1
|
Wong MS, Wells M, Zamanzadeh D, Akre S, Pevnick JM, Bui AAT, Gregory KD. Applying Automated Machine Learning to Predict Mode of Delivery Using Ongoing Intrapartum Data in Laboring Patients. Am J Perinatol 2024; 41:e412-e419. [PMID: 35752169 DOI: 10.1055/a-1885-1697] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 01/24/2023]
Abstract
OBJECTIVE This study aimed to develop and validate a machine learning (ML) model to predict the probability of a vaginal delivery (Partometer) using data iteratively obtained during labor from the electronic health record. STUDY DESIGN A retrospective cohort study of deliveries at an academic, tertiary care hospital was conducted from 2013 to 2019 who had at least two cervical examinations. The population was divided into those delivered by physicians with nulliparous term singleton vertex (NTSV) cesarean delivery rates <23.9% (Partometer cohort) and the remainder (control cohort). The cesarean rate among this population of lower risk patients is a standard metric by which to compare provider rates; <23.9% was the Healthy People 2020 goal. A supervised automated ML approach was applied to generate a model for each population. The primary outcome was accuracy of the model developed on the Partometer cohort at 4 hours from admission to labor and delivery. Secondary outcomes included discrimination ability (receiver operating characteristics-area under the curve [ROC-AUC]), precision-recall AUC, and calibration of the Partometer. To assess generalizability, we compared the performance and clinical predictors identified by the Partometer to the control model. RESULTS There were 37,932 deliveries during the study period; after exclusions, 9,385 deliveries were included in the Partometer cohort and 19,683 in the control cohort. Accuracy of predicting vaginal delivery at 4 hours was 87.1% for the Partometer (ROC-AUC: 0.82). Clinical predictors of greatest importance in the stacked Intrapartum Partometer Model included the Admission Model prediction and ongoing measures of dilatation and station which mirrored those found in the control population. CONCLUSION Using automated ML and intrapartum factors improved the accuracy of prediction of probability of a vaginal delivery over both previously published models based on logistic regression. Harnessing real-time data and ML could represent the bridge to generating a truly prescriptive tool to augment clinical decision-making, predict labor outcomes, and reduce maternal and neonatal morbidity. KEY POINTS · Our ML-based model yielded accurate predictions of mode of delivery early in labor.. · Predictors for models created on populations with high and low cesarean rates were the same.. · A ML-based model may provide meaningful guidance to clinicians managing labor..
Collapse
Affiliation(s)
- Melissa S Wong
- Department of Obstetrics and Gynecology, Cedars-Sinai Medical Center, Los Angeles, California
- Division of Informatics, Department of Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California
| | - Matthew Wells
- Enterprise Data Intelligence, Cedars-Sinai Medical Center, Los Angeles, California
| | - Davina Zamanzadeh
- Medical and Imaging Informatics (MII) Group, Department of Radiological Sciences, University of California Los Angeles, Los Angeles, California
| | - Samir Akre
- Medical and Imaging Informatics (MII) Group, Department of Radiological Sciences, University of California Los Angeles, Los Angeles, California
| | - Joshua M Pevnick
- Division of Informatics, Department of Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California
- Division of General Internal Medicine, Department of Medicine, Cedars-Sinai Medical Center, Los Angeles, California
| | - Alex A T Bui
- Medical and Imaging Informatics (MII) Group, Department of Radiological Sciences, University of California Los Angeles, Los Angeles, California
- Department of Bioengineering, University of California Los Angeles, Los Angeles, California
| | - Kimberly D Gregory
- Department of Obstetrics and Gynecology, Cedars-Sinai Medical Center, Los Angeles, California
- Department of Obstetrics and Gynecology, David Geffen School of Medicine at University of California Los Angeles, Los Angeles, California
- Department of Community Health Sciences, Los Angeles, California Fielding School of Public Health, Los Angeles, California
| |
Collapse
|
2
|
Spurlock EJ, Pickler RH, Ruiz RJ, Ford J, Gillespie S, Kue J. Acculturation, Acculturative Stress, Experience of Discrimination, and Cesarean Birth in Mexican American Women. HISPANIC HEALTH CARE INTERNATIONAL 2023; 21:184-194. [PMID: 36949611 DOI: 10.1177/15404153231164369] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 03/24/2023]
Abstract
Introduction: There is evidence that Mexican Americans are more likely to have cesarean birth than non-Hispanic White Americans. The purpose of this study was to identify factors related to acculturation along with psychological and sociodemographic factors associated with birth mode in a prospective cohort of Mexican American women in Texas. Methods: This secondary analysis included 244 Mexican American pregnant women. Women with a prior cesarean birth were excluded. Variable selection was guided by Berry's Theoretical Framework of Acculturation. Correlations and logistic regression were used to examine relationships and predict risk of cesarean birth. Mediators and moderators were also considered. Results: Eighty women birthed by cesarean. Analytic and parent samples were similar in all demographics. After controlling for parity in logistic regression, greater Spanish language-related acculturative stress (adjusted odds ratio [AOR], 1.06, 95% confidence interval [CI] [1.01, 1.11], p = .028) and experience of discrimination (AOR, 1.18, 95% CI [1.00, 1.38], p = .044) increased the odds of cesarean birth. The relationship between acculturative stress and birth mode was moderated by birth facility. Conclusion: Acculturative stress and discrimination may play a role in birth mode for Mexican American women birthing in Texas. Birth facility and acculturative stress may be interacting in ways that have clinical significance but are yet unexplored.
Collapse
Affiliation(s)
| | - Rita H Pickler
- College of Nursing, The Ohio State University, Columbus, OH, USA
| | | | - Jodi Ford
- College of Nursing, The Ohio State University, Columbus, OH, USA
| | | | - Jennifer Kue
- College of Nursing, The Ohio State University, Columbus, OH, USA
| |
Collapse
|
3
|
Spurlock EJ, Kue J, Gillespie S, Ford J, Ruiz RJ, Pickler RH. Integrative Review of Disparities in Mode of Birth and Related Complications among Mexican American Women. J Midwifery Womens Health 2021; 67:95-106. [PMID: 34958159 DOI: 10.1111/jmwh.13288] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/20/2021] [Revised: 08/04/2021] [Accepted: 08/12/2021] [Indexed: 11/30/2022]
Abstract
INTRODUCTION Cesarean rates are particularly high among Hispanic women in some regions of the United States, placing a disproportionate health burden on women and their newborns. This integrative review synthesized the literature on mode of birth (vaginal vs cesarean) and related childbirth complications (hemorrhage, surgical site infection, perineal trauma) among Mexican American women living in the United States. METHODS Four electronic databases, PubMed, Embase, CINAHL, and SCOPUS, were searched to identify studies meeting the inclusion criteria, research studies that included Mexican American women who were pregnant or postpartum. Results were limited to English language and publications that were peer-reviewed and published before May 2020. Covidence was used in article identification, screening, and assessment. Critical appraisal of the research was performed using the Quality Assessment Tool for Studies with Diverse Designs. RESULTS Ten articles met inclusion criteria. In some studies, Mexican American women born in the United States were more likely to have cesareans than women born in Mexico; in other studies, these findings were reversed. Mexican American women often had lower unadjusted cesarean rates compared with non-Hispanic white women, but adjusting for birth facility (some facilities perform more cesareans than others), sociodemographic, and risk factors often revealed Mexican American women have a higher adjusted risk for cesarean birth. Women with higher socioeconomic status had higher cesarean rates compared with women with lower socioeconomic status. In studies of birth outcome by level of acculturation, women who were US-oriented had higher rates of cesarean and more frequent perinatal complications. By ethnic subgroup, rates of cesarean and complications varied among Hispanic women. DISCUSSION Birth facility was associated with perinatal outcomes for Mexican American women; those who gave birth at higher-performing facilities had better outcomes when compared with women who gave birth at lower-performing facilities. After adjusting for pregnancy complications, Mexican American women had a greater risk for cesarean birth compared with non-Hispanic white women, a finding that may have clinical practice implications. Level of acculturation affected birth outcomes, but more research using precise instruments is needed.
Collapse
Affiliation(s)
| | - Jennifer Kue
- College of Nursing, University of South Florida, Tampa, Florida
| | | | - Jodi Ford
- College of Nursing, The Ohio State University, Columbus, Ohio
| | | | - Rita H Pickler
- College of Nursing, The Ohio State University, Columbus, Ohio
| |
Collapse
|
4
|
McDonald JA, Amatya A, Gard CC. Multi-level Drivers of Disparities in Hispanic Cesarean Delivery Rates in US-Mexico Border States. J Racial Ethn Health Disparities 2019; 7:238-250. [PMID: 31686370 DOI: 10.1007/s40615-019-00652-2] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/05/2019] [Revised: 10/03/2019] [Accepted: 10/10/2019] [Indexed: 11/24/2022]
Abstract
BACKGROUND Hispanic women living along the US-Mexico border have higher cesarean delivery rates than non-Hispanic white women, African American women, and other Hispanic women in the USA. Their rates also exceed those of other Hispanic women in states that border Mexico and non-Hispanic white women along the border. Our objective was to determine the causes of the disparities in border Hispanic cesarean rates. METHODS Using the 2015 birth certificate file and other sources, we performed a twofold Oaxaca-Blinder decomposition analysis of the disparities in low-risk primary and repeat cesarean rates between Hispanic and non-Hispanic white women in the US-Mexico border counties and Hispanic women residing in nonborder counties of border states. RESULTS Rates of low-risk primary cesarean among border Hispanic, nonborder Hispanic, and border non-Hispanic white women were 21.1%, 15.0%, and 16.5%, respectively. Higher Hispanic concentration in county of residence, a larger proportion of for-profit hospital beds, and greater poverty accounted for 24.7%, 22.1%, and 11.1% of the border-nonborder Hispanic difference, respectively. No other variable explained more than 5% of the difference. Higher Hispanic concentration, more for-profit beds, less attendance by an MD, higher BMI, and greater poverty explained 60.6%, 42.4%, 42.4%, 27.4%, and 21.3%, respectively, of the Hispanic-non-Hispanic white difference. Hispanic concentration and for-profit beds were also important explanatory variables for low-risk repeat cesareans. CONCLUSION Efforts to address potentially unnecessary cesareans among Hispanic women on the border should recognize that community demographic and health delivery system characteristics are more influential than maternal medical risk factors.
Collapse
Affiliation(s)
- Jill A McDonald
- Department of Public Health Sciences, College of Health and Social Services, New Mexico State University, PO Box 30001, Las Cruces, NM, 88003-8001, USA.
- Southwest Institute for Health Disparities Research, College of Health and Social Services, New Mexico State University, PO Box 30001, Las Cruces, NM, 88003-8001, USA.
| | - Anup Amatya
- Department of Public Health Sciences, College of Health and Social Services, New Mexico State University, PO Box 30001, Las Cruces, NM, 88003-8001, USA
- Southwest Institute for Health Disparities Research, College of Health and Social Services, New Mexico State University, PO Box 30001, Las Cruces, NM, 88003-8001, USA
| | - Charlotte C Gard
- Southwest Institute for Health Disparities Research, College of Health and Social Services, New Mexico State University, PO Box 30001, Las Cruces, NM, 88003-8001, USA
- Department of Economics, College ofBusiness, Applied Statistics, and International Business, New Mexico State University, Las Cruces, NM, USA
| |
Collapse
|
5
|
McDonald JA, Amatya A, Gard CC, Sigala J. In States That Border Mexico, Cesarean Rates Were Highest For Hispanic Women Living In Border Counties In 2015. Health Aff (Millwood) 2019; 38:276-286. [DOI: 10.1377/hlthaff.2018.05369] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/23/2022]
Affiliation(s)
- Jill A. McDonald
- Jill A. McDonald is the Stan Fulton Endowed Chair in Health Disparities Research; director of the Southwest Institute for Health Disparities Research; and a professor in the Department of Public Health Sciences, College of Health and Social Services, New Mexico State University, in Las Cruces
| | - Anup Amatya
- Anup Amatya is an associate professor in the Department of Public Health Sciences; is a member of the Biostatistics and Epidemiology Research Design Core of the Mountain West Idea Clinical and Translational Research–Infrastructure Network (CTR-IN); and is affiliated with the Southwest Institute for Health Disparities Research, College of Health and Social Services, New Mexico State University
| | - Charlotte C. Gard
- Charlotte C. Gard is an associate professor in the Department of Economics, Applied Statistics, and International Business and is affiliated with the Southwest Institute for Health Disparities Research, College of Business, New Mexico State University
| | - Jesus Sigala
- Jesus Sigala is a graduate student in the Department of Economics, Applied Statistics, and International Business and is affiliated with the Southwest Institute for Health Disparities Research, College of Business, New Mexico State University
| |
Collapse
|