediction model. This model, developed by the Genesis study, identified five key predictive factors for emergency Cesarean delivery: maternal age, maternal height, BMI, fetal head circumference (HC) and fetal abdominal circumference (AC).

STUDY DESIGN

This prospective, observational study was conducted in two tertiary referral perinatal centers. Inclusion criteria were as follows: primiparous women with a singleton, cephalic presentation fetus in the absence of fetal growth restriction (FGR), oligohydramnios, pre-eclampsia, pre-existing diabetes mellitus or an indication for planned Cesarean delivery. Between 38 + 0 and 40 + 6 weeks' gestational age, participants attended for prenatal assessment that enabled the determination of an individualized risk calculation for emergency Cesarean delivery during labour based on maternal height, BMI, fetal HC and AC, with crucially both participants and care providers being blinded to the resultant risk prediction score. Labor, delivery and postnatal outcomes were ascertained. Calibration and receiver operator curves were generated to determine the predictive capacity for emergency Cesarean delivery of the Genesis risk prediction model in this cohort.

RESULTS

559 primiparous participants were enrolled from May 2017 to April 2019, of whom 142 (25 %) had an emergency Cesarean delivery during labour. Participants with a low predicted risk score (<10 %) had a mean predicted rate of 8% (+/- standard deviation of 2%) and a similarly low actual observed rate of Cesarean delivery (8%). Participants with a high predicted risk (>50 %) had a mean predicted Cesarean delivery rate of 64 % (+/- standard deviation of 9%) and also had a high actual observed Cesarean delivery rate (62 %). The calibration curve and receiver operating characteristic curve demonstrated that this validation study had comparable discriminatory power for emergency Cesarean delivery to that described in the original Genesis study. The Area Under the Curve (AUC) in Genesis was 0.69, whereas the AUC in RECIPE was 0.72, which reflects good predictive capacity of the risk prediction model.

CONCLUSION

The accuracy of the Genesis Cesarean delivery prediction tool is supported by this validation study.

Impact on perinatal health and cost-effectiveness of risk-based care in obstetrics: a before-after study

BACKGROUND

Obstetric health care relies on an adequate antepartum risk selection. Most guidelines used for risk stratification, however, do not assess absolute risks. In 2017, a prediction tool was implemented in a Dutch region. This tool combines first trimester prediction models with obstetric care paths tailored to the individual risk profile, enabling risk-based care.

OBJECTIVE

To assess impact and cost-effectiveness of risk-based care compared to care-as-usual in a general population.

METHODS

A before-after study was conducted using 2 multicenter prospective cohorts. The first cohort (2013-2015) received care-as-usual; the second cohort (2017-2018) received risk-based care. Health outcomes were (1) a composite of adverse perinatal outcomes and (2) maternal quality-adjusted life-years. Costs were estimated using a health care perspective from conception to 6 weeks after the due date. Mean costs per woman, cost differences between the 2 groups, and incremental cost effectiveness ratios were calculated. Sensitivity analyses were performed to evaluate the robustness of the findings.

RESULTS

In total 3425 women were included. In nulliparous women there was a significant reduction of perinatal adverse outcomes among the risk-based care group (adjusted odds ratio, 0.56; 95% confidence interval, 0.32-0.94), but not in multiparous women. Mean costs per pregnant woman were significantly lower for risk-based care (mean difference, -€2766; 95% confidence interval, -€3700 to -€1825). No differences in maternal quality of life, adjusted for baseline health, were observed.

CONCLUSION

In the Netherlands, risk-based care in nulliparous women was associated with improved perinatal outcomes as compared to care-as-usual. Furthermore, risk-based care was cost-effective compared to care-as-usual and resulted in lower health care costs.

The RECIPE study: reducing emergency Caesareans and improving the Primiparous experience: a blinded, prospective, observational study

Background

The RECIPE study aims to validate a risk prediction model for intrapartum caesarean delivery which has been developed by our group. The Genesis study was a prospective observational study carried out by the Perinatal Ireland Research Consortium across 7 clinical centres in Ireland between October 2012 and June 2015. Genesis investigated a range of maternal and fetal parameters in a prospective blinded study of 2336 singleton pregnancies between 39 + 0–41 + 0 weeks’ gestational age. This resulted in the development of a risk prediction model for Caesarean Delivery in nulliparous women at term. The RECIPE study now proposes to provide external validation of this risk prediction tool.

Methods

In order to externally validate the model, we aim to include a centre which was not involved in the original study. We propose a trial of risk-assignment for intrapartum caesarean amongst nulliparous women with a singleton pregnancy between 38 + 0 and 40 + 6 weeks’ gestational age who are planning a vaginal birth. Results of the risk prediction tool will be concealed from participants and from midwives and doctors providing labour care.. Participants will be invited for an ultrasound scan and delivery details will be collated postnatally. The principal aim of this study is to externally validate the risk prediction model. This prediction model holds the potential to accurately identify nulliparous women who are likely to achieve an uncomplicated vaginal birth and those at high prospect of requiring an unplanned caesarean delivery.

Discussion

Validation of the Genesis prediction model would enable more accurate counselling for women in the antenatal setting regarding their own likelihood of requiring an intrapartum Caesarean section. It would also provide valuable personalised information to women about the anticipated course of their own labour. We believe that this is an issue of national relevance that will impact positively on obstetric practice, and will positively empower women to make considered, personalised choices surrounding labour and delivery.

Prediction of vaginal birth after cesarean deliveries using machine learning

BACKGROUND

Efforts to reduce cesarean delivery rates to 12-15% have been undertaken worldwide. Special focus has been directed towards parturients who undergo a trial of labor after cesarean delivery to reduce the burden of repeated cesarean deliveries. Complication rates are lowest when a vaginal birth is achieved and highest when an unplanned cesarean delivery is performed, which emphasizes the need to assess, in advance, the likelihood of a successful vaginal birth after cesarean delivery. Vaginal birth after cesarean delivery calculators have been developed in different populations; however, some limitations to their implementation into clinical practice have been described. Machine-learning methods enable investigation of large-scale datasets with input combinations that traditional statistical analysis tools have difficulty processing.

OBJECTIVE

The aim of this study was to evaluate the feasibility of using machine-learning methods to predict a successful vaginal birth after cesarean delivery.

STUDY DESIGN

The electronic medical records of singleton, term labors during a 12-year period in a tertiary referral center were analyzed. With the use of gradient boosting, models that incorporated multiple maternal and fetal features were created to predict successful vaginal birth in parturients who undergo a trial of labor after cesarean delivery. One model was created to provide a personalized risk score for vaginal birth after cesarean delivery with the use of features that are available as early as the first antenatal visit; a second model was created that reassesses this score after features are added that are available only in proximity to delivery.

RESULTS

A cohort of 9888 parturients with 1 previous cesarean delivery was identified, of which 75.6% of parturients (n=7473) attempted a trial of labor, with a success rate of 88%. A machine-learning-based model to predict when vaginal delivery would be successful was developed. When features that are available at the first antenatal visit are used, the model showed a receiver operating characteristic curve with area under the curve of 0.745 (95% confidence interval, 0.728-0.762) that increased to 0.793 (95% confidence interval, 0.778-0.808) when features that are available in proximity to the delivery process were added. Additionally, for the later model, a risk stratification tool was built to allocate parturients into low-, medium-, and high-risk groups for failed trial of labor after cesarean delivery. The low- and medium-risk groups (42.4% and 25.6% of parturients, respectively) showed a success rate of 97.3% and 90.9%, respectively. The high-risk group (32.1%) had a vaginal delivery success rate of 73.3%. Application of the model to a cohort of parturients who elected a repeat cesarean delivery (n=2145) demonstrated that 31% of these parturients would have been allocated to the low- and medium-risk groups had a trial of labor been attempted.

CONCLUSION

Trial of labor after cesarean delivery is safe for most parturients. Success rates are high, even in a population with high rates of trial of labor after cesarean delivery. Application of a machine-learning algorithm to assign a personalized risk score for a successful vaginal birth after cesarean delivery may help in decision-making and contribute to a reduction in cesarean delivery rates. Parturient allocation to risk groups may help delivery process management.

A new model for screening for early-onset preeclampsia

BACKGROUND

Early identification of women with an increased risk for preeclampsia is of utmost importance to minimize adverse perinatal events. Models developed until now (mainly multiparametric algorithms) are thought to be overfitted to the derivation population, which may affect their reliability when applied to other populations. Options allowing adaptation to a variety of populations are needed.

OBJECTIVE

The objective of the study was to assess the performance of a first-trimester multivariate Gaussian distribution model including maternal characteristics and biophysical/biochemical parameters for screening of early-onset preeclampsia (delivery <34 weeks of gestation) in a routine care low-risk setting.

STUDY DESIGN

Early-onset preeclampsia screening was undertaken in a prospective cohort of singleton pregnancies undergoing routine first-trimester screening (8 weeks 0/7 days to 13 weeks 6/7 days of gestation), mainly using a 2-step scheme, at 2 hospitals from March 2014 to September 2017. A multivariate Gaussian distribution model including maternal characteristics (a priori risk), serum pregnancy-associated plasma protein-A and placental growth factor assessed at 8 weeks 0/7 days to 13 weeks 6/7 days and mean arterial pressure and uterine artery pulsatility index measured at 11.0-13.6 weeks was used.

RESULTS

A total of 7908 pregnancies underwent examination, of which 6893 were included in the analysis. Incidence of global preeclampsia was 2.3% (n = 161), while of early-onset preeclampsia was 0.2% (n = 17). The combination of maternal characteristics, biophysical parameters, and placental growth factor showed the best detection rate, which was 59% for a 5% false-positive rate and 94% for a 10% false-positive rate (area under the curve, 0.96, 95% confidence interval, 0.94-0.98). The addition of placental growth factor to biophysical markers significantly improved the detection rate from 59% to 94%.

CONCLUSION

The multivariate Gaussian distribution model including maternal factors, early placental growth factor determination (at 8 weeks 0/7 days to 13 weeks 6/7 days), and biophysical variables (mean arterial pressure and uterine artery pulsatility index) at 11 weeks 0/7 days to 13 weeks 6/7 days is a feasible tool for early-onset preeclampsia screening in the routine care setting. Performance of this model should be compared with predicting models based on regression analysis.

Early pregnancy prediction of gestational diabetes mellitus risk using prenatal screening biomarkers in nulliparous women

AIMS

To evaluate the clinical utility of first and second trimester prenatal screening biomarkers for early pregnancy prediction of gestational diabetes mellitus (GDM) risk in nulliparous women.

METHODS

We conducted a population-based cohort study of nulliparous women participating in the California Prenatal Screening Program from 2009 to 2011 (n = 105,379). GDM was ascertained from hospital discharge records or birth certificates. Models including maternal characteristics and prenatal screening biomarkers were developed and validated. Risk stratification and reclassification were performed to assess clinical utility of the biomarkers.

RESULTS

Decreased levels of first trimester pregnancy-associated plasma protein A (PAPP-A) and increased levels of second trimester unconjugated estriol (uE₃) and dimeric inhibin A (INH) were associated with GDM. The addition of PAPP-A only and PAPP-A, uE₃, and INH to maternal characteristics resulted in small, yet significant, increases in area under the receiver operating characteristic curve (AUC) (maternal characteristics only: AUC 0.714 (95% CI 0.703-0.724), maternal characteristics + PAPP-A: AUC 0.718 (95% CI 0.707-0.728), maternal characteristics + PAPP-A, uE₃, and INH: AUC 0.722 (0.712-0.733)); however, no net improvement in classification was observed.

CONCLUSIONS

PAPP-A, uE₃, and INH have limited clinical utility for prediction of GDM risk in nulliparous women. Utility of other readily accessible clinical biomarkers in predicting GDM risk warrants further investigation.

The Unrealised Potential for Predicting Pregnancy Complications in Women with Gestational Diabetes: A Systematic Review and Critical Appraisal

Gestational diabetes (GDM) increases the risk of pregnancy complications. However, these risks are not the same for all affected women and may be mediated by inter-related factors including ethnicity, body mass index and gestational weight gain. This study was conducted to identify, compare, and critically appraise prognostic prediction models for pregnancy complications in women with gestational diabetes (GDM). A systematic review of prognostic prediction models for pregnancy complications in women with GDM was conducted. Critical appraisal was conducted using the prediction model risk of bias assessment tool (PROBAST). Five prediction modelling studies were identified, from which ten prognostic models primarily intended to predict pregnancy complications related to GDM were developed. While the composition of the pregnancy complications predicted varied, the delivery of a large-for-gestational age neonate was the subject of prediction in four studies, either alone or as a component of a composite outcome. Glycaemic measures and body mass index were selected as predictors in four studies. Model evaluation was limited to internal validation in four studies and not reported in the fifth. Performance was inadequately reported with no useful measures of calibration nor formal evaluation of clinical usefulness. Critical appraisal using PROBAST revealed that all studies were subject to a high risk of bias overall driven by methodologic limitations in statistical analysis. This review demonstrates the potential for prediction models to provide an individualised absolute risk of pregnancy complications for women affected by GDM. However, at present, a lack of external validation and high risk of bias limit clinical application. Future model development and validation should utilise the latest methodological advances in prediction modelling to achieve the evolution required to create a useful clinical tool. Such a tool may enhance clinical decision-making and support a risk-stratified approach to the management of GDM. Systematic review registration: PROSPERO CRD42019115223.

Systematic review of prediction models for gestational hypertension and preeclampsia

INTRODUCTION

Prediction models for gestational hypertension and preeclampsia have been developed with data and assumptions from developed countries. Their suitability and application for low resource settings have not been tested. This review aimed to identify and assess the methodological quality of prediction models for gestational hypertension and pre-eclampsia with reference to their application in low resource settings.

METHODS

Using combinations of keywords for gestational hypertension, preeclampsia and prediction models seven databases were searched to identify prediction models developed with maternal data obtained before 20 weeks of pregnancy and including at least three predictors (Prospero registration CRD 42017078786). Prediction model characteristics and performance measures were extracted using the CHARMS, STROBE and TRIPOD checklists. The National Institute of Health quality assessment tools for observational cohort and cross-sectional studies were used for study quality appraisal.

RESULTS

We retrieved 8,309 articles out of which 40 articles were eligible for review. Seventy-seven percent of all the prediction models combined biomarkers with maternal clinical characteristics. Biomarkers used as predictors in most models were pregnancy associated plasma protein-A (PAPP-A) and placental growth factor (PlGF). Only five studies were conducted in a low-and middle income country.

CONCLUSIONS

Most of the studies evaluated did not completely follow the CHARMS, TRIPOD and STROBE guidelines in prediction model development and reporting. Adherence to these guidelines will improve prediction modelling studies and subsequent application of prediction models in clinical practice. Prediction models using maternal characteristics, with good discrimination and calibration, should be externally validated for use in low and middle income countries where biomarker assays are not routinely available.

Population screening for gestational hypertensive disorders using maternal, fetal and placental characteristics: A population-based prospective cohort study

Objective

To determine screening performance of maternal, fetal and placental characteristics for selecting pregnancies at risk of gestational hypertension and preeclampsia in a low‐risk multi‐ethnic population.

Method

In a prospective population‐based cohort among 7124 pregnant women, we collected maternal characteristics including body mass index, ethnicity, parity, smoking and blood pressure in early‐pregnancy. Fetal characteristics included second and third trimester estimated fetal weight and sex determined by ultrasound. Placental characteristics included first and second trimester placental growth factor concentrations and second and third trimester uterine artery resistance indices.

Results

Maternal characteristics provided the best screening result for gestational hypertension (area‐under‐the‐curve [AUC] 0.79 [95% Confidence interval {CI} 0.76‐0.81]) with 40% sensitivity at 90% specificity. For preeclampsia, the maternal characteristics model led to a screening performance of AUC 0.74 (95% CI 0.70‐0.78) with 33% sensitivity at 90% specificity. Addition of second and third trimester placental ultrasound characteristics only improved screening performance for preeclampsia (AUC 0.78 [95% CI 0.75‐0.82], with 48% sensitivity at 90% specificity).

Conclusion

Routinely measured maternal characteristics, known at the start of pregnancy, can be used in screening for pregnancies at risk of gestational hypertension or preeclampsia within a low‐risk multi‐ethnic population. Addition of combined second and third trimester placental ultrasound characteristics only improved screening for preeclampsia.

The impact of periconceptional maternal lifestyle on clinical features and biomarkers of placental development and function: a systematic review

BACKGROUND

Worldwide, placenta-related complications contribute to adverse pregnancy outcomes, such as pre-eclampsia, fetal growth restriction and preterm birth, with implications for the future health of mothers and offspring. The placenta develops in the periconception period and forms the interface between mother and embryo/fetus. An unhealthy periconceptional maternal lifestyle, such as smoking, alcohol and under- and over-nutrition, can detrimentally influence placental development and function.

OBJECTIVE AND RATIONALE

The impact of maternal lifestyle on placental health is largely unknown. Therefore, we aim to summarize the evidence of the impact of periconceptional maternal lifestyle on clinical features and biomarkers of placental development and function throughout pregnancy.

SEARCH METHODS

A comprehensive search in Medline, Embase, Pubmed, The Cochrane Library Web of Science and Google Scholar was conducted. The search strategy included keywords related to the maternal lifestyle, smoking, alcohol, caffeine, nutrition (including folic acid supplement intake) and body weight. For placental markers throughout pregnancy, keywords related to ultrasound imaging, serum biomarkers and histological characteristics were used. We included randomized controlled trials and observational studies published between January 2000 and March 2017 and restricted the analysis to singleton pregnancies and maternal periconceptional lifestyle. Methodological quality was scored using the ErasmusAGE tool. A protocol of this systematic review has been registered in PROSPERO International prospective register of systematic reviews (PROSPERO 2016:CRD42016045596).

OUTCOMES

Of 2593 unique citations found, 82 studies were included. The median quality score was 5 (range: 0-10). The findings revealed that maternal smoking was associated with lower first-trimester placental vascularization flow indices, higher second- and third-trimester resistance of the uterine and umbilical arteries and lower resistance of the middle cerebral artery. Although a negative impact of smoking on placental weight was expected, this was less clear. Alcohol use was associated with a lower placental weight. One study described higher second- and third-trimester placental growth factor (PlGF) levels after periconceptional alcohol use. None of the studies looked at caffeine intake. Adequate nutrition in the first trimester, periconceptional folic acid supplement intake and strong adherence to a Mediterranean diet, were all associated with a lower resistance of the uterine and umbilical arteries in the second and third trimester. A low caloric intake resulted in a lower placental weight, length, breadth, thickness, area and volume. Higher maternal body weight was associated with a larger placenta measured by ultrasound in the second and third trimester of pregnancy or weighed at birth. In addition, higher maternal body weight was associated with decreased PlGF-levels.

WIDER IMPLICATIONS

Evidence of the impact of periconceptional maternal lifestyle on placental health was demonstrated. However, due to poorly defined lifestyle exposures and time windows of investigation, unstandardized measurements of placenta-related outcomes and small sample sizes of the included studies, a cautious interpretation of the effect estimates is indicated. We suggest that future research should focus more on physiological consequences of unhealthy lifestyle during the critical periconception window. Moreover, we foresee that new evidence will support the development of lifestyle interventions to improve the health of mothers and their offspring from the earliest moment in life.

Modifiable risk factors for spontaneous preterm birth in nulliparous women: a prospective study

Background Spontaneous preterm birth (sPTB) has a detrimental impact on perinatal outcome, as well as on the resources of health systems in high-income countries. Thus, the objective of the current study was to determine the incidence of modifiable risk factors in pregnancy and their impact on the rate of sPTB. Methods All nulliparous pregnant women, in singleton pregnancies, with free medical and obstetric history, were eligible to participate in this study. The primary outcome of interest was the incidence of specific modifiable risk factors for sPTB. The correlations between these risk factors and sPTB were also investigated. Results Overall, 2027 women were eligible for the study and agreed to participate. The incidence of sPTB was 4.9%; 25.5% (n = 518) of the participants were in extreme maternal age (<20 or >35 years), 34.5% (n = 701) had an abnormal body mass index (BMI) (<18.5 or ≥25 kg/m2), 4.4% (n = 89) reported use of assisted reproductive technology (ART) and 10.9% (n = 220) reported themselves as smokers in pregnancy. In the multivariate analysis, sPTB was significantly correlated with ART conception [odds ratio (OR): 2.494, 95% confidence interval (CI): 1.196-5.199]. Conclusion Approximately one in 20 primiparous pregnant women in the study had a sPTB. The study population included a high percentage of women of advanced maternal age, with abnormal BMI and smokers, but these characteristics did not affect the incidence of sPTB. On the other hand, conception following ART increased two-fold the risk of sPTB.

Calibration: the Achilles heel of predictive analytics

BACKGROUND

The assessment of calibration performance of risk prediction models based on regression or more flexible machine learning algorithms receives little attention.

MAIN TEXT

Herein, we argue that this needs to change immediately because poorly calibrated algorithms can be misleading and potentially harmful for clinical decision-making. We summarize how to avoid poor calibration at algorithm development and how to assess calibration at algorithm validation, emphasizing balance between model complexity and the available sample size. At external validation, calibration curves require sufficiently large samples. Algorithm updating should be considered for appropriate support of clinical practice.

CONCLUSION

Efforts are required to avoid poor calibration when developing prediction models, to evaluate calibration when validating models, and to update models when indicated. The ultimate aim is to optimize the utility of predictive analytics for shared decision-making and patient counseling.

Implementing a Preeclampsia Prediction Model in Obstetrics: Cutoff Determination and Health Care Professionals' Adherence

Background. Despite improved management, preeclampsia remains an important cause of maternal and neonatal mortality and morbidity. Low-dose aspirin (LDA) lowers the risk of preeclampsia. Although several guidelines recommend LDA prophylaxis in women at increased risk, they disagree about the definition of high risk. Recently, an externally validated prediction model for preeclampsia was implemented in a Dutch region combined with risk-based obstetric care paths. Objectives. To demonstrate the selection of a risk threshold and to evaluate the adherence of obstetric health care professionals to the prediction tool. Study Design. Using a survey (n = 136) and structured meetings among health care professionals, possible cutoff values at which LDA should be discussed were proposed. The prediction model, with chosen cutoff and corresponding risk-based care paths, was embedded in an online tool. Subsequently, a prospective multicenter cohort study (n = 850) was performed to analyze the adherence of health care professionals. Patient questionnaires, linked to the individual risk profiles calculated by the online tool, were used to evaluate adherence. Results. Health care professionals agreed upon employing a tool with a high detection rate (cutoff: 3.0%; sensitivity 75%, specificity 64%) followed by shared decision between patients and health care professionals on LDA prophylaxis. Of the 850 enrolled women, 364 women had an increased risk of preeclampsia. LDA was discussed with 273 of these women, resulting in an 81% adherence rate. Conclusion. Consensus regarding a suitable risk cutoff threshold was reached. The adherence to this recommendation was 81%, indicating adequate implementation.

Predictive analytics in health care: how can we know it works?

There is increasing awareness that the methodology and findings of research should be transparent. This includes studies using artificial intelligence to develop predictive algorithms that make individualized diagnostic or prognostic risk predictions. We argue that it is paramount to make the algorithm behind any prediction publicly available. This allows independent external validation, assessment of performance heterogeneity across settings and over time, and algorithm refinement or updating. Online calculators and apps may aid uptake if accompanied with sufficient information. For algorithms based on "black box" machine learning methods, software for algorithm implementation is a must. Hiding algorithms for commercial exploitation is unethical, because there is no possibility to assess whether algorithms work as advertised or to monitor when and how algorithms are updated. Journals and funders should demand maximal transparency for publications on predictive algorithms, and clinical guidelines should only recommend publicly available algorithms.

FIRST TRIMESTER SCREENING FOR PREECLAMPSIA - A SYSTEMATIC REVIEW

Objective: To increase the detection rate of preterm preeclampsia (PE) first trimester combined screening tests are being developed. The aim of this review is to create an overview of the currently investigated screening markers, algorithms and their validations.Methods: Comprehensive review of the literature concerning first trimester screening for PEResults and conclusions: Studies investigating a total of 160 biochemical, 6 biophysical and 14 ultrasound markers could be identified. Of the 21 algorithms published, mainly the algorithm published by the Fetal Medicine Foundation London has been validated. This algorithm performes significantly better than screening by anamnestic risk factors only.

Prognostic models for adverse pregnancy outcomes in low-income and middle-income countries: a systematic review

INTRODUCTION

Ninety-nine per cent of all maternal and neonatal deaths occur in low-income and middle-income countries (LMIC). Prognostic models can provide standardised risk assessment to guide clinical management and can be vital to reduce and prevent maternal and perinatal mortality and morbidity. This review provides a comprehensive summary of prognostic models for adverse maternal and perinatal outcomes developed and/or validated in LMIC.

METHODS

A systematic search in four databases (PubMed/Medline, EMBASE, Global Health Library and The Cochrane Library) was conducted from inception (1970) up to 2 May 2018. Risk of bias was assessed with the PROBAST tool and narratively summarised.

RESULTS

1741 articles were screened and 21 prognostic models identified. Seventeen models focused on maternal outcomes and four on perinatal outcomes, of which hypertensive disorders of pregnancy (n=9) and perinatal death including stillbirth (n=4) was most reported. Only one model was externally validated. Thirty different predictors were used to develop the models. Risk of bias varied across studies, with the item 'quality of analysis' performing the least.

CONCLUSION

Prognostic models can be easy to use, informative and low cost with great potential to improve maternal and neonatal health in LMIC settings. However, the number of prognostic models developed or validated in LMIC settings is low and mirrors the 10/90 gap in which only 10% of resources are dedicated to 90% of the global disease burden. External validation of existing models developed in both LMIC and high-income countries instead of developing new models should be encouraged.

PROSPERO REGISTRATION NUMBER

CRD42017058044.

Prediction model of postnatal renal function in fetuses with lower urinary tract obstruction (LUTO)-Development and internal validation

OBJECTIVE

To develop a prediction model of postnatal renal function in fetuses with lower urinary tract obstruction (LUTO) based on fetal ultrasound parameters and amniotic fluid volume.

METHODS

Retrospective nationwide cohort study of fetuses with postnatally confirmed LUTO and known eGFR. Fetuses treated with fetal interventions such as vesico-amniotic shunting or cystoscopy were excluded. Logistic regression analysis was used to identify prognostic ultrasound variables with respect to renal outcome following multiple imputation of missing data. On the basis of these fetal renal parameters and amniotic fluid volume, a model was developed to predict postnatal renal function in fetuses with LUTO. The main study outcome was an eGFR less than 60 mL/min * 1.73 m² based on the creatinine nadir during the first year following diagnosis. Model performance was evaluated by receiver operator characteristic (ROC) curve analysis, calibration plots, and bootstrapping.

RESULTS

Hundred one fetuses with a confirmed diagnosis of LUTO were included, eGFR less than 60 was observed in 40 (39.6%) of them. Variables predicting an eGFR less than 60 mL/min * 1.73m² included the following sonographic parameters: hyperechogenicity of the renal cortex and abnormal amniotic fluid volume. The model showed fair discrimination, with an area under the ROC curve of 0.70 (95% confidence interval, 0.59-0.81, 0.66 after bootstrapping) and was overall well-calibrated.

CONCLUSION

This study shows that a prediction model incorporating ultrasound parameters such as cortical appearance and abnormal amniotic fluid volume can fairly discriminate an eGFR above or below 60 mL/min * 1.73m² . This clinical information can be used in identifying fetuses eligible for prenatal interventions and improve counseling of parents.

Methodological standards for the development and evaluation of clinical prediction rules: a review of the literature

Clinical prediction rules (CPRs) that predict the absolute risk of a clinical condition or future outcome for individual patients are abundant in the medical literature; however, systematic reviews have demonstrated shortcomings in the methodological quality and reporting of prediction studies. To maximise the potential and clinical usefulness of CPRs, they must be rigorously developed and validated, and their impact on clinical practice and patient outcomes must be evaluated. This review aims to present a comprehensive overview of the stages involved in the development, validation and evaluation of CPRs, and to describe in detail the methodological standards required at each stage, illustrated with examples where appropriate. Important features of the study design, statistical analysis, modelling strategy, data collection, performance assessment, CPR presentation and reporting are discussed, in addition to other, often overlooked aspects such as the acceptability, cost-effectiveness and longer-term implementation of CPRs, and their comparison with clinical judgement. Although the development and evaluation of a robust, clinically useful CPR is anything but straightforward, adherence to the plethora of methodological standards, recommendations and frameworks at each stage will assist in the development of a rigorous CPR that has the potential to contribute usefully to clinical practice and decision-making and have a positive impact on patient care.

The use of rigorous methods was strongly warranted among prognostic prediction models for obstetric care

OBJECTIVE

The objective of the study was to examine methodological characteristics about the design and conduct in prognostic prediction models used for obstetric care.

STUDY DESIGN AND SETTING

We searched PubMed for studies on prognostic prediction models for obstetric care, published in top general medicine or major specialty journals between January 2011 and February 2018. Teams of method-trained investigators independently screened titles and abstracts and collected data using a prespecified, pilot-tested, structured questionnaire.

RESULTS

In total, 91 studies were eligible, of which two were published in top general medicine journals, 20 (22.0%) involved an epidemiologist or statistician, 18 (19.4%) published study protocols, 53 (58.2%) did not include any model validation, 20 (22.0%) did not clearly state the intended timing of use, 23 (25.3%) had no eligibility criteria, 15 (16.5%) did not use clear criteria for ascertaining outcome, and 69 (75.82%) did not apply blinding to outcome assessment. Among those models, 11 (12.1%) included participants fewer than 200 events, 41 (48.8%) had fewer than 100 events, and 19 (24.7%) had fewer than 10 events per variable.

CONCLUSION

The prognostic prediction models have important limitations in design and conduct. Substantial efforts are needed to strengthen the production of reliable prognostic prediction models for obstetric care.

Prediction of pre-eclampsia: review of reviews

OBJECTIVE

Primary studies and systematic reviews provide estimates of varying accuracy for different factors in the prediction of pre-eclampsia. The aim of this study was to review published systematic reviews to collate evidence on the ability of available tests to predict pre-eclampsia, to identify high-value avenues for future research and to minimize future research waste in this field.

METHODS

MEDLINE, EMBASE and The Cochrane Library including DARE (Database of Abstracts of Reviews of Effects) databases, from database inception to March 2017, and bibliographies of relevant articles were searched, without language restrictions, for systematic reviews and meta-analyses on the prediction of pre-eclampsia. The quality of the included reviews was assessed using the AMSTAR tool and a modified version of the QUIPS tool. We evaluated the comprehensiveness of search, sample size, tests and outcomes evaluated, data synthesis methods, predictive ability estimates, risk of bias related to the population studied, measurement of predictors and outcomes, study attrition and adjustment for confounding.

RESULTS

From 2444 citations identified, 126 reviews were included, reporting on over 90 predictors and 52 prediction models for pre-eclampsia. Around a third (n = 37 (29.4%)) of all reviews investigated solely biochemical markers for predicting pre-eclampsia, 31 (24.6%) investigated genetic associations with pre-eclampsia, 46 (36.5%) reported on clinical characteristics, four (3.2%) evaluated only ultrasound markers and six (4.8%) studied a combination of tests; two (1.6%) additional reviews evaluated primary studies investigating any screening test for pre-eclampsia. Reviews included between two and 265 primary studies, including up to 25 356 688 women in the largest review. Only approximately half (n = 67 (53.2%)) of the reviews assessed the quality of the included studies. There was a high risk of bias in many of the included reviews, particularly in relation to population representativeness and study attrition. Over 80% (n = 106 (84.1%)) summarized the findings using meta-analysis. Thirty-two (25.4%) studies lacked a formal statement on funding. The predictors with the best test performance were body mass index (BMI) > 35 kg/m² , with a specificity of 92% (95% CI, 89-95%) and a sensitivity of 21% (95% CI, 12-31%); BMI > 25 kg/m² , with a specificity of 73% (95% CI, 64-83%) and a sensitivity of 47% (95% CI, 33-61%); first-trimester uterine artery pulsatility index or resistance index > 90^th centile (specificity 93% (95% CI, 90-96%) and sensitivity 26% (95% CI, 23-31%)); placental growth factor (specificity 89% (95% CI, 89-89%) and sensitivity 65% (95% CI, 63-67%)); and placental protein 13 (specificity 88% (95% CI, 87-89%) and sensitivity 37% (95% CI, 33-41%)). No single marker had a test performance suitable for routine clinical use. Models combining markers showed promise, but none had undergone external validation.

CONCLUSIONS

This review of reviews calls into question the need for further aggregate meta-analysis in this area given the large number of published reviews subject to the common limitations of primary predictive studies. Prospective, well-designed studies of predictive markers, preferably randomized intervention studies, and combined through individual-patient data meta-analysis are needed to develop and validate new prediction models to facilitate the prediction of pre-eclampsia and minimize further research waste in this field. Copyright © 2018 ISUOG. Published by John Wiley & Sons Ltd.

The association between first trimester AFP to PAPP-A ratio and placentally-related adverse pregnancy outcome

INTRODUCTION

Low maternal serum levels of pregnancy-associated plasma protein A (PAPP-A) measured in the first trimester and high levels of alpha fetoprotein (AFP) measured in the second trimester have been associated with adverse pregnancy outcomes reflective of placental insufficiency, and there is a synergistic relationship between the two. We investigated the utility as a screening test of a simple ratio of maternal serum AFP to PAPP-A (AFP:PAPP-A) measured in the first trimester.

METHODS

We studied 4057 nulliparous women with a singleton pregnancy from the Pregnancy Outcome Prediction (POP) study. We studied the predictive ability for adverse outcome of the AFP:PAPP-A ratio measured in the first trimester with and without correction for maternal weight and gestational age at measurement. We compared the AFP:PAPP-A ratio with corrected AFP and PAPP-A on their own and in combination.

RESULTS

An AFP:PAPP-A ratio >10 was associated with placentally-related adverse outcomes, including fetal growth restriction (risk ratio (RR) 3.74, 95% confidence interval (CI) 2.30-6.09), severe preeclampsia (RR 2.12, 95% CI 1.39-3.25) and stillbirth (RR 5.05, 95% CI 1.48-17.18). The ratio performed favorably in predicting adverse pregnancy outcomes when compared with corrected measurements of either AFP or PAPP-A, and was equivalent to a model combining the two. Its predictive ability was not affected by correction for maternal weight or gestational age at measurement.

DISCUSSION

An elevated maternal AFP:PAPP-A ratio in the first trimester is associated with placentally-related adverse outcomes in a cohort of unselected nulliparous women.

External validation of prognostic models for preeclampsia in a Dutch multicenter prospective cohort

OBJECTIVE

To perform an external validation of all published prognostic models for first-trimester prediction of the risk of developing preeclampsia (PE).

METHODS

Women <14 weeks of pregnancy were recruited in the Netherlands. All systematically identified prognostic models for PE that contained predictors commonly available were eligible for external validation.

RESULTS

3,736 women were included; 87 (2.3%) developed PE. Calibration was poor due to overestimation. Discrimination of 9 models for LO-PE ranged from 0.58 to 0.71 and of 9 models for all PE from 0.55 to 0.75.

CONCLUSION

Only a few easily applicable prognostic models for all PE showed discrimination above 0.70, which is considered an acceptable performance.

Untapped potential of multicenter studies: a review of cardiovascular risk prediction models revealed inappropriate analyses and wide variation in reporting

BACKGROUND

Clinical prediction models are often constructed using multicenter databases. Such a data structure poses additional challenges for statistical analysis (clustered data) but offers opportunities for model generalizability to a broad range of centers. The purpose of this study was to describe properties, analysis, and reporting of multicenter studies in the Tufts PACE Clinical Prediction Model Registry and to illustrate consequences of common design and analyses choices.

METHODS

Fifty randomly selected studies that are included in the Tufts registry as multicenter and published after 2000 underwent full-text screening. Simulated examples illustrate some key concepts relevant to multicenter prediction research.

RESULTS

Multicenter studies differed widely in the number of participating centers (range 2 to 5473). Thirty-nine of 50 studies ignored the multicenter nature of data in the statistical analysis. In the others, clustering was resolved by developing the model on only one center, using mixed effects or stratified regression, or by using center-level characteristics as predictors. Twenty-three of 50 studies did not describe the clinical settings or type of centers from which data was obtained. Four of 50 studies discussed neither generalizability nor external validity of the developed model.

CONCLUSIONS

Regression methods and validation strategies tailored to multicenter studies are underutilized. Reporting on generalizability and potential external validity of the model lacks transparency. Hence, multicenter prediction research has untapped potential.

REGISTRATION

This review was not registered.

A simple model to predict the complicated operative vaginal deliveries using vacuum or forceps

BACKGROUND

Complicated operative vaginal deliveries are associated with high neonatal morbidity and maternal trauma, especially if the procedure is unsuccessful and a cesarean delivery is needed. The decision to perform an operative vaginal delivery has traditionally been based on a subjective assessment by digital vaginal examination combined with the clinical expertise of the obstetrician. Currently there is no method for objectively quantifying the likelihood of successful delivery. Intrapartum ultrasound has been introduced in clinical practice to help predict the progression and final method of delivery.

OBJECTIVE

The aim of this study was to compare predictive models for identifying complicated operative vaginal deliveries (vacuum or forceps) based on intrapartum transperineal ultrasound in nulliparous women.

STUDY DESIGN

We performed a prospective cohort study in nulliparous women at term with singleton pregnancies and full dilatation who underwent intrapartum transperineal ultrasound evaluation prior to operative vaginal delivery. Managing obstetricians were blinded to the ultrasound data. Intrapartum transperineal ultrasound (angle of progression, progression distance, and midline angle) was performed immediately before instrument application, both at rest and concurrently with pushing. Intrapartum evaluation of fetal biometric parameters (estimated fetal weight, head circumference, and biparietal diameter) was also carried out. An operative vaginal delivery was classified as complicated when 1 or more of the following complications occurred: ≥3 tractions needed; third- to fourth-degree perineal tear; severe bleeding during episiotomy repair (decrease of ≥2.5 g/dL in the hemoglobin level); or significant traumatic neonatal lesion (subdural-intracerebral hemorrhage, epicranial subaponeurotic hemorrhage, skeletal injuries, injuries to spine and spinal cord, or peripheral and cranial nerve injuries). Six predictive models were evaluated (information available in Table 2).

RESULTS

We recruited 84 nulliparous patients, of whom 5 were excluded because of the difficulty of adequately evaluating the biparietal diameter and head circumference. A total of 79 nulliparous patients were studied (47 vacuum deliveries, 32 forceps deliveries) with 13 cases in the occiput-posterior position. We identified 31 cases of complicated operative vaginal deliveries (19 vacuum deliveries and 12 forceps deliveries). No differences were identified in obstetric, neonatal, or intrapartum characteristics between the 2 study groups (operative uncomplicated vaginal delivery vs operative complicated vaginal delivery), with the following exceptions: estimated fetal weight (3243 ± 425 g vs 3565 ± 330 g; P = .001), biparietal diameter (93.2 ± 2.1 vs 95.2 ± 2.3 mm; P = .001), head circumference (336 ± 12 vs 348 ± 6.4 mm; P = .001), sex (female 62.5% vs 29.0%; P = .010), newborn weight (3258 ± 472 g vs 3499 ± 383 g; P = .027), and number of tractions (median, interquartile range) (1 [1-2] vs 4 [3-5]; P < .0005). To predict complicated operative deliveries, all 6 of the studied models presented an area under the receiver-operating characteristics curve between 0.863 and 0.876 (95% confidence intervals, 0.775-0.950 and 0.790-0.963; P < .0005). The results of the study met the criteria of interpretability and parsimony (simplicity), allowing us to identify a binary logistic regression model based on the angle of progression and head circumference; this model has an area under the receiver-operating characteristics curve of 0.876 (95% confidence interval, 0.790-0.963; P < .0005) and a calibration slope B of 0.984 (95% confidence interval, 0.0.726-1.243; P < .0005).

CONCLUSION

The combination of the angle of progression and the head circumference can predict 87% of complicated operative vaginal deliveries and can be performed in the delivery room.

PROBAST: A Tool to Assess Risk of Bias and Applicability of Prediction Model Studies: Explanation and Elaboration

Prediction models in health care use predictors to estimate for an individual the probability that a condition or disease is already present (diagnostic model) or will occur in the future (prognostic model). Publications on prediction models have become more common in recent years, and competing prediction models frequently exist for the same outcome or target population. Health care providers, guideline developers, and policymakers are often unsure which model to use or recommend, and in which persons or settings. Hence, systematic reviews of these studies are increasingly demanded, required, and performed. A key part of a systematic review of prediction models is examination of risk of bias and applicability to the intended population and setting. To help reviewers with this process, the authors developed PROBAST (Prediction model Risk Of Bias ASsessment Tool) for studies developing, validating, or updating (for example, extending) prediction models, both diagnostic and prognostic. PROBAST was developed through a consensus process involving a group of experts in the field. It includes 20 signaling questions across 4 domains (participants, predictors, outcome, and analysis). This explanation and elaboration document describes the rationale for including each domain and signaling question and guides researchers, reviewers, readers, and guideline developers in how to use them to assess risk of bias and applicability concerns. All concepts are illustrated with published examples across different topics. The latest version of the PROBAST checklist, accompanying documents, and filled-in examples can be downloaded from www.probast.org.

Predictions of Preterm Birth from Early Pregnancy Characteristics: Born in Guangzhou Cohort Study

Preterm birth (PTB, <37 weeks) is the leading cause of death in children <5 years of age. Early risk prediction for PTB would enable early monitoring and intervention. However, such prediction models have been rarely reported, especially in low- and middle-income areas. We used data on a number of easily accessible predictors during early pregnancy from 9044 women in Born in Guangzhou Cohort Study, China to generate prediction models for overall PTB and spontaneous, iatrogenic, late (34–36 weeks), and early (<34 weeks) PTB. Models were constructed using the Cox proportional hazard model, and their performance was evaluated by Harrell’s c and D statistics and calibration plot. We further performed a systematic review to identify published models and validated them in our population. Our new prediction models had moderate discrimination, with Harrell’s c statistics ranging from 0.60–0.66 for overall and subtypes of PTB. Significant predictors included maternal age, height, history of preterm delivery, amount of vaginal bleeding, folic acid intake before pregnancy, and passive smoking during pregnancy. Calibration plots showed good fit for all models except for early PTB. We validated three published models, all of which were from studies conducted in high-income countries; the area under receiver operating characteristic for these models ranged from 0.50 to 0.56. Based on early pregnancy characteristics, our models have moderate predictive ability for PTB. Future studies should consider inclusion of laboratory markers for the prediction of PTB.

External Validation and Clinical Usefulness of First Trimester Prediction Models for the Risk of Preeclampsia: A Prospective Cohort Study

INTRODUCTION

This study assessed the external validity of all published first trimester prediction models for the risk of preeclampsia (PE) based on routinely collected maternal predictors. Moreover, the potential utility of the best-performing models in clinical practice was evaluated.

MATERIAL AND METHODS

Ten prediction models were systematically selected from the literature. We performed a multicenter prospective cohort study in the Netherlands between July 1, 2013, and December 31, 2015. Eligible pregnant women completed a web-based questionnaire before 16 weeks' gestation. The outcome PE was established using postpartum questionnaires and medical records. Predictive performance of each model was assessed by means of discrimination (c-statistic) and a calibration plot. Clinical usefulness was evaluated by means of decision curve analysis and by calculating the potential impact at different risk thresholds.

RESULTS

The validation cohort contained 2,614 women of whom 76 developed PE (2.9%). Five models showed moderate discriminative performance with c-statistics ranging from 0.73 to 0.77. Adequate calibration was obtained after refitting. The best models were clinically useful over a small range of predicted probabilities.

DISCUSSION

Five of the ten included first trimester prediction models for PE showed moderate predictive performance. The best models may provide more benefit compared to risk selection as used in current guidelines.

Prediction models for the risk of spontaneous preterm birth based on maternal characteristics: a systematic review and independent external validation

Introduction

Prediction models may contribute to personalized risk‐based management of women at high risk of spontaneous preterm delivery. Although prediction models are published frequently, often with promising results, external validation generally is lacking. We performed a systematic review of prediction models for the risk of spontaneous preterm birth based on routine clinical parameters. Additionally, we externally validated and evaluated the clinical potential of the models.

Material and methods

Prediction models based on routinely collected maternal parameters obtainable during first 16 weeks of gestation were eligible for selection. Risk of bias was assessed according to the CHARMS guidelines. We validated the selected models in a Dutch multicenter prospective cohort study comprising 2614 unselected pregnant women. Information on predictors was obtained by a web‐based questionnaire. Predictive performance of the models was quantified by the area under the receiver operating characteristic curve (AUC) and calibration plots for the outcomes spontaneous preterm birth <37 weeks and <34 weeks of gestation. Clinical value was evaluated by means of decision curve analysis and calculating classification accuracy for different risk thresholds.

Results

Four studies describing five prediction models fulfilled the eligibility criteria. Risk of bias assessment revealed a moderate to high risk of bias in three studies. The AUC of the models ranged from 0.54 to 0.67 and from 0.56 to 0.70 for the outcomes spontaneous preterm birth <37 weeks and <34 weeks of gestation, respectively. A subanalysis showed that the models discriminated poorly (AUC 0.51–0.56) for nulliparous women. Although we recalibrated the models, two models retained evidence of overfitting. The decision curve analysis showed low clinical benefit for the best performing models.

Conclusions

This review revealed several reporting and methodological shortcomings of published prediction models for spontaneous preterm birth. Our external validation study indicated that none of the models had the ability to predict spontaneous preterm birth adequately in our population. Further improvement of prediction models, using recent knowledge about both model development and potential risk factors, is necessary to provide an added value in personalized risk assessment of spontaneous preterm birth.

Implementation and Effects of Risk-Dependent Obstetric Care in the Netherlands (Expect Study II): Protocol for an Impact Study

BACKGROUND

Recently, validated risk models predicting adverse obstetric outcomes combined with risk-dependent care paths have been made available for early antenatal care in the southeastern part of the Netherlands. This study will evaluate implementation progress and impact of the new approach in obstetric care.

OBJECTIVE

The objective of this paper is to describe the design of a study evaluating the impact of implementing risk-dependent care. Validated first-trimester prediction models are embedded in daily clinical practice and combined with risk-dependent obstetric care paths.

METHODS

A multicenter prospective cohort study consisting of women who receive risk-dependent care is being performed from April 2017 to April 2018 (Expect Study II). Obstetric risk profiles will be calculated using a Web-based tool, the Expect prediction tool. The primary outcomes are the adherence of health care professionals and compliance of women. Secondary outcomes are patient satisfaction and cost-effectiveness. Outcome measures will be established using Web-based questionnaires. The secondary outcomes of the risk-dependent care cohort (Expect II) will be compared with the outcomes of a similar prospective cohort (Expect I). Women of this similar cohort received former care-as-usual and were prospectively included between July 1, 2013 and December 31, 2015 (Expect I).

RESULTS

Currently, women are being recruited for the Expect Study II, and a total of 300 women are enrolled.

CONCLUSIONS

This study will provide information about the implementation and impact of a new approach in obstetric care using prediction models and risk-dependent obstetric care paths.

TRIAL REGISTRATION

Netherlands Trial Register NTR4143; http://www.trialregister.nl/trialreg/admin/rctview.asp?TC=4143 (Archived by WebCite at http://www.webcitation.org/6t8ijtpd9).