Safety of low weight gain or weight loss in pregnancies with class 1, 2, and 3 obesity: a population-based cohort study.

BACKGROUND: There are concerns that current gestational weight gain recommendations for women with obesity are too high and that guidelines should differ on the basis of severity of obesity. In this study we investigated the safety of gestational weight gain below current recommendations or weight loss in pregnancies with obesity, and evaluated whether separate guidelines are needed for different obesity classes. METHODS: In this population-based cohort study, we used electronic medical records from the Stockholm-Gotland Perinatal Cohort study to identify pregnancies with obesity (early pregnancy BMI before 14 weeks' gestation ≥30 kg/m2) among singleton pregnancies that delivered between Jan 1, 2008, and Dec 31, 2015. The pregnancy records were linked with Swedish national health-care register data up to Dec 31, 2019. Gestational weight gain was calculated as the last measured weight before or at delivery minus early pregnancy weight (at <14 weeks' gestation), and standardised for gestational age into z-scores. We used Poisson regression to assess the association of gestational weight gain z-score with a composite outcome of: stillbirth, infant death, large for gestational age and small for gestational age at birth, preterm birth, unplanned caesarean delivery, gestational diabetes, pre-eclampsia, excess postpartum weight retention, and new-onset longer-term maternal cardiometabolic disease after pregnancy, weighted to account for event severity. We calculated rate ratios (RRs) for our composite adverse outcome along the weight gain z-score continuum, compared with a reference of the current lower limit for gestational weight gain recommended by the US Institute of Medicine (IOM; 5 kg at term). RRs were adjusted for confounding factors (maternal age, height, parity, early pregnancy BMI, early pregnancy smoking status, prepregnancy cardiovascular disease or diabetes, education, cohabitation status, and Nordic country of birth). FINDINGS: Our cohort comprised 15 760 pregnancies with obesity, followed up for a median of 7·9 years (IQR 5·8-9·4). 11 667 (74·0%) pregnancies had class 1 obesity, 3160 (20·1%) had class 2 obesity, and 933 (5·9%) had class 3 obesity. Among these pregnancies, 1623 (13·9%), 786 (24·9%), and 310 (33·2%), respectively, had weight gain during pregnancy below the lower limit of the IOM recommendation (5 kg). In pregnancies with class 1 or 2 obesity, gestational weight gain values below the lower limit of the IOM recommendation or weight loss did not increase risk of the adverse composite outcome (eg, at weight gain z-score -2·4, corresponding to 0 kg at 40 weeks: adjusted RR 0·97 [95% CI 0·89-1·06] in obesity class 1 and 0·96 [0·86-1·08] in obesity class 2). In pregnancies with class 3 obesity, weight gain values below the IOM limit or weight loss were associated with reduced risk of the adverse composite outcome (eg, adjusted RR 0·81 [0·71-0·89] at weight gain z-score -2·4, or 0 kg). INTERPRETATION: Our findings support calls to lower or remove the lower limit of current IOM recommendations for pregnant women with obesity, and suggest that separate guidelines for class 3 obesity might be warranted. FUNDING: Karolinska Institutet and the Eunice Kennedy Shriver National Institute of Child Health and Human Development.

Pseudo-Random Number Generator Influences on Average Treatment Effect Estimates Obtained with Machine Learning.

BACKGROUND: Use of machine learning to estimate exposure effects introduces a dependence between the results of an empirical study and the value of the seed used to fix the pseudo-random number generator. METHODS: We used data from 10,038 pregnant women and a 10% subsample (N = 1,004) to examine the extent to which the risk difference for the relation between fruit and vegetable consumption and preeclampsia risk changes under different seed values. We fit an augmented inverse probability weighted estimator with two Super Learner algorithms: a simple algorithm including random forests and single layer neural networks and a more complex algorithm with a mix of tree-based, regression based, penalized and simple algorithms. We evaluated the distributions of risk differences, standard errors, and p values that result from 5,000 different seed value selections. RESULTS: Our findings suggest important variability in the risk difference estimates, as well as an important effect of the stacking algorithm used. The interquartile range width (IQRw) of the risk differences in the full sample with the simple algorithm was 13 per 1000. However, all other IQRs were roughly an order of magnitude lower. The medians of the distributions of risk differences differed according to the sample size and the algorithm used. CONCLUSIONS: Our findings add another dimension of concern regarding the potential for "p-hacking", and further warrants the need to move away from simplistic evidentiary thresholds in empirical research. When empirical results depend on pseudo-random number generator seed values, caution is warranted in interpreting these results.

Application of a Web-based Tool for Quantitative Bias Analysis: The Example of Misclassification Due to Self-reported Body Mass Index.

BACKGROUND: We describe the use of Apisensr, a web-based application that can be used to implement quantitative bias analysis for misclassification, selection bias, and unmeasured confounding. We apply Apisensr using an example of exposure misclassification bias due to use of self-reported body mass index (BMI) to define obesity status in an analysis of the relationship between obesity and diabetes. METHODS: We used publicly available data from the National Health and Nutrition Examination Survey. The analysis consisted of: (1) estimating bias parameter values (sensitivity, specificity, negative predictive value, and positive predictive value) for self-reported obesity by sex, age, and race-ethnicity compared to obesity defined by measured BMI, and (2) using Apisensr to adjust for exposure misclassification. RESULTS: The discrepancy between self-reported and measured obesity varied by demographic group (sensitivity range: 75%-89%; specificity range: 91%-99%). Using Apisensr for quantitative bias analysis, there was a clear pattern in the results: the relationship between obesity and diabetes was underestimated using self-report in all age, sex, and race-ethnicity categories compared to measured obesity. For example, in non-Hispanic White men aged 40-59 years, prevalence odds ratios for diabetes were 3.06 (95% confidence inerval = 1.78, 5.30) using self-reported BMI and 4.11 (95% confidence interval = 2.56, 6.75) after bias analysis adjusting for misclassification. CONCLUSION: Apisensr is an easy-to-use, web-based Shiny app designed to facilitate quantitative bias analysis. Our results also provide estimates of bias parameter values that can be used by other researchers interested in examining obesity defined by self-reported BMI.

Validation of Long-term Recall of Pregnancy-related Weight in the Life-course Experiences And Pregnancy Study.

BACKGROUND: Prepregnancy body mass index (BMI) and gestational weight gain (GWG) are determinants of maternal and child health. However, many studies of these factors rely on error-prone self-reported measures. METHODS: Using data from Life-course Experiences And Pregnancy (LEAP), a US-based cohort, we assessed the validity of prepregnancy BMI and GWG recalled on average 8 years postpartum against medical record data treated as alloyed gold standard ("true") values. We calculated probabilities of being classified into a self-reported prepregnancy BMI or GWG category conditional on one's true category (analogous to sensitivities and specificities) and probabilities of truly being in each prepregnancy BMI or GWG category conditional on one's self-reported category (analogous to positive and negative predictive values). RESULTS: There was a tendency toward under-reporting prepregnancy BMI. Self-report misclassified 32% (95% confidence interval [CI] = 19%, 48%) of those in LEAP with truly overweight and 13% (5%, 27%) with obesity into a lower BMI category. Self-report correctly predicted the truth for 72% (55%, 84%) with self-reported overweight to 100% (90%, 100%) with self-reported obesity. For GWG, both under- and over-reporting were common; self-report misclassified 32% (15%, 55%) with truly low GWG as having moderate GWG and 50% (28%, 72%) with truly high GWG as moderate or low GWG. Self-report correctly predicted the truth for 45% (25%, 67%) with self-reported high GWG to 85% (76%, 91%) with self-reported moderate GWG. Misclassification of BMI and GWG varied across maternal characteristics. CONCLUSION: Findings can be used in quantitative bias analyses to estimate bias-adjusted associations with prepregnancy BMI and GWG.

Maternal history of childhood maltreatment and pregnancy weight outcomes.

BACKGROUND: Childhood maltreatment is associated with elevated adult weight. It is unclear whether this association extends to pregnancy, a critical window for the development of obesity. METHODS: We examined associations of childhood maltreatment histories with pre-pregnancy BMI and gestational weight gain among women who had participated for >20 years in a longitudinal cohort.At age 26-35 participants reported childhood maltreatment (physical, sexual, and emotional abuse; emotional neglect) and, 5 years later, about pre-pregnancy weight and gestational weight gain for previous pregnancies (n=656). Modified Poisson regression models were used to estimate associations of maltreatment history with pre-pregnancy BMI and gestational weight gain z-scores, adjusting for sociodemographics. We used Multivariate Imputation by Chained Equations to adjust outcome measures for misclassification using data from an internal validation study. RESULTS: Before misclassification adjustment, results indicated a higher risk of pre-pregnancy BMI ≥30 kg/m2 in women with certain types of maltreatment (e.g., emotional abuse RR=2.4; 95% CI: 1.5, 3.7) compared with women without that maltreatment type. After misclassification adjustment, estimates were attenuated but still modestly elevated (e.g., emotional abuse RR=1.7; 95% CI: 1.1, 2.7). Misclassification-adjusted estimates for maltreatment associations with gestational weight gain z-scores were close to the null and imprecise. CONCLUSIONS: Findings suggest an association of maltreatment with pre-pregnancy BMI ≥30 kg/m2 but not with high gestational weight gain. Results suggest a potential need for equitable interventions that can support all women, including those with maltreatment histories, as they enter pregnancy.

Periconceptional Dietary Patterns and Adverse Pregnancy and Birth Outcomes.

BACKGROUND: The periconceptional period is a critical window for the origins of adverse pregnancy and birth outcomes, yet little is known about the dietary patterns that promote perinatal health. OBJECTIVE: We used machine learning methods to determine the effect of periconceptional dietary patterns on risk of preeclampsia, gestational diabetes, preterm birth, small-for-gestational-age (SGA) birth, and a composite of these outcomes. METHODS: We used data from 8259 participants in the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be (8 US medical centers, 2010‒2013). Usual daily periconceptional intake of 82 food groups was estimated from a food frequency questionnaire. We used k-means clustering with a Euclidean distance metric to identify dietary patterns. We estimated the effect of dietary patterns on each perinatal outcome using targeted maximum likelihood estimation and an ensemble of machine learning algorithms, adjusting for confounders including health behaviors and psychological, neighborhood, and sociodemographic factors. RESULTS: The 4 dietary patterns that emerged from our data were identified as "Sandwiches and snacks" (34% of the sample); "High fat, sugar, and sodium" (29%); "Beverages, refined grains, and mixed dishes" (21%); and "High fruits, vegetables, whole grains, and plant proteins" (16%). One-quarter of pregnancies had preeclampsia (8% incidence), gestational diabetes (5%), preterm birth (8%), or SGA birth (8%). Compared with the "High fat, sugar, and sodium" pattern, there were 3.3 to 4.3 fewer cases of the composite adverse outcome per 100 pregnancies among participants following the "Beverages, refined grains and mixed dishes" pattern (risk difference -0.043; 95% confidence interval -0.078, -0.009), "High fruits, vegetables, whole grains and plant proteins" pattern (-0.041; 95% confidence interval -0.078, -0.004), and "Sandwiches and snacks" pattern (-0.033; 95% confidence interval -0.065, -0.002). CONCLUSIONS: Our results highlight that there are a variety of periconceptional dietary patterns that are associated with perinatal health and reinforce the negative health implications of diets high in fat, sugars, and sodium.

The impact of cohort inclusion/exclusion criteria on pregnancy weight gain chart percentiles.

Pregnancy weight gain standards are charts describing percentiles of weight gain among participants with no risk factors that could adversely affect weight gain. This detailed information is burdensome to collect. We investigated the extent to which exclusion of various pre-pregnancy, pregnancy and postpartum factors impacted the values of pregnancy weight gain percentiles. We examined pregnancy weight gain (kg) among 3178 participants of the US nuMoM2b-Heart Health Study (HHS). We identified five groups of potential exclusion criteria for pregnancy weight gain standards: socio-economic characteristics (group 1), maternal morbidities (group 2), lifestyle/behaviour factors (group 3), adverse neonatal outcomes (group 4) and longer-term adverse outcomes (group 5). We established the impact of different exclusion criteria by comparing the median, 25th and 75th percentiles of weight gain in the full cohort with the values after applying each of the five exclusion criteria groups. Differences > 0·75 kg were considered meaningful. Excluding participants with group 1, 2, 3 or 4 exclusion criteria had no impact on the 25th, median or 75th percentiles of pregnancy weight gain. Percentiles were only meaningfully different after excluding participants in group 5 (longer-term adverse outcomes), which shifted the upper end of the weight gain distribution to lower values (e.g. 75th percentile decreased from 19·6 kg to 17·8 kg). This shift was due to exclusion of participants with excess postpartum weight retention > 5 kg or > 10 kg. Except for excess postpartum weight retention, most potential exclusion criteria for pregnancy weight gain standards did not meaningfully impact chart percentiles.

Should gestational weight gain charts exclude individuals with excess postpartum weight retention?

BACKGROUND: High gestational weight gain is associated with excess postpartum weight retention, yet excess postpartum weight retention is not an exclusion criterion for current gestational weight gain charts. We aimed to assess the impact of excluding individuals with high interpregnancy weight change (a proxy for excess postpartum weight retention) on gestational weight gain distributions. METHODS: We included individuals with an index birth from 2008 to 2014 and a subsequent birth before 2019, in the population-based Stockholm-Gotland Perinatal Cohort. We estimated gestational weight gain (kg) at 25 and 37 weeks, using weight at first prenatal visit (<14 weeks) as the reference. We calculated high interpregnancy weight change (≥10 kg and ≥5 kg) using the difference between weight at the start of an index and subsequent pregnancy. We compared gestational weight gain distributions and percentiles (stratified by early-pregnancy body mass index) before and after excluding participants with high interpregnancy weight change. RESULTS: Among 55,723 participants, 17% had ≥10 kg and 34% had ≥5 kg interpregnancy weight change. The third, tenth, 50th, 90th and 97th percentiles of gestational weight gain were similar (largely within 1 kg) before versus after excluding participants with high interpregnancy weight change, at both 25 and 37 weeks. For example, among normal weight participants at 37 weeks, the 50th and 97th percentiles were 14 kg and 23 kg including versus 13 kg and 23 kg excluding participants with ≥5 kg interpregnancy weight change. CONCLUSIONS: Excluding individuals with excess postpartum weight retention from normative gestational weight gain charts may not meaningfully impact the charts' percentiles.

Investigating the Shape and Strength of the Relationship Between Maternal Weight Gain and Gestational Age at Delivery in Twin and Singleton Pregnancies.

Both inadequate and excessive maternal weight gain are correlated with preterm delivery in singleton pregnancies, yet this relationship has not been adequately studied in twins. We investigated the relationship between time-varying maternal weight gain and gestational age at delivery in twin pregnancies and compared it with that in singletons delivered in the same study population. We used serial weight measurements abstracted from charts for twin and singleton pregnancies delivered during 1998-2013 in Pittsburgh, Pennsylvania. Our exposure was time-varying weight gain z score, calculated using gestational age-standardized and prepregnancy body mass index-stratified twin- and singleton-specific charts, and our outcome was gestational age at delivery. Our analyses used a flexible extension of the Cox proportional hazards model that allowed for nonlinear and time-dependent effects. We found a U-shaped relationship between weight gain z score and gestational age at delivery among twin pregnancies (lowest hazard of delivery observed at z score = 1.2), which we attributed to increased hazard of early preterm spontaneous delivery among pregnancies with low weight gain and increased hazard of late preterm delivery without labor among pregnancies with high weight gain. Our findings may be useful for updating provisional guidelines for maternal weight gain in twin pregnancies.

Are Detailed Behavioral, Psychosocial, and Environmental Variables Necessary to Control for Confounding in Pregnancy Weight Gain Research?

BACKGROUND: Associations between pregnancy weight gain and adverse outcomes may be spurious owing to confounding by factors not typically measured in cohort studies. We determined the extent to which the addition of detailed behavioral, psychosocial, and environmental measurements to commonly available covariates improved control of confounding. METHODS: We used data from a prospective US pregnancy cohort study (2010-2013, n = 8978). We calculated two propensity scores for low and high pregnancy weight gain (vs. adequate gain) using 11 standard confounders (e.g., age and education). We examined the balance of characteristics between weight gain groups before and after propensity score matching. We used negative binomial regression to estimate the association between weight gain and small- and large-for-gestational-age birth, preterm birth, and unplanned cesarean delivery, controlling for propensity score. To this model, we then added 17 detailed behavioral, psychosocial, and environmental measurements ("fully adjusted"). We calculated the risk ratio owing to confounding as the ratio of the standard confounder-adjusted risk ratio to the fully adjusted risk ratio. RESULTS: There were minimal imbalances between weight gain groups in detailed measures after matching for a propensity score of standard covariates. Accordingly, the inclusion of detailed covariates had minimal impact on estimated associations between low or high pregnancy weight gain and adverse pregnancy outcomes: risk ratios owing to confounding were null for all outcomes (e.g., 1.1 [95% CI = 1.0, 1.1] for low weight gain and preterm birth). CONCLUSIONS: Adjustment for detailed behavioral, psychosocial, and environmental measurements had minimal impact on estimated associations between pregnancy weight gain and adverse perinatal outcomes.

Experiences of Racial Discrimination and Periconceptional Diet Quality.

BACKGROUND: Racism is a key determinant of perinatal health disparities. Poor diet may contribute to this effect, but research on racism and dietary patterns is limited. OBJECTIVE: We aimed to describe the relation between experiences of racial discrimination and adherence to the 2015‒2020 Dietary Guidelines for Americans. METHODS: We used data from a prospective pregnancy cohort study conducted at 8 United States medical centers (2010‒2013). At 6‒13 weeks of gestation, 10,038 nulliparous people with singleton pregnancies were enrolled. Participants completed a Block food frequency questionnaire, assessing usual diet in the 3 mo around conception, and the Krieger Experiences of Discrimination Scale, assessing the number of situational domains (e.g., at school and on the street) in which participants ever experienced racial discrimination. Alignment of dietary intake with the 2015-2020 Dietary Guidelines for Americans was assessed using the Healthy Eating Index (HEI)-2015. RESULTS: The study showed that 49%, 44%, 35%, and 17% of the Asian, Black, Hispanic, and White participants reported experiences of racial discrimination in any domain. Most participants experienced discrimination in 1 or 2 situational domains. There were no meaningful differences in HEI-2015 total or component scores in any racial or ethnic group according to count of self-reported domains in which individuals experienced discrimination. For example, mean total scores were 57‒59 among Black, 61‒66 among White, 61‒63 among Hispanic, and 66‒69 among Asian participants across the count of racial discrimination domains. CONCLUSIONS: This null association stresses the importance of going beyond interpersonal racial discrimination to consider the institutions, systems, and practices affecting racialized people to eliminate persistent inequalities in diet and perinatal health.

The association between multidimensional sleep health and gestational weight gain.

BACKGROUND: Although poor sleep health is associated with weight gain and obesity in the non-pregnant population, research on the impact of sleep health on weight change among pregnant people using a multidimensional sleep health framework is needed. OBJECTIVES: This secondary data analysis of the Nulliparous Pregnancy Outcome Study: Monitoring Mothers-to-be Sleep Duration and Continuity Study (n = 745) examined associations between mid-pregnancy sleep health indicators, multidimensional sleep health and gestational weight gain (GWG). METHODS: Sleep domains (i.e. regularity, nap duration, timing, efficiency and duration) were assessed via actigraphy between 16 and 21 weeks of gestation. We defined 'healthy' sleep in each domain with empirical thresholds. Multidimensional sleep health was based on sleep profiles derived from latent class analysis and composite score defined as the sum of healthy sleep domains. Total GWG, the difference between self-reported pre-pregnancy weight and the last measured weight before delivery, was converted to z-scores using gestational age- and BMI-specific charts. GWG was defined as low (<-1 SD), moderate (-1 or +1 SD) and high (>+1 SD). RESULTS: Nearly 50% of the participants had a healthy sleep profile (i.e. healthy sleep in most domains), whereas others had a sleep profile defined as having varying degrees of unhealthy sleep in each domain. The individual sleep domains were associated with a 20%-30% lower risk of low or high GWG. Each additional healthy sleep indicator was associated with a 10% lower risk of low (vs. moderate), but not high, GWG. Participants with late timing, long duration and low efficiency (vs. healthy) profiles had the strongest risk of low GWG (relative risk 1.5, 95% confidence interval 0.9, 2.4). Probabilistic bias analysis suggested that most associations between individual sleep health indicators, sleep health profiles and GWG were biased towards the null. CONCLUSIONS: Future research should determine whether sleep health is an intervention target for healthy GWG.

Severe Maternal Morbidity in Twins.

OBJECTIVE: While twin gestations are at increased risk of severe maternal morbidity (SMM), there is limited information about timing and causes of SMM in twins. Furthermore, existing data rely on screening definitions of SMM because a gold standard approach requires chart review. We sought to determine the timing and cause of SMM in twins using a gold standard definition outlined by the American College of Obstetricians and Gynecologists (ACOG). STUDY DESIGN: We used a perinatal database to identify all twin deliveries from 1998 to 2013 at a single academic medical center (n = 2,367). Deliveries were classified as screen positive for SMM if they met any of the following criteria: (1) one of the Centers for Disease Control and Prevention (CDC) International Classification of Diseases Ninth Revision diagnosis and procedure codes for SMM; (2) a prolonged postpartum length of stay (>3 standard deviations beyond mean length of stay by mode of delivery); or (3) maternal intensive care unit admission. We identified true cases of SMM through medical record review of all screen-positive deliveries using the definition of SMM outlined in the ACOG Obstetric Care Consensus. We also determined cause and timing of SMM. RESULTS: A total of 165 (7%) of twin deliveries screened positive for SMM. After chart review of all screen-positive cases, 2.4% (n = 56) were classified as a true case of SMM using the ACOG definition for a positive predictive value of 34%. The majority of SMM occurred postpartum (65%). Hemorrhage was the most common cause of SMM, followed by hypertensive and pulmonary etiologies. CONCLUSION: Commonly used approaches to screen for SMM perform poorly in twins. This has important implications for quality initiatives and epidemiologic studies that rely on screening definitions of maternal morbidity. Our study demonstrates that the immediate postpartum period is a critical time for maternal health among women with twin pregnancies. KEY POINTS: · Screening approaches for SMM have low positive predictive value in twins.. · Hemorrhage, hypertensive, and pulmonary complications were the most common morbidities.. · SMM was most common postpartum..

Early Gestational Weight Gain and the Risk of Preeclampsia in Dichorionic Twin Pregnancies.

OBJECTIVE: The purpose of our study was to evaluate the body mass index (BMI)-specific association between early gestational weight gain (GWG) in dichorionic twin pregnancies and the risk of preeclampsia. STUDY DESIGN: We conducted a retrospective cohort study of all dichorionic twin pregnancies from 1998 to 2013. Data were obtained from a perinatal database and chart abstraction. Prepregnancy BMI was categorized as normal (18.5-24.9 kg/m2), overweight (25-29.9 kg/m2), and obese (≥30 kg/m2). Early GWG was defined as the last measured weight from 160/7 to 196/7weeks' gestation minus prepregnancy weight. GWG was standardized for gestational duration using BMI-specific z-score charts for dichorionic pregnancies. Preeclampsia was diagnosed using American College of Obstetricians and Gynecologists criteria and identified with International Classification of Diseases-9 coding. Early GWG z-score was modeled as a three-level categorical variable (≤ - 1 standard deviation [SD], 0, 3 +1 SD), where -1 to +1 was the referent group. We estimated risk differences and 95% confidence intervals (CIs) via marginal standardization. RESULTS: We included 1,693 dichorionic twin pregnancies in the cohort. In adjusted analysis, the incidence of preeclampsia increased with increasing early GWG among women with normal BMI. Women with normal BMI and a GWG z-score < - 1 (equivalent to 2.6 kg by 20 weeks) had 2.5 fewer cases of preeclampsia per 100 births (95% CI: -4.7 to - 0.3) compared with the referent; those with GWG z-score > +1 (equivalent to gaining 9.8 kg by 20 weeks) had 2.8 more cases of preeclampsia per 100 (95 % CI: 0.1-5.5) compared with the referent. In adjusted analyses, early GWG had minimal impact on the risk of preeclampsia in women with overweight or obesity. CONCLUSION: GWG of 2.6 kg or less by 20 weeks was associated with a decreased risk of preeclampsia among women pregnant with dichorionic twins and normal prepregnancy BMI. Current GWG guidelines focus on optimizing fetal weight and gestational length. Our findings demonstrate the importance of considering other outcomes when making GWG recommendations for twin pregnancy. KEY POINTS: · Early GWG decreased with increasing BMI category.. · Among women with normal weight, as early GWG increased so did the risk of preeclampsia.. · There was no association between early GWG and preeclampsia among women with overweight or obesity..

The Role of the Natural Course in Causal Analysis.

The average causal effect compares counterfactual outcomes if everyone had been exposed versus if everyone had been unexposed, which can be an unrealistic contrast. Alternatively, we can target effects that compare counterfactual outcomes against the factual outcomes observed in the sample (i.e., we can compare against the natural course). Here, we demonstrate how the natural course can be estimated and used in causal analyses for model validation and effect estimation. Our example is an analysis assessing the impact of taking aspirin on pregnancy, 26 weeks after randomization, in the Effects of Aspirin in Gestation and Reproduction trial (United States, 2006-2012). To validate our models, we estimated the natural course using g-computation and then compared that against the observed incidence of pregnancy. We observed good agreement between the observed and model-based natural courses. We then estimated an effect that compared the natural course against the scenario in which participants assigned to aspirin always complied. If participants had always complied, there would have been 5.0 (95% confidence interval: 2.2, 7.8) more pregnancies per 100 women than was observed. It is good practice to estimate the natural course for model validation when using parametric models, but whether one should estimate a natural course contrast depends on the underlying research questions.

Performance Evaluation of Parametric and Nonparametric Methods When Assessing Effect Measure Modification.

Effect measure modification is often evaluated using parametric models. These models, although efficient when correctly specified, make strong parametric assumptions. While nonparametric models avoid important functional form assumptions, they often require larger samples to achieve a given accuracy. We conducted a simulation study to evaluate performance tradeoffs between correctly specified parametric and nonparametric models to detect effect modification of a binary exposure by both binary and continuous modifiers. We evaluated generalized linear models and doubly robust (DR) estimators, with and without sample splitting. Continuous modifiers were modeled with cubic splines, fractional polynomials, and nonparametric DR-learner. For binary modifiers, generalized linear models showed the greatest power to detect effect modification, ranging from 0.42 to 1.00 in the worst and best scenario, respectively. Augmented inverse probability weighting had the lowest power, with an increase of 23% when using sample splitting. For continuous modifiers, the DR-learner was comparable to flexible parametric models in capturing quadratic and nonlinear monotonic functions. However, for nonlinear, nonmonotonic functions, the DR-learner had lower integrated bias than splines and fractional polynomials, with values of 141.3, 251.7, and 209.0, respectively. Our findings suggest comparable performance between nonparametric and correctly specified parametric models in evaluating effect modification.

Adverse Cardiovascular Events Following Severe Maternal Morbidity.

Severe maternal morbidity (SMM) affects 50,000 women annually in the United States, but its consequences are not well understood. We aimed to estimate the association between SMM and risk of adverse cardiovascular events during the 2 years postpartum. We analyzed 137,140 deliveries covered by the Pennsylvania Medicaid program (2016-2018), weighted with inverse probability of censoring weights to account for nonrandom loss to follow-up. SMM was defined as any diagnosis on the Centers for Disease Control and Prevention list of SMM diagnoses and procedures and/or intensive care unit admission occurring at any point from conception through 42 days postdelivery. Outcomes included heart failure, ischemic heart disease, and stroke/transient ischemic attack up to 2 years postpartum. We used marginal standardization to estimate average treatment effects. We found that SMM was associated with increased risk of each adverse cardiovascular event across the follow-up period. Per 1,000 deliveries, relative to no SMM, SMM was associated with 12.1 (95% confidence interval (CI): 6.2, 18.0) excess cases of heart failure, 6.4 (95% CI: 1.7, 11.2) excess cases of ischemic heart disease, and 8.2 (95% CI: 3.2, 13.1) excess cases of stroke/transient ischemic attack at 26 months of follow-up. These results suggest that SMM identifies a group of women who are at high risk of adverse cardiovascular events after delivery. Women who survive SMM may benefit from more comprehensive postpartum care linked to well-woman care.

Use of a Doubly Robust Machine-Learning-Based Approach to Evaluate Body Mass Index as a Modifier of the Association Between Fruit and Vegetable Intake and Preeclampsia.

The Dietary Guidelines for Americans rely on summaries of the effect of dietary pattern on disease risk, independent of other population characteristics. We explored the modifying effect of prepregnancy body mass index (BMI; weight (kg)/height (m)2) on the relationship between fruit and vegetable density (cup-equivalents/1,000 kcal) and preeclampsia using data from a pregnancy cohort study conducted at 8 US medical centers (n = 9,412; 2010-2013). Usual daily periconceptional intake of total fruits and total vegetables was estimated from a food frequency questionnaire. We quantified the effects of diets with a high density of fruits (≥1.2 cups/1,000 kcal/day vs. <1.2 cups/1,000 kcal/day) and vegetables (≥1.3 cups/1,000 kcal/day vs. <1.3 cups/1,000 kcal/day) on preeclampsia risk, conditional on BMI, using a doubly robust estimator implemented in 2 stages. We found that the protective association of higher fruit density declined approximately linearly from a BMI of 20 to a BMI of 32, by 0.25 cases per 100 women per each BMI unit, and then flattened. The protective association of higher vegetable density strengthened in a linear fashion, by 0.3 cases per 100 women for every unit increase in BMI, up to a BMI of 30, where it plateaued. Dietary patterns with a high periconceptional density of fruits and vegetables appear more protective against preeclampsia for women with higher BMI than for leaner women.

Can Ensemble Machine Learning Improve the Accuracy of Severe Maternal Morbidity Screening in a Perinatal Database?

BACKGROUND: Severe maternal morbidity (SMM) is an important maternal health indicator, but existing tools to identify SMM have substantial limitations. Our objective was to retrospectively identify true SMM status using ensemble machine learning in a hospital database and to compare machine learning algorithm performance with existing tools for SMM identification. METHODS: We screened all deliveries occurring at Magee-Womens Hospital, Pittsburgh, PA (2010-2011 and 2013-2017) using the Centers for Disease Control and Prevention list of diagnoses and procedures for SMM, intensive care unit admission, and/or prolonged postpartum length of stay. We performed a detailed medical record review to confirm case status. We trained ensemble machine learning (SuperLearner) algorithms, which "stack" predictions from multiple algorithms to obtain optimal predictions, on 171 SMM cases and 506 non-cases from 2010 to 2011, then evaluated the performance of these algorithms on 160 SMM cases and 337 non-cases from 2013 to 2017. RESULTS: Some SuperLearner algorithms performed better than existing screening criteria in terms of positive predictive value (0.77 vs. 0.64, respectively) and balanced accuracy (0.99 vs. 0.86, respectively). However, they did not perform as well as the screening criteria in terms of true-positive detection rate (0.008 vs. 0.32, respectively) and performed similarly in terms of negative predictive value. The most important predictor variables were intensive care unit admission and prolonged postpartum length of stay. CONCLUSIONS: Ensemble machine learning did not globally improve the ascertainment of true SMM cases. Our results suggest that accurate identification of SMM likely will remain a challenge in the absence of a universal definition of SMM or national obstetric surveillance systems.

Gestational Diabetes in Twin Versus Singleton Pregnancies With Normal Weight or Overweight Pre-Pregnancy Body Mass Index: The Mediating Role of Mid-Pregnancy Weight Gain.

BACKGROUND: Gestational diabetes might be more common in twin versus singleton pregnancies, yet the reasons for this are unclear. We evaluated the extent to which this relationship is explained by higher mid-pregnancy weight gain within normal weight and overweight pre-pregnancy body mass index (BMI) strata. METHODS: We analyzed serial weights and glucose screening and diagnostic data abstracted from medical charts for twin (n = 1397) and singleton (n = 3117) pregnancies with normal or overweight pre-pregnancy BMI delivered from 1998 to 2013 at Magee-Womens Hospital in Pennsylvania. We used causal mediation analyses to estimate the total effect of twin versus singleton pregnancy on gestational diabetes, as well as those mediated (natural indirect effect) and not mediated (natural and controlled direct effects) by pathways involving mid-pregnancy weight gain. RESULTS: Odds of gestational diabetes were higher among twin pregnancies [odds ratios (ORs) for total effect = 2.83 (95% CI = 1.54, 5.19) for normal weight and 2.09 (95% CI = 1.16, 3.75) for overweight pre pregnancy BMI], yet there was limited evidence that this relationship was mediated by mid-pregnancy weight gain [ORs for natural indirect effect = 1.21 (95% CI = 0.90, 1.24) for normal weight and 1.06 (95% CI = 0.92, 1.21) for overweight pre-pregnancy BMI] and more evidence of mediation via other pathways [ORs for natural direct effect = 2.34 (95% CI = 1.24, 4.40) for normal weight and 1.97 (95% CI = 1.08, 3.60) for overweight pre-pregnancy BMI]. CONCLUSIONS: While twin pregnancies with normal weight or overweight pre-pregnancy BMI experienced higher odds of gestational diabetes versus singletons, most of this effect was explained by pathways not involving mid-pregnancy weight gain.