Machine learning prediction of non-attendance to postpartum glucose screening and subsequent risk of type 2 diabetes following gestational diabetes

Periyathambi, Nishanthi; Parkhi, Durga; Ghebremichael‐Weldeselassie, Yonas; Patel, Vinod; Sukumar, Nithya; Siddharthan, Rahul; Narlikar, Leelavati; Saravanan, Ponnusamy

doi:10.1371/journal.pone.0264648

Cited by 9 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While rates were lower than the near-universal adherence to the early GTT in the original prospective cohort study (83% rather than 98%), we still achieved significantly higher rates of postpartum glucose testing than the 30 to 60% reported in other studies at the currently recommended timing. [10][11][12][13] Among postpartum individuals who elected early testing, rates of impaired glucose metabolism and overt diabetes were comparable to what has been found in study settings suggesting that these rates are likely representative and generalizable to other populations. Given that prospective studies have demonstrated clinical equipoise between early and deferred testing for identification of dysglycemia, the fact that fewer individuals had impaired glucose metabolism and diabetes in the group that deferred testing is likely reflective of differences in underlying risk for dysglycemia and overt diabetes between the populations electing for early versus standard testing timing.…”

Section: Discussionsupporting

confidence: 59%

Implementation of In-Hospital Postpartum Glucose Tolerance Testing for People with Gestational Diabetes

Ayala,

Fain,

Smith

et al. 2024

Am J Perinatol

View full text Add to dashboard Cite

Objective We aimed to evaluate uptake of the glucose tolerance test performed during delivery hospitalization as part of routine clinical care. Study Design This is a retrospective cohort study of people with GDM at a tertiary center. We collected 9 months of postimplementation data after the in-hospital (“early”) glucose tolerance test was adopted as a routine screening option. Adherence was compared between those who elected early glucose tolerance testing versus those who deferred testing to the standard postpartum period. Bivariable statistics including demographics, care team, and postpartum testing/visit attendance were compared between those who received early testing and those who did not using chi-square, Fisher's exact, and t-tests. Results A total of 681 patients with GDM delivered during the study period. Of those who had an early glucose tolerance test ordered (n = 408), 340 (83.3%) completed the test. Among those who did not complete an early glucose tolerance test (ordered and not completed or never ordered), only 104/341 (30.5%) completed any postpartum glucose testing in the first 12 months of postpartum. There were significant differences in characteristics in terms of race/ethnicity, insurance, type of gestational diabetes (A1GDM vs. A2GDM), diabetes medications, obstetric care provider, and delivery mode. Among those who completed early testing, 43.7% of participants had impaired glucose metabolism and 6.5% had values concerning for overt diabetes mellitus. Among those who deferred testing to the standard 6- to 12-week period, 24.0% had impaired glucose metabolism and none had overt diabetes. Those who completed an early glucose tolerance test had a lower rate of postpartum visit attendance compared with those who deferred (75.6 vs. 91.5%, p < 0.01). Conclusion In this cohort, when the early glucose tolerance test is offered in clinical practice, adherence rates are higher than when the test is deferred until the postpartum visit. Key Points

show abstract

Section: Discussionsupporting

confidence: 59%

Implementation of In-Hospital Postpartum Glucose Tolerance Testing for People with Gestational Diabetes

Ayala,

Fain,

Smith

et al. 2024

Am J Perinatol

View full text Add to dashboard Cite

show abstract

“…Whether this can be replicated in real-world settings in SA and SEA will require additional studies. A targeted approach for postpartum screening may be a better approach similar to a study by Nishanthi et al [ 27 ] A simple machine learning approach using the routinely available antenatal factors could may identify who is unlikely to attend postpartum screening and enable better, targetted follow-up. The simple risk calculator proposed by Nishanthi et al [ 27 ] is easy to use for healthcare providers.…”

Section: Discussionmentioning

confidence: 99%

Prevalence of prediabetes and type 2 diabetes mellitus in south and southeast Asian women with history of gestational diabetes mellitus: Systematic review and meta-analysis

et al. 2022

Self Cite

View full text Add to dashboard Cite

Background The burden of Gestational Diabetes Mellitus (GDM) is very high in south Asia (SA) and southeast Asia (SEA). Thus, there is a need to understand the prevalence and risk factors for developing prediabetes and type 2 diabetes mellitus (T2DM) postpartum, in this high-risk population. Aim To conduct a systematic review and meta-analysis to estimate the prevalence of prediabetes and T2DM among the women with history of GDM in SA and SEA. Methods A comprehensive literature search was performed in the following databases: Medline, EMBASE, Web of Knowledge and CINHAL till December 2021. Studies that had reported greater than six weeks of postpartum follow-up were included. The pooled prevalence of diabetes and prediabetes were estimated by random effects meta-analysis model and I2 statistic was used to assess heterogeneity. Results Meta-analysis of 13 studies revealed that the prevalence of prediabetes and T2DM in post-GDM women were 25.9% (95%CI 18.94 to 33.51) and 29.9% (95%CI 17.02 to 44.57) respectively. Women with history of GDM from SA and SEA seem to have higher risk of developing T2DM than women without GDM (RR 13.2, 95%CI 9.52 to 18.29, p<0.001). The subgroup analysis showed a rise in the prevalence of T2DM with increasing duration of follow-up. Conclusion The conversion to T2DM and prediabetes is very high among women with history of GDM in SA and SEA. This highlights the need for follow-up of GDM women for early identification of dysglycemia and to plan interventions to prevent/delay the progression to T2DM.

show abstract

“…All women diagnosed with GDM during pregnancy are recommended to have annual screening, 19, 25 although the compliance is currently poor. 5, 26 Therefore, we can allow for more false positives than false negatives and propose c out = 0.140 as the optimal cut-off for classification. However, in low-resource settings, we can primarily focus on women with P (prediabetes) ≥ c in = 0.381 and then consider women with P (prediabetes) ≥ c in−out = 0.260 in the following step.…”

Section: Discussionmentioning

confidence: 99%

Machine learning prediction of early postpartum prediabetes in women with gestational diabetes mellitus

Parkhi

Periyathambi

Ghebremichael‐Weldeselassie

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Background: Early onset of type 2 diabetes and cardiovascular disease are common complications for women diagnosed with gestational diabetes. About half of the women with gestational diabetes develop postpartum prediabetes within 10 years of the index pregnancy. These women also have double the risk of developing cardiovascular disease than women without a history of gestational diabetes. Currently, there is no accurate way of knowing which women with gestational diabetes are likely to develop postpartum prediabetes. This study aims to predict the risk of postpartum prediabetes in women diagnosed with gestational diabetes. Methods: We build a sparse logistic regression-based machine learning model to learn key variables significant for the prediction of postpartum prediabetes, from antenatal data with maternal anthropometric and biochemical variables as well as neonatal characteristics of 607 UK women diagnosed with gestational diabetes. We evaluate the performance of the proposed model in addition to other more advanced machine learning methods using established metrics such as the area under the receiver operating characteristic curve and specificity for pre-determined values of sensitivity. We use K-L divergence and information graphs to evaluate and compare different thresholds of classification for targeted screening options in resource-constrained settings. We also perform a decision curve analysis to study the net standardized benefit of our model compared to the universal screening approach. Results: Strikingly, our sparse logistic regression approach selects only two variables as relevant but gives an area under the receiver operating characteristic curve of 0.72, outperforming all other methods. It can identify postpartum prediabetes in women with gestational diabetes using the Rule-in test with 92\% specificity at an optimal probability threshold of 0.381 and using the Rule-out test with 92\% sensitivity at an optimal probability threshold of 0.140. Conclusion: We propose a simple logistic regression model, which needs only the antenatal fasting glucose at OGTT and HbA1c soon after the diagnosis of GDM, to predict, with remarkable accuracy, the probability of postpartum prediabetes in women with gestational diabetes. We envision this to be a practical solution, which coupled with a targeted follow-up of high-risk women, could yield better cardiometabolic outcomes in women with a history of GDM.

show abstract

Machine learning prediction of non-attendance to postpartum glucose screening and subsequent risk of type 2 diabetes following gestational diabetes

Cited by 9 publications

References 26 publications

Implementation of In-Hospital Postpartum Glucose Tolerance Testing for People with Gestational Diabetes

Implementation of In-Hospital Postpartum Glucose Tolerance Testing for People with Gestational Diabetes

Prevalence of prediabetes and type 2 diabetes mellitus in south and southeast Asian women with history of gestational diabetes mellitus: Systematic review and meta-analysis

Machine learning prediction of early postpartum prediabetes in women with gestational diabetes mellitus

Contact Info

Product

Resources

About