Adjusting for multiple-misclassified variables in a study using birth certificates

Pregnancy Outcomes after Preeclampsia: The Effects of Interpregnancy Weight Change

OBJECTIVE

This study aimed to examine the effects of interpregnancy weight change on pregnancy outcomes, including recurrent preeclampsia, preterm birth, small-for-gestational age (SGA), large-for-gestational age (LGA), and cesarean delivery, among women with a history of preeclampsia. We also evaluated whether these associations were modified by prepregnancy body mass index (BMI) category in the first pregnancy (BMI < 25 vs. ≥25 kg/m²) and if associations were present among women who maintained a healthy BMI category in both pregnancies.

STUDY DESIGN

We conducted a population-based retrospective cohort study including 15,108 women who delivered their first two nonanomalous singleton live births in Missouri (1989-2005) and experienced preeclampsia in the first pregnancy. We performed Poisson regression with robust error variance to estimate relative risks and 95% confidence intervals for outcomes of interest after controlling for potential confounders.

RESULTS

Interpregnancy weight gain was associated with increased risk of recurrent preeclampsia, LGA, and cesarean delivery. These risks increased in a "dose-response" manner with increasing magnitude of interpregnancy weight gain and were generally more pronounced among women who were underweight or normal weight in the first pregnancy. Interpregnancy weight loss exceeding 1 BMI unit was associated with increased risk of SGA among underweight and normal weight women, while interpregnancy weight loss exceeding 2 BMI units was associated with reduced risk of recurrent preeclampsia among overweight and obese women.

CONCLUSION

Even small changes in interpregnancy weight may significantly affect pregnancy outcomes among formerly preeclamptic women. Appropriate weight management between pregnancies has the potential to attenuate such risks.

KEY POINTS

· Interpregnancy weight change among formerly preeclamptic women significantly affects pregnancy outcomes.. · Interpregnancy weight gain is associated with increased risk of recurrent preeclampsia, large-for-gestational-age and cesarean delivery.. · Interpregnancy weight loss is associated with increased risk of small-for-gestational age and recurrent preeclampsia..

A systematic review of quantitative bias analysis applied to epidemiological research

BACKGROUND

Quantitative bias analysis (QBA) measures study errors in terms of direction, magnitude and uncertainty. This systematic review aimed to describe how QBA has been applied in epidemiological research in 2006-19.

METHODS

We searched PubMed for English peer-reviewed studies applying QBA to real-data applications. We also included studies citing selected sources or which were identified in a previous QBA review in pharmacoepidemiology. For each study, we extracted the rationale, methodology, bias-adjusted results and interpretation and assessed factors associated with reproducibility.

RESULTS

Of the 238 studies, the majority were embedded within papers whose main inferences were drawn from conventional approaches as secondary (sensitivity) analyses to quantity-specific biases (52%) or to assess the extent of bias required to shift the point estimate to the null (25%); 10% were standalone papers. The most common approach was probabilistic (57%). Misclassification was modelled in 57%, uncontrolled confounder(s) in 40% and selection bias in 17%. Most did not consider multiple biases or correlations between errors. When specified, bias parameters came from the literature (48%) more often than internal validation studies (29%). The majority (60%) of analyses resulted in >10% change from the conventional point estimate; however, most investigators (63%) did not alter their original interpretation. Degree of reproducibility related to inclusion of code, formulas, sensitivity analyses and supplementary materials, as well as the QBA rationale.

CONCLUSIONS

QBA applications were rare though increased over time. Future investigators should reference good practices and include details to promote transparency and to serve as a reference for other researchers.

Gestational Weight Gain in the First Two Pregnancies and Perinatal Outcomes in the Second Pregnancy

BACKGROUND

Gestational Weight Gain (GWG) below or above the Institute of Medicine (IOM) recommendations increases the risk of adverse pregnancy outcomes. However, it remains unknown whether the risk of adverse outcomes is affected by GWG in a previous pregnancy. We examined associations between GWG in the index (second) pregnancy and pregnancy outcomes, including preterm delivery and small for gestational age (SGA), while taking into consideration GWG in the first pregnancy.

METHODS

In a population-based cohort study (n = 210 564), using the Missouri maternally-linked birth registry (1989-2005), we used multivariable Poisson regression with robust error variance stratified by prepregnancy body mass index (BMI) to evaluate associations between GWG in the index pregnancy and a composite indicator of GWG in the first and second pregnancies and our outcomes of interest, after controlling for sociodemographic and pregnancy-related confounders.

RESULTS

Associations between GWG in the index pregnancy and pregnancy outcomes were moderated by GWG in the first pregnancy. Despite having GWG within recommendations in the index pregnancy, women had increased risk of preterm delivery and SGA if they had suboptimal GWG in their first pregnancy. Also, women having suboptimal GWG in the index pregnancy had increased risk of preterm delivery only if their GWG in the first pregnancy was also suboptimal.

CONCLUSIONS

The observation that women who have GWG within recommendations in a current pregnancy may still have increased risk of adverse outcomes if they had suboptimal GWG in the first pregnancy has considerable clinical and public health implications.

Racial/Ethnic Differences in Correlates of Spontaneous and Medically-Indicated Late Preterm Births among Adolescents

STUDY OBJECTIVE

To investigate the racial/ethnic differences in the correlates of spontaneous and medically-indicated late preterm birth (LPTB), defined as deliveries between 34 0/7 and 36 6/7 weeks gestation, among US adolescents.

DESIGN

Population-based, retrospective cohort study.

SETTING

Births in the United States to adolescents in 2012.

PARTICIPANTS

Adolescents (younger than 20 years; n = 171,573) who delivered nonanomalous singleton first births between 34 and 44 weeks of gestation.

INTERVENTIONS AND MAIN OUTCOME MEASURES

Bivariate and multivariable logistic regression were used to evaluate the associations between maternal risk factors and spontaneous and medically-indicated LPTB, stratified according to maternal race/ethnicity.

RESULTS

Risk factors for spontaneous LPTB included single marital status among Asian adolescents; no insurance coverage among whites, Asian, and Hispanic adolescents; inadequate prenatal care among all racial/ethnic groups except American Indian, and adequate plus prenatal care among all races/ethnicities; prenatal smoking among whites and black adolescents; insufficient gestational weight gain among all racial/ethnic groups except American Indian; and prepregnancy underweight among white, black, and Hispanic adolescents. Risk factors for medically-indicated LPTB included inadequate prenatal care among white, black, and Hispanic adolescents, and adequate plus prenatal care among all racial/ethnic groups except Asian; insufficient gestational weight gain among white, black, and Hispanic adolescents; and prepregnancy overweight and obesity among white, black, and Hispanic adolescents.

CONCLUSION

Our results show racial/ethnic differences in the correlates of spontaneous and medically-indicated LPTB among US adolescents and support the need for risk-specific interventions among different racial/ethnic groups.

Quantifying and Adjusting for Disease Misclassification Due to Loss to Follow-Up in Historical Cohort Mortality Studies

The purpose of this analysis was to quantify and adjust for disease misclassification from loss to follow-up in a historical cohort mortality study of workers where exposure was categorized as a multi-level variable. Disease classification parameters were defined using 2008 mortality data for the New Zealand population and the proportions of known deaths observed for the cohort. The probability distributions for each classification parameter were constructed to account for potential differences in mortality due to exposure status, gender, and ethnicity. Probabilistic uncertainty analysis (bias analysis), which uses Monte Carlo techniques, was then used to sample each parameter distribution 50,000 times, calculating adjusted odds ratios (ORDM-LTF) that compared the mortality of workers with the highest cumulative exposure to those that were considered never-exposed. The geometric mean ORDM-LTF ranged between 1.65 (certainty interval (CI): 0.50-3.88) and 3.33 (CI: 1.21-10.48), and the geometric mean of the disease-misclassification error factor (εDM-LTF), which is the ratio of the observed odds ratio to the adjusted odds ratio, had a range of 0.91 (CI: 0.29-2.52) to 1.85 (CI: 0.78-6.07). Only when workers in the highest exposure category were more likely than those never-exposed to be misclassified as non-cases did the ORDM-LTF frequency distributions shift further away from the null. The application of uncertainty analysis to historical cohort mortality studies with multi-level exposures can provide valuable insight into the magnitude and direction of study error resulting from losses to follow-up.

Prepregnancy body mass index in a first uncomplicated pregnancy and outcomes of a second pregnancy

OBJECTIVE

This study examined the effect of body mass index (BMI) before a first uncomplicated pregnancy on maternal and fetal outcomes in a subsequent pregnancy, including preterm births, preeclampsia, cesarean delivery, small for gestational age, large for gestational age, and neonatal deaths.

STUDY DESIGN

We conducted a population-based cohort study (n = 121,092) using the Missouri maternally linked birth registry (1989 through 2005). Multivariable binary logistic regression models were fit to estimate odds ratios and 95% confidence intervals for the parameters of interest after controlling for sociodemographic and pregnancy-related confounders in the second pregnancy.

RESULTS

Compared to women with a normal BMI in their first pregnancy, those who were underweight prepregnancy had increased odds for preterm birth by 20% and small for gestational age by 40% in their second pregnancy, while those with prepregnancy obesity had increased odds for large for gestational age, preeclampsia, cesarean delivery, and neonatal deaths in their second pregnancy by 54%, 156%, 85%, and 37%, respectively.

CONCLUSION

Women starting a first pregnancy with suboptimal BMI may be at risk of adverse maternal and fetal outcomes in a subsequent pregnancy, even if their first pregnancy was uncomplicated or if they reached a normal weight by their second pregnancy. The long-term consequences of suboptimal BMI carry considerable public health implications.

Misclassification in administrative claims data: quantifying the impact on treatment effect estimates

Misclassification is present in nearly every epidemiologic study, yet is rarely quantified in analysis in favor of a focus on random error. In this review, we discuss past and present wisdom on misclassification and what measures should be taken to quantify this influential bias, with a focus on bias in pharmacoepidemiologic studies. To date, pharmacoepidemiology primarily utilizes data obtained from administrative claims, a rich source of prescription data but susceptible to bias from unobservable factors including medication sample use, medications filled but not taken, health conditions that are not reported in the administrative billing data, and inadequate capture of confounders. Due to the increasing focus on comparative effectiveness research, we provide a discussion of misclassification in the context of an active comparator, including a demonstration of treatment effects biased away from the null in the presence of nondifferential misclassification. Finally, we highlight recently developed methods to quantify bias and offer these methods as potential options for strengthening the validity and quantifying uncertainty of results obtained from pharmacoepidemiologic research.

Misclassification in administrative claims data: quantifying the impact on treatment effect estimates

Misclassification is present in nearly every epidemiologic study, yet is rarely quantified in analysis in favor of a focus on random error. In this review, we discuss past and present wisdom on misclassification and what measures should be taken to quantify this influential bias, with a focus on bias in pharmacoepidemiologic studies. To date, pharmacoepidemiology primarily utilizes data obtained from administrative claims, a rich source of prescription data but susceptible to bias from unobservable factors including medication sample use, medications filled but not taken, health conditions that are not reported in the administrative billing data, and inadequate capture of confounders. Due to the increasing focus on comparative effectiveness research, we provide a discussion of misclassification in the context of an active comparator, including a demonstration of treatment effects biased away from the null in the presence of nondifferential misclassification. Finally, we highlight recently developed methods to quantify bias and offer these methods as potential options for strengthening the validity and quantifying uncertainty of results obtained from pharmacoepidemiologic research.

Accuracy of prenatal smoking data from Washington State birth certificates in a population-based sample with cotinine measurements

PURPOSE

To assess the accuracy of smoking data in contemporary U.S. birth certificates.

METHODS

We compared data on prenatal smoking as reported on Washington State birth certificates to cotinine measured in archived newborn screening dried blood spots for 200 infants born in 2007 (100 randomly selected from births to self-reported nonsmokers and 100 born to self-reported smokers). We estimated the sensitivity of the birth certificate data to identify prenatal smokers and the precision with which self-identified third trimester smokers report smoking levels.

RESULTS

Infants born to two (2%) mothers who reported they did not smoke during the pregnancy had whole blood cotinine concentrations consistent with active smoking by the mother (sensitivity 85%). Sensitivity of the birth certificate to identify reported smokers who continued to smoke throughout pregnancy was similar (89%). Among self-identified third trimester smokers whose infants' specimens were collected shortly after delivery, Spearman rho between infant cotinine and maternal-reported cigarettes/day in the third trimester was 0.54.

CONCLUSIONS

Birth certificates may represent a viable option for assessing prenatal smoking status, and possibly smoking cessation and level among smokers, in epidemiologic studies sufficiently powered to overcome a moderate amount of exposure measurement error.