Lisa M Bodnar1,2,3, Abigail R Cartus1, Sharon I Kirkpatrick4, Katherine P Himes2,3, Edward H Kennedy5, Hyagriv N Simhan2,3, William A Grobman6, Jennifer Y Duffy7, Robert M Silver8, Samuel Parry9, Ashley I Naimi1. 1. Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, PA, USA. 2. Department of Obstetrics, Gynecology, and Reproductive Sciences, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA. 3. Magee-Womens Research Institute, Pittsburgh, PA, USA. 4. School of Public Health and Health Systems, University of Waterloo, Waterloo, Ontario, Canada. 5. Department of Statistics & Data Science, Carnegie Mellon University, Pittsburgh, PA, USA. 6. Department of Obstetrics and Gynecology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA. 7. Department of Obstetrics & Gynecology, School of Medicine, University of California, Irvine, Irvine, CA, USA. 8. Department of Obstetrics and Gynecology, University of Utah, Salt Lake City, UT, USA. 9. Department of Obstetrics and Gynecology, University of Pennsylvania School of Medicine, Philadelphia, PA, USA.
Abstract
BACKGROUND: Conventional analytic approaches for studying diet patterns assume no dietary synergy, which can lead to bias if incorrectly modeled. Machine learning algorithms can overcome these limitations. OBJECTIVES: We estimated associations between fruit and vegetable intake relative to total energy intake and adverse pregnancy outcomes using targeted maximum likelihood estimation (TMLE) paired with the ensemble machine learning algorithm Super Learner, and compared these with results generated from multivariable logistic regression. METHODS: We used data from 7572 women in the Nulliparous Pregnancy Outcomes Study: monitoring mothers-to-be. Usual daily periconceptional intake of total fruits and total vegetables was estimated from an FFQ. We calculated the marginal risk of preterm birth, small-for-gestational-age (SGA) birth, gestational diabetes, and pre-eclampsia according to density of fruits and vegetables (cups/1000 kcal) ≥80th percentile compared with <80th percentile using multivariable logistic regression and Super Learner with TMLE. Models were adjusted for confounders, including other Healthy Eating Index-2010 components. RESULTS: Using logistic regression, higher fruit and high vegetable densities were associated with 1.1% and 1.4% reductions in pre-eclampsia risk compared with lower densities, respectively. They were not associated with the 3 other outcomes. Using Super Learner with TMLE, high fruit and vegetable densities were associated with fewer cases of preterm birth (-4.0; 95% CI: -4.9, -3.0 and -3.7; 95% CI: -5.0, -2.3), SGA (-1.7; 95% CI: -2.9, -0.51 and -3.8; 95% CI: -5.0, -2.5), and pre-eclampsia (-3.2; 95% CI: -4.2, -2.2 and -4.0; 95% CI: -5.2, -2.7) per 100 births, respectively, and high vegetable densities were associated with a 0.9% increase in risk of gestational diabetes. CONCLUSIONS: The differences in results between Super Learner with TMLE and logistic regression suggest that dietary synergy, which is accounted for in machine learning, may play a role in pregnancy outcomes. This innovative methodology for analyzing dietary data has the potential to advance the study of diet patterns.
BACKGROUND: Conventional analytic approaches for studying diet patterns assume no dietary synergy, which can lead to bias if incorrectly modeled. Machine learning algorithms can overcome these limitations. OBJECTIVES: We estimated associations between fruit and vegetable intake relative to total energy intake and adverse pregnancy outcomes using targeted maximum likelihood estimation (TMLE) paired with the ensemble machine learning algorithm Super Learner, and compared these with results generated from multivariable logistic regression. METHODS: We used data from 7572 women in the Nulliparous Pregnancy Outcomes Study: monitoring mothers-to-be. Usual daily periconceptional intake of total fruits and total vegetables was estimated from an FFQ. We calculated the marginal risk of preterm birth, small-for-gestational-age (SGA) birth, gestational diabetes, and pre-eclampsia according to density of fruits and vegetables (cups/1000 kcal) ≥80th percentile compared with <80th percentile using multivariable logistic regression and Super Learner with TMLE. Models were adjusted for confounders, including other Healthy Eating Index-2010 components. RESULTS: Using logistic regression, higher fruit and high vegetable densities were associated with 1.1% and 1.4% reductions in pre-eclampsia risk compared with lower densities, respectively. They were not associated with the 3 other outcomes. Using Super Learner with TMLE, high fruit and vegetable densities were associated with fewer cases of preterm birth (-4.0; 95% CI: -4.9, -3.0 and -3.7; 95% CI: -5.0, -2.3), SGA (-1.7; 95% CI: -2.9, -0.51 and -3.8; 95% CI: -5.0, -2.5), and pre-eclampsia (-3.2; 95% CI: -4.2, -2.2 and -4.0; 95% CI: -5.2, -2.7) per 100 births, respectively, and high vegetable densities were associated with a 0.9% increase in risk of gestational diabetes. CONCLUSIONS: The differences in results between Super Learner with TMLE and logistic regression suggest that dietary synergy, which is accounted for in machine learning, may play a role in pregnancy outcomes. This innovative methodology for analyzing dietary data has the potential to advance the study of diet patterns.
Authors: Ramkripa Raghavan; Carol Dreibelbis; Brittany L Kingshipp; Yat Ping Wong; Barbara Abrams; Alison D Gernand; Kathleen M Rasmussen; Anna Maria Siega-Riz; Jamie Stang; Kellie O Casavale; Joanne M Spahn; Eve E Stoody Journal: Am J Clin Nutr Date: 2019-03-01 Impact factor: 7.045
Authors: Ali H Mokdad; Katherine Ballestros; Michelle Echko; Scott Glenn; Helen E Olsen; Erin Mullany; Alex Lee; Abdur Rahman Khan; Alireza Ahmadi; Alize J Ferrari; Amir Kasaeian; Andrea Werdecker; Austin Carter; Ben Zipkin; Benn Sartorius; Berrin Serdar; Bryan L Sykes; Chris Troeger; Christina Fitzmaurice; Colin D Rehm; Damian Santomauro; Daniel Kim; Danny Colombara; David C Schwebel; Derrick Tsoi; Dhaval Kolte; Elaine Nsoesie; Emma Nichols; Eyal Oren; Fiona J Charlson; George C Patton; Gregory A Roth; H Dean Hosgood; Harvey A Whiteford; Hmwe Kyu; Holly E Erskine; Hsiang Huang; Ira Martopullo; Jasvinder A Singh; Jean B Nachega; Juan R Sanabria; Kaja Abbas; Kanyin Ong; Karen Tabb; Kristopher J Krohn; Leslie Cornaby; Louisa Degenhardt; Mark Moses; Maryam Farvid; Max Griswold; Michael Criqui; Michelle Bell; Minh Nguyen; Mitch Wallin; Mojde Mirarefin; Mostafa Qorbani; Mustafa Younis; Nancy Fullman; Patrick Liu; Paul Briant; Philimon Gona; Rasmus Havmoller; Ricky Leung; Ruth Kimokoti; Shahrzad Bazargan-Hejazi; Simon I Hay; Simon Yadgir; Stan Biryukov; Stein Emil Vollset; Tahiya Alam; Tahvi Frank; Talha Farid; Ted Miller; Theo Vos; Till Bärnighausen; Tsegaye Telwelde Gebrehiwot; Yuichiro Yano; Ziyad Al-Aly; Alem Mehari; Alexis Handal; Amit Kandel; Ben Anderson; Brian Biroscak; Dariush Mozaffarian; E Ray Dorsey; Eric L Ding; Eun-Kee Park; Gregory Wagner; Guoqing Hu; Honglei Chen; Jacob E Sunshine; Jagdish Khubchandani; Janet Leasher; Janni Leung; Joshua Salomon; Jurgen Unutzer; Leah Cahill; Leslie Cooper; Masako Horino; Michael Brauer; Nicholas Breitborde; Peter Hotez; Roman Topor-Madry; Samir Soneji; Saverio Stranges; Spencer James; Stephen Amrock; Sudha Jayaraman; Tejas Patel; Tomi Akinyemiju; Vegard Skirbekk; Yohannes Kinfu; Zulfiqar Bhutta; Jost B Jonas; Christopher J L Murray Journal: JAMA Date: 2018-04-10 Impact factor: 56.272
Authors: Jason D Morgenstern; Laura C Rosella; Andrew P Costa; Russell J de Souza; Laura N Anderson Journal: Adv Nutr Date: 2021-06-01 Impact factor: 8.701
Authors: Geraldine Lo Siou; Alianu K Akawung; Nathan M Solbak; Kathryn L McDonald; Ala Al Rajabi; Heather K Whelan; Sharon I Kirkpatrick Journal: Nutr J Date: 2021-05-08 Impact factor: 3.271
Authors: Ashley I Naimi; Jacqueline E Rudolph; Edward H Kennedy; Abigail Cartus; Sharon I Kirkpatrick; David M Haas; Hyagriv Simhan; Lisa M Bodnar Journal: Epidemiology Date: 2021-03-01 Impact factor: 4.860
Authors: Mark R Cullen; Michael Baiocchi; Lisa Chamberlain; Isabella Chu; Ralph I Horwitz; Michelle Mello; Amy O'Hara; Sam Roosz Journal: SSM Popul Health Date: 2022-03-01