(Abstracted from Am J Obstet Gynecol 2024;230:440.e1–440.e13) High reported maternal death rates in the United States have garnered attention and skepticism, especially with reports indicating a near doubling of maternal mortality between 2018 and 2021. Some attribute this increase to changes in reporting criteria for maternal mortality developed in 2003 by the National Center for Health Statistics (NCHS).

Objective: There is uncertainty regarding the effect of the COVID-19 pandemic on population rates of stillbirth. We quantified pandemicassociated changes in stillbirth rates in Canada and the United States. Methods: We carried out a retrospective study that included all live births and stillbirths in Canada and the United States from 2015 to 2020. The primary analysis was based on all stillbirths and live births at >= 20 weeks gestation. Stillbirth rates were analyzed by month, with March 2020 considered to be the month of pandemic onset. Interrupted time series analyses were used to determine pandemic effects. Results: The study population included 18 475 stillbirths and 2 244 240 live births in Canada and 134 883 stillbirths and 22 963 356 live births in the United States (8.2 and 5.8 stillbirths per 1000 total births, respectively). In Canada, pandemic onset was associated with an increase in stillbirths at >= 20 weeks gestation of 1.01 (95% confidence interval [CI] 0.56-1.46) per 1000 total births and an increase in stillbirths at >= 28 weeks gestation of 0.35 (95% CI 0.16-0.54) per 1000 total births. In the United States, pandemic onset was associated with an increase in stillbirths at >= 20 weeks gestation of 0.48 (95% CI 0.22-0.75) per 1000 total births and an increase in stillbirths at >= 28 weeks gestation of 0.22 (95% CI 0.12-0.32) per 1000 total births. The increase in stillbirths at pandemic onset returned to pre -pandemic levels in subsequent months. Conclusion: The COVID-19 pandemic's onset was associated with a transitory increase in stillbirth rates in Canada and the United States.

BACKGROUND: National Vital Statistics System reports show that maternal mortality rates in the United States have nearly doubled, from 17.4 in 2018 to 32.9 per 100,000 live births in 2021. However, these high and rising rates could reflect issues unrelated to obstetrical factors, such as changes in maternal medical conditions or maternal mortality surveillance (eg, due to introduction of the pregnancy checkbox). OBJECTIVE: This study aimed to assess if the high and rising rates of maternal mortality in the United States reflect changes in obstetrical factors, maternal medical conditions, or maternal mortality surveillance. STUDY DESIGN: The study was based on all deaths in the United States from 1999 to 2021. Maternal deaths were identified using the following 2 approaches: (1) per National Vital Statistics System methodology, as deaths in pregnancy or in the postpartum period, including deaths identified solely because of a positive pregnancy checkbox, and (2) under an alternative formulation, as deaths in pregnancy or in the postpartum period, with at least 1 mention of pregnancy among the multiple causes of death on the death certificate. The frequencies of major cause-of-death categories among deaths of female patients aged 15 to 44 years, maternal deaths, deaths due to obstetrical causes (ie, direct obstetrical deaths), and deaths due to maternal medical conditions aggravated by pregnancy or its management (ie, indirect obstetrical deaths) were quantified. RESULTS: Maternal deaths, per National Vital Statistics System methodology, increased by 144% (95% confidence interval, 130-159) from 9.65 in 1999-2002 (n=1550) to 23.6 per 100,000 live births in 2018-2021 (n=3489), with increases occurring among all race and ethnicity groups. Direct obstetrical deaths increased from 8.41 in 1999-2002 to 14.1 per 100,000 live births in 2018-2021, whereas indirect obstetrical deaths increased from 1.24 to 9.41 per 100,000 live births: 38% of direct obstetrical deaths and 87% of indirect obstetrical deaths in 2018-2021 were identified because of a positive pregnancy checkbox. The pregnancy checkbox was associated with increases in less specific and incidental causes of death. For example, maternal deaths with malignant neoplasms listed as a multiple cause of death increased 46-fold from 0.03 in 1999-2002 to 1.42 per 100,000 live births in 2018-2021. Under the alternative formulation, the maternal mortality rate was 10.2 in 1999-2002 and 10.4 per 100,000 live births in 2018-2021; deaths from direct obstetrical causes decreased from 7.05 to 5.82 per 100,000 live births. Deaths due to preeclampsia, eclampsia, postpartum hemorrhage, puerperal sepsis, venous complications, and embolism decreased, whereas deaths due to adherent placenta, renal and unspecified causes, cardiomyopathy, and preexisting hypertension increased. Maternal mortality increased among non-Hispanic White women and decreased among non-Hispanic Black and Hispanic women. However, rates were disproportionately higher among non-Hispanic Black women, with large disparities evident in several causes of death (eg, cardiomyopathy). CONCLUSION: The high and rising rates of maternal mortality in the United States are a consequence of changes in maternal mortality surveillance, with reliance on the pregnancy checkbox leading to an increase in misclassified maternal deaths. Identifying maternal deaths by requiring mention of pregnancy among the multiple causes of death shows lower, stable maternal mortality rates and declines in maternal deaths from direct obstetrical causes.

BACKGROUND: In vitro fertilization (IVF) as a fertility treatment is associated with adverse perinatal outcomes. Racial/ethnic disparity in severe maternal morbidity (SMM) in women who conceived by IVF is understudied. OBJECTIVE: To examine differences in the association between race/ethnicity and SMM between women who conceived spontaneously and those who conceived using IVF. METHODS: We included all singleton live births and stillbirths in the United States, 2016-2021; data were obtained from the National Center for Health Statistics. Maternal race/ethnicity included non-Hispanic White (NHW), non-Hispanic Black (NHB), American Indian and Alaska Native (AIAN), Asian, Pacific Islander (PI), Hispanic, and mixed-race categories. The SMM composite outcome included eclampsia, uterine rupture, peripartum hysterectomy, blood transfusion, and intensive care unit (ICU) admission. We used logistic regression to adjust for potential confounders (such as age, education, parity, prepregnancy body mass index, smoking during pregnancy, chronic hypertension, and preexisting diabetes) and to assess modification of the association between race/ethnicity and SMM by IVF. RESULTS: The study population included 21,585,015 women: 52% were NHW, 15% NHB, 0.8% AIAN, 6% Asian, 0.2% PI, 24% Hispanic, and 2% were of mixed race. IVF was used by 183,662 (0.85%) women; the rate of the SMM composite outcome was 18.5 per 1000 deliveries and 7.9 per 1000 deliveries in the IVF and spontaneous conception groups, respectively (unadjusted rate ratio 2.34, 95% confidence interval [CI] 2.26-2.43). In women with spontaneous conception, NHB, Asian and mixed-race women had elevated odds of SMM compared with NHW women (adjusted odds ratio [aOR]=1.39, 95% CI 1.37-1.41; aOR=1.04, 95% CI 1.02-1.07; and aOR=1.42, 95% CI 1.38-1.46, respectively). Racial/ethnic disparities in SMM and its components were not different between the IVF and spontaneous conception groups for the mixed-race category. NHB and Hispanic women had significantly higher aORs for uterine rupture/intrapartum hysterectomy compared with NHW women in the IVF group, while Asian women had a higher aOR for ICU admission compared with NHW women in the IVF group. CONCLUSION: Women who conceived by IVF have a greater than two-fold higher risk of SMM and this higher risk is evident across all racial/ ethnic groups. However, NHB and Hispanic women who conceived by IVF had a higher risk of uterine rupture/hysterectomy, and Asian women who conceived by IVF had a higher risk of ICU admission. Our results warrant further investigation examining pregnancy and postpartum care issues among racial/ethnic minority women who conceive using IVF.

OBJECTIVE:To identify the factors underlying the recent increase in maternal mortality ratios (maternal deaths per 100,000 live births) in the United States.METHODS:We carried out a retrospective study with data on maternal deaths and live births in the United States from 1993 to 2014 obtained from the birth and death files of the Centers for Disease Control and Prevention. Underlying causes of death were examined between 1999 and 2014 using International Classification of Diseases, 10th Revision (ICD-10) codes. Poisson regression was used to estimate maternal mortality rate ratios (RRs) and 95% confidence intervals (CIs) after adjusting for the introduction of a separate pregnancy question and the standard pregnancy checkbox on death certificates and adoption of ICD-10.RESULTS:Maternal mortality ratios increased from 7.55 in 1993, to 9.88 in 1999, and to 21.5 per 100,000 live births in 2014 (RR 2014 compared with 1993 2.84, 95% CI 2.49-3.24; RR 2014 compared with 1999 2.17, 95% CI 1.93-2.45). The increase in maternal deaths from 1999 to 2014 was mainly the result of increases in maternal deaths associated with two new ICD-10 codes (O26.8, ie, primarily renal disease; and O99, ie, other maternal diseases classifiable elsewhere); exclusion of such deaths abolished the increase in mortality (RR 1.09, 95% CI 0.94-1.27). Regression adjustment for improvements in surveillance also abolished the temporal increase in maternal mortality ratios (adjusted maternal mortality ratios 7.55 in 1993, 8.00 per 100,000 live births in 2013; adjusted RR 2013 compared with 1993 1.06, 95% CI 0.90-1.25).CONCLUSION:Recent increases in maternal mortality ratios in the United States are likely an artifact of improvements in surveillance and highlight past underestimation of maternal death. Complete ascertainment of maternal death in populations remains a challenge even in countries with good systems for civil registration and vital statistics.

Background:Reports of high and rising maternal mortality ratios (MMR) in the United States have caused serious concern. We examined spatiotemporal patterns in cause-specific MMRs, in order to obtain insights into the cause for the increase. Methods:The study included all maternal deaths recorded by the Centers for Disease Control and Prevention from 1999 to 2021. Changes in overall and cause-specific MMRs were quantified nationally; in low-vs high-MMR states (i.e., MMRs <20 vs ≥26 per 100,000 live births in 2018-2021); and in California vs Texas (populous states with low vs high MMRs). Cause-specific MMRs included those due to unambiguous causes (e.g., selected obstetric causes such as pre-eclampsia/eclampsia) and less-specific/potentially incidental causes (e.g., "other specified pregnancy-related conditions", chronic hypertension, and malignant neoplasms). Findings:MMRs increased from 9.60 (n = 1543) in 1999-2002 to 23.5 (n = 3478) per 100,000 live births in 2018-2021. The temporal increase in MMRs was smaller in low-MMR states (from 7.82 to 14.1 per 100,000 live births) compared with high-MMR states (from 11.1 to 31.4 per 100,000 live births). MMRs due to selected obstetric causes decreased to a similar extent in low-vs high-MMR states, whereas the increase in MMRs from less-specific/potentially incidental causes was smaller in low- vs high-MMR states (MMR ratio (RR) 5.57, 95% CI 4.28, 7.25 vs 7.07, 95% CI 5.91, 8.46), and in California vs Texas (RR 1.67, 95% CI 1.03, 2.69 vs 10.8, 95% CI 6.55, 17.7). The change in malignant neoplasm-associated MMRs was smaller in California vs Texas (RR 1.21, 95% CI 0.08, 19.3 vs 91.2, 95% CI 89.2, 94.8). MMRs from less-specific/potentially incidental causes increased in all race/ethnicity groups. Interpretation:Spatiotemporal patterns of cause-specific MMRs, including similar reductions in unambiguous obstetric causes of death and variable increases in less-specific/potentially incidental causes, suggest misclassified maternal deaths and overestimated maternal mortality in some US states. Funding:This work received no funding.

Importance The prevalence of overweight and obesity (body mass index [BMI] ≥25) has increased globally, and high BMI has been linked to higher rates of twin birth. However, evidence from large population-based studies is lacking; the issue needs careful study, as women with obesity are also more likely to use assisted reproductive technology (ART), which frequently results in twin pregnancy. Objective To examine the association between BMI and twin birth and the role of ART as a potential mediator in this association. Design, Setting, and Participants This retrospective cohort study included all live births and stillbirths with gestational age of 20 weeks or longer in British Columbia, Canada, from 2008 to 2020, using data from the British Columbia Perinatal Database Registry. Data analysis was conducted from November 2022 to June 2023. Exposures Prepregnancy BMI, calculated as weight in kilograms divided by height in meters squared, and use of ART. Main Outcomes and Measures The study assessed whether prepregnancy BMI is associated with the rate of twin vs singleton delivery and whether this association is explained by the differential use of ART in women with obesity. Results A total of 524 845 deliveries at 20 weeks’ or longer gestation occurred in British Columbia during the study period, and 392 046 women had complete data on prepregnancy BMI. The median (IQR) age was 31.4 (27.7-35.0) years, approximately half were nulliparous (243 443 [46.4%]) and less than 10% smoked during pregnancy (36 894 [7.1%]). Overall, 8295 women had a twin delivery (15.8 per 1000 deliveries), and rates per 1000 deliveries by prepregnancy BMI categories were 11.9 (underweight), 15.1 (normal), 16.0 (overweight), 16.0 (obesity class I), 16.7 (obesity class II), and 18.9 (obesity class III). After adjustment for other covariates, women with underweight had relatively 16% fewer twins compared with women with normal BMI (adjusted risk ratio [aRR], 0.84; 95% CI, 0.74-0.95), while women with overweight, class I obesity, class II obesity, and class III obesity had 14% (aRR, 1.14; 95% CI, 1.07-1.21), 16% (aRR, 1.16; 95% CI, 1.06-1.27), 17% (aRR, 1.17; 95% CI, 1.02-1.34), and 41% higher rates (aRR, 1.41; 95% CI, 1.19-1.66), respectively. The proportion of women who conceived by ART increased with increasing BMI, and ART was associated with nearly a 12-fold higher rate of twin delivery (aRR, 11.80; 95% CI 11.10-12.54). ART explained about a quarter of the association between obesity class I and II and twin delivery (eg, obesity class I, 23% mediated; 95% CI, 7%-39% mediated), but none of this association was mediated by ART in women with class III obesity. Conclusions and relevance In this cohort study of 524 845 births, the rate of twin birth increased with increasing prepregnancy BMI. In women with a BMI between 30 and 40, approximately one-quarter of this association was explained by higher use of ART; however, there was no evidence of such mediation in women with BMI of 40 or greater.

BACKGROUNDSurvival analysis methods are increasingly used to model the gestational age-specific risk of perinatal phenomena such as stillbirth.OBJECTIVESTo compare two types of survival analysis models, and highlight differences by estimating the relationships between pre-pregnancy BMI and gestational age-specific rates of stillbirth.METHODSThe study was based on singleton live births and stillbirths in the United States in 2016-2017, with data obtained from the natality and fetal death files of the National Center for Health Statistics. We compared Cox regression versus piecewise exponential additive mixed models (PAMMs) for modelling the relationship between BMI and stillbirth across gestational age. In a second analysis, we illustrated the performance of both models for assessing the relationship between the trimester-specific number of cigarettes smoked, a time-dependent covariate, and stillbirth.RESULTSThe study population included 7,567,316 births, of which 42,739 were stillbirths (5.6 per 1000 total births). Stillbirth rates increased with increasing pre-pregnancy BMI and increasing gestational age. In analyses with BMI as a categorical variable, the Cox model and PAMM models yielded similar results. Analyses of BMI as a continuous variable also showed similar results when BMI associations were assumed to be linear, and the changes in gestational age-specific rates were modelled parametrically. However, results differed slightly when PAMMs, modelled with data-driven approaches, were used to estimate changes in BMI effects across gestational age; PAMMs provided a more nuanced modelling of time-varying effects. PAMM models showed an approximately linear increase in the effect of smoking on stillbirth with increasing gestational age.CONCLUSIONSFor survival analyses using the foetuses-at-risk approach, PAMMs provide a valuable alternative to the traditional Cox model, with increased modelling flexibility when proportional hazards assumptions are violated.

Background:Recommendations for deliveries of pregnant patients with a previous cesarean delivery and the type of hospitals deemed safe for these deliveries have evolved in recent years, although no studies have examined hospital factors and associated safety. We sought to evaluate maternal and neonatal outcomes among patients with a previous cesarean delivery by hospital tier and volume. Methods:We carried out an ecological study of singleton live births delivered at term gestation to patients with a previous cesarean delivery in all Canadian hospitals (excluding Quebec), 2013-2019. We obtained data from the Discharge Abstract Database of the Canadian Institute for Health Information. The primary outcomes were severe maternal morbidity or mortality (SMMM), and serious neonatal morbidity or mortality (SNMM). We used regression modelling to examine hospital tier (tier 4 hospitals being those that provide the highest level of care) and volume; we also identified hospitals with high rates of SMMM and SNMM using within-tier comparisons and comparisons with the overall rate. Results:We included 235 442 deliveries to patients with a previous cesarean delivery; SMMM and SNMM rates were 14.6 per 1000 deliveries and 4.6 per 1000 live births, respectively. Among patients with a parity of 1, SMMM rates were lower in tier 1 hospitals (adjusted incidence rate ratio [IRR] 0.68, 95% confidence interval [CI] 0.52-0.89) and higher in tier 4 hospitals (adjusted IRR 1.41, 95% CI 1.05-1.91) than in tier 2 hospitals; SNMM rates did not differ by hospital tier. Rates of SNMM increased with increasing hospital volume (adjusted IRR 1.02, 95% CI 1.00-1.04) and increasing rates of vaginal birth after cesarean delivery (adjusted IRR 1.02, 95% CI 1.01-1.04). Most hospitals had relatively low SMMM and SNMM rates, although a few hospitals in each tier and volume category had significantly higher rates than others. Interpretation:Adverse maternal and neonatal outcomes among patients with a previous cesarean delivery showed no clear pattern of decreasing SMMM and SNMM with increasing tiers of service and hospital volume. All hospitals, irrespective of tier or size, should continually review their rates of adverse maternal and neonatal outcomes.

BACKGROUND The assessment of birthweight for gestational age and the identification of small- and large-for-gestational age (SGA and LGA) infants remain contentious, despite the recent creation of the Intergrowth 21st Project and World Health Organisation (WHO) birthweight-for-gestational age standards. OBJECTIVE We carried out a study to identify birthweight-for-gestational age cut-offs, and corresponding population-based, Intergrowth 21st and WHO centiles associated with higher risks of adverse neonatal outcomes, and to evaluate their ability to predict serious neonatal morbidity and neonatal mortality (SNMM) at term gestation. METHODS The study population was based on non-anomalous, singleton live births between 37 and 41 weeks' gestation in the United States from 2003 to 2017. SNMM included 5-min Apgar score <4, neonatal seizures, need for assisted ventilation, and neonatal death. Birthweight-specific SNMM was modelled by gestational week using penalised B-splines. The birthweights at which SNMM odds were minimised (and higher by 10%, 50% and 100%) were estimated, and the corresponding population, Intergrowth 21st, and WHO centiles were identified. The clinical performance and population impact of these cut-offs for predicting SNMM were evaluated. RESULTS The study included 40,179,663 live births and 991,486 SNMM cases. Among female singletons at 39 weeks' gestation, SNMM odds was lowest at 3203 g birthweight, and 10% higher at 2835 g and 3685 g (population centiles 11th and 82nd, Intergrowth centiles 17th and 88th and WHO centiles 15th and 85th). Birthweight cut-offs were poor predictors of SNMM, for example, the cut-offs associated with 10% and 50% higher odds of SNMM among female singletons at 39 weeks' gestation resulted in a sensitivity, specificity, and population attributable fraction of 12.5%, 89.4%, and 2.1%, and 2.9%, 98.4% and 1.3%, respectively. CONCLUSIONS Reference- and standard-based birthweight-for-gestational age indices and centiles perform poorly for predicting adverse neonatal outcomes in individual infants, and their associated population impact is also small.

Hocquette and Zeitlin1 and Grantz and Zhang2 highlight a few issues with regard to our paper3 on the performance of birthweight-for-gestational age charts and birthweight centiles at term gestation. In this counterpoint, we discuss the points raised, including the choice of outcome for evaluating birthweight-for-gestational age charts, the potential impact of obstetrical intervention(s) on such assessments, and the emerging perspective on the utility of abnormal fetal and newborn growth indices. The choice of outcome for assessing fetal and newborn weight-for-gestational age charts requires consideration of the purpose of monitoring fetal and newborn growth status. In fact, the rationale varies depending on when growth is assessed, whether in utero or at birth. In utero estimation of fetal weight-for-gestational age provides information on general fetal health status and malnutrition, including restricted and excessive growth. Such assessment occurs in real-time and permits remedial intervention, although assessment is limited by potential inaccuracies in the estimation of fetal weight. On the other hand, birthweight-for-gestational age enables several assessments, including (i) a retrospective assessment of the cumulative in-utero growth experience; (ii) a cross-sectional assessment of general health status at birth; (iii) setting prognosis with regard to neonatal complications (e.g. hypoglycaemia and hyperbilirubinaemia); and (iv) obtaining a population perspective which addresses newborn growth distributions in different subpopulations. Whereas the retrospective outlook deals with obstetrical issues (e.g. by relating pregnancy complications to growth status at birth), and the prognostic viewpoint addresses the neonatal outlook (e.g. by relating growth status at birth to subsequent complications), the cross-sectional assessment at birth permits a more accurate quantification of the relationship between newborn growth status and general health status (as opposed to the in utero assessment since weight and health status are more accurately ascertained in infants). The utility of the latter assessment is predicated on two assumptions: (i) that health status at birth is best assessed using immediate findings (e.g. low 5-min Apgar score) and delayed manifestations (e.g. neonatal seizures or death) to comprehensively identify overt and hidden health conditions; and (ii) that the relation between birthweight-for-gestational age and newborn health status is generalisable, at least partly, to the estimated fetal weight-for-age and health status relation. No single outcome can address all the purposes of monitoring fetal and newborn growth, and assessments of (multi-dimensional) general health status are best achieved using a composite outcome. For our study,3 which focused on the assessment of health status at birth, we used a composite outcome that included 5-min Apgar <4, need for assisted ventilation, neonatal seizures, and neonatal death. Hocquette and Zeitlin1 advocate the evaluation of fetal and newborn growth charts based on neonatal morbidity and mortality but restrict that evaluation to small for gestational age (SGA) and large for gestational age (LGA) infants. However, as they point out,1 such evaluation excludes neonatal morbidity and mortality among appropriate-for-gestational age (AGA) infants. This is problematic because the majority of neonatal morbidity and mortality occurs among AGA infants,3 and the restriction to fetuses or infants deemed SGA and LGA fails to address the health status of all fetuses or infants. Grantz and Zhang2 also highlight the need to assess specific morbidity such as neonatal hypoglycaemia. Using information on infant growth status at birth for predicting neonatal hypoglycaemia and other morbidity is a legitimate clinical objective. However, we suspect that SGA and LGA, while risk factors for hypoglycaemia, will fail to identify the majority of hypoglycaemia cases, which will likely occur among the substantially larger population of AGA infants. One epidemiologic study4 that routinely screened 3595 newborn infants for early hypoglycaemia showed that only 13 of 124 infants with a blood glucose <40 mg/dL were SGA and 16 were LGA, while 95 were AGA. Both commentaries1, 2 provide a cautionary note regarding potential modification of the association between birthweight-for-gestational age and adverse perinatal outcomes by obstetrical and other interventions. This is a pervasive problem in non-experimental perinatal research, although some relationships are likely more impacted than others. For instance, evidence suggests that preeclampsia rates in specific populations have decreased in recent years owing to increases in iatrogenic early delivery. On the other hand, it is uncommon for iatrogenic early delivery to be based solely on SGA status. Both the GRIT randomised trial, which contrasted immediate or deferred delivery following signs of impaired foetal health in the presence of suspected growth restriction at 24–36 weeks' gestation, and the DIGITAT randomised trial, which examined the effect of labour induction versus expectant management for suspected intrauterine growth restriction at 36 weeks' gestation, showed no difference in neonatal morbidity/mortality or long-term developmental outcomes. Current clinical guidelines5 advocate iatrogenic early delivery only in the small subset of cases in whom suspected foetal growth restriction is associated with additional risk factors (such as ultrasound demonstrated absent or reversed umbilical artery blood flow), as this is associated with a reduction in perinatal death. Foetal growth restriction and excessive growth are considered 'pathological' conditions, although they are defined in abstract terms—as conditions affecting foetuses that fail to reach their biological growth potential or who exceed their growth potential, respectively, for their gestational age. Operationalisation of these concepts typically involves the use of foetal growth indices, namely, SGA and LGA, based on weight-for-gestational age cut-offs obtained from references/standards. However, recent studies (e.g. 3, 6, 7) have raised fundamental questions about SGA and LGA: Do they define diseases? Should they be used as screening criteria? Are they predictors of neonatal morbidity and mortality? Or, as some experts have argued—is it time to abandon SGA altogether?8 SGA and LGA fetuses and infants comprise a heterogeneous group with diverse aetiologies, including chromosomal abnormalities, other congenital anomalies, placental dysfunction, and constitutionally small (normal) foetuses and infants. Although such heterogeneity means that abnormal foetal and newborn growth indices do not represent diseases (which are characterised by an overt or hidden somatic anomaly9), a case can be made that they represent disease heuristics, that is, they identify individuals at high risk for disease based on empirically derived biomarker cut-offs. Hypertension and osteoporosis are examples of such heuristically defined diseases,9 which according to contemporary medical practice warrant specific therapy. Alternatively, it could be argued that abnormal fetal and newborn growth indices can be used as a first-step surveillance screen to identify fetuses and newborns at high risk for perinatal mortality or serious neonatal morbidity. This implies that screen-positive individuals are at risk of serious morbidity or mortality, but true- and false-positive individuals need to be identified through a second-stage diagnostic procedure. Unfortunately, as many recent studies have shown (e.g. 6, 7) and our study3 confirms, SGA and LGA indices fit neither the disease nor the screening criteria profile as they cannot discriminate between fetuses and infants who are, and who are not, at high risk of perinatal death or serious neonatal morbidity. The ability of a dichotomised biomarker to discriminate between individuals who have (or will develop) a disease, and those who do not, can be illustrated by contrasting systolic hypertension in relation to stroke death versus birthweight-for-gestational age in relation to serious neonatal morbidity or neonatal mortality (SNMM). Systolic hypertension is a risk factor for stroke: the 12-year follow-up of the Multiple Risk Factor Intervention Trial10 showed that stroke mortality was 4.2 times higher among males with a systolic blood pressure (SBP) of 140–149 mm Hg, 6.5 times higher among males with a SBP of 150–159 mm Hg, etc., compared with those with a SBP <110 mm Hg. Similarly, low birthweight-for-gestational age is a risk factor for SNMM: in our study,3 SNMM rates were 1.6 times higher among female singleton infants at 39 weeks' gestation with birthweights between 2283 and 2509 g, and 2.9 times higher among infants whose birthweights were <2283 g (compared with infants whose birthweights were 2850 to 3670 g). Figure 1A shows that the distribution of SBP among adults who suffered a stroke death differs substantially from the SBP distribution among all adults. In contrast, Figure 1B shows that the birthweight-for-gestational age distribution of infants with low 5-min Apgar scores differs only marginally from the same distribution among all infants. The overlapping distributions of birthweight-for-gestational age among infants with a low versus normal 5-min Apgar mean that birthweight-for-gestational age cut-offs have a limited ability to discriminate between infants at high versus low risk for such SNMM. Nevertheless, these differences only partly address the reasons why hypertension is viewed as a 'disease', while SGA and LGA face a more uncertain status. Pertinent issues in this context include the strong relationship between hypertension and other common diseases of older adults (including coronary heart disease and death from coronary heart disease10), and also evidence showing reductions in stroke and coronary heart disease mortality following anti-hypertensive therapy. This contrasts with the results of the GRIT and DIGITAT trials, which failed to show the benefit of intervention for SGA in terms of short- and long-term pregnancy and child outcomes. The accumulating evidence on abnormal fetal and newborn growth indices shows that growth centiles are 'dose-dependent' predictors of perinatal mortality and serious neonatal morbidity, although they perform poorly when used in isolation as disease proxies or screening criteria (e.g. 3, 6, 7). Nevertheless, estimated fetal weight and birthweight centiles in multivariable prediction functions3 could aid in the accurate identification of compromised fetuses and newborns at high risk of perinatal death or serious neonatal mortality, and facilitate the rational use of iatrogenic early delivery,8 and intensive neonatal care. Additionally, scientific and clinical communication and universal use of such multivariable prognostic functions would be facilitated if the same estimated fetal weight and birthweight-for-gestational age charts were used globally. KSJ and SJ proposed the response and wrote the first draft of the Counterpoint. JF, SL and MSK provided critically feedback and all authors approved the final version of the manuscript. The authors declare that they have no conflicts of interest in connection with this artilce. The data that support the findings of this study are openly available in the cited manuscripts (Reference 10) and in the NCHS linked births-infant death files (https://www.cdc.gov/nchs/data_access/vitalstatsonline.htm).

OBJECTIVE: To quantify pandemic-related changes in obstetric intervention and perinatal outcomes in the United States. METHODS: We carried out a retrospective study of all live births and fetal deaths in the United States, 2015–2021, with data obtained from the natality, fetal death, and linked live birth–infant death files of the National Center for Health Statistics. Analyses were carried out among all singletons; singletons of patients with prepregnancy diabetes, prepregnancy hypertension, and hypertensive disorders of pregnancy; and twins. Outcomes of interest included preterm birth, preterm labor induction or preterm cesarean delivery, macrosomia, postterm birth, and perinatal death. Interrupted time series analyses were used to estimate changes in the prepandemic period (January 2015–February 2020), at pandemic onset (March 2020), and in the pandemic period (March 2020–December 2021). RESULTS: The study population included 26,604,392 live births and 155,214 stillbirths. The prepandemic period was characterized by temporal increases in preterm birth and preterm labor induction or cesarean delivery rates and temporal reductions in macrosomia, postterm birth, and perinatal mortality. Pandemic onset was associated with absolute decreases in preterm birth (decrease of 0.322/100 live births, 95% CI 0.506–0.139) and preterm labor induction or cesarean delivery (decrease of 0.190/100 live births, 95% CI 0.334–0.047) and absolute increases in macrosomia (increase of 0.046/100 live births), postterm birth (increase of 0.015/100 live births), and perinatal death (increase of 0.501/1,000 total births, 95% CI 0.220–0.783). These changes were larger in subpopulations at high risk (eg, among singletons of patients with prepregnancy diabetes). Among singletons of patients with prepregnancy diabetes, pandemic onset was associated with a decrease in preterm birth (decrease of 1.634/100 live births) and preterm labor induction or cesarean delivery (decrease of 1.521/100 live births) and increases in macrosomia (increase of 0.328/100 live births) and perinatal death (increase of 9.840/1,000 total births, 95% CI 3.933–15.75). Most changes were reversed in the months after pandemic onset. CONCLUSION: The onset of the coronavirus disease 2019 (COVID-19) pandemic was associated with a transient decrease in obstetric intervention (especially preterm labor induction or cesarean delivery) and a transient increase in perinatal mortality.

Over three decades ago, Arnold Relman, the then Editor of the N Engl J Med, predicted a 'third revolution in medical care'.1 Relman's thesis was that a focus on the quality of services and cost control would lead to dramatic improvements in health care. Such 'assessment and accountability' would be made possible by 'linking medical management decisions to new, systematic information about outcomes of treatment' using computerised health information systems.1 Although Relman's prediction has not come to pass, digital information systems that capture vast amounts of medical and related data are now on the verge of revolutionising the work of perinatal and other epidemiologists. This issue of Paediatric and Perinatal Epidemiology includes two articles that highlight recent developments in health information systems: Suárez-Idueta et al.2 present the results of linking live birth and death registration data (for infants and children <5 years of age) in Mexico, while Johansson et al.3 discuss avant-garde population-based database linkages that bring together detailed clinical and related information on pregnant women in the Stockholm-Gotland cohort. The linkage of 24 million live births to deaths in infancy and early childhood represents a significant epidemiologic success for Mexico, despite challenges related to the underestimation of deaths, and unlinked deaths.2 A similar attempt in Canada in the 1990s resulted in a 25% rate of unlinked infant deaths in Ontario, the largest province in Canada.4 Several investigations aimed at identifying the cause for the missing birth registrations in the vital statistics data proved unsuccessful, and ultimately led to a completely revamped perinatal program in Ontario, and live birth-infant death linkages using an alternative hospitalisation database.5 The previous development and recent consolidation of detailed database linkages in the Stockholm-Gotland cohort are highly inspirational even against the Swedish backdrop, where longstanding deterministic linkages between databases (based on unique personal identifiers) have already yielded numerous epidemiologic insights. The availability of longitudinal clinical, laboratory and other information on the mother, fetus and infant will provide boundless opportunities for groundbreaking epidemiologic research. The explosion in detailed health information within databases and through database linkages has led to many different disciplines entering a field that has been, until recently, the almost exclusive prerogative of epidemiologists. Undoubtedly, the involvement of students of health informatics and data science (among others) in health research will facilitate and enrich medical research. The increasing use of machine learning techniques, and artificial intelligence, more generally, will provide a substantial impetus to diagnostic and non-causal prognostic research both by epidemiologists and other professionals. The benefits of an influx of new researchers into the health research arena notwithstanding, inter-disciplinary semantic and methodologic differences could pose communication and other challenges. As members of the discipline traditionally involved with analysing and interpreting health data using non-experimental designs, epidemiologists have a duty to help manage this expansion in professional diversity. At the very least, epidemiologists have an obligation to introduce newcomers to the lessons learned and caveats derived from decades of working with non-experimental data. One crucial general lesson is the importance of integrating substantive (clinical) understanding and methodologic principles, while arguably the most outstanding particular caveat is the problem of confounding by indication. This latter phenomenon haunts non-experimental studies of therapeutic efficacy when the outcome of interest is an intended effect.6, 7 It is generally accepted that the efficacy of therapy, whether drug, surgery, or other intervention, is best assessed through randomised trials because confounding by indication in non-experimental studies is generally considered intractable. Higher maternal mortality rates following caesarean delivery (as compared with vaginal delivery) epitomise this issue. The Sixth Report of the Confidential Enquiry into Maternal and Child Health in the United Kingdom8 highlighted this problem of confounding by indication by stating that 'it is almost impossible to disentangle the consequences of a caesarean section from the indication for the operation'. The bias of confounding by indication is also evident in the crude positive associations between admission to intensive care units (compared with admission to general wards) and death; between people with hypertension on medication (compared with normotensive people and even those with hypertension not on medication) and stroke and between hospital births (compared with home births) and perinatal death.6, 7 One of the more exciting developments associated with databases and linked database research is the ability to examine clinically and socially relevant determinants and outcomes that may be otherwise difficult to study. Rare outcomes and outcomes that require short- or long-term follow-up can be more easily studied by accessing large population databases or by linking different databases. Successive pregnancy and sibling studies are a particular genre of perinatal research that exemplify the utility of such database linkages.9, 10 The population-based nature of some databases can address problems that may arise due to more selective inclusion into a study population, and the large study sizes, which typify many such databases, are also a clear benefit. Access to information on a given determinant, confounder or outcome that may be more accurate or complete is another potential advantage of linked database research. Figure 1A shows how linking the birth database to subsequent hospitalisation and outpatient-visit databases in Sweden provides a more complete picture of congenital malformations. Figure 1B shows the frequency of cyanotic congenital heart disease infant deaths in the United States, with more deaths of infants whose heart defect was not diagnosed at birth compared with the death of infants whose heart defect was diagnosed at birth. These figures provide a sense of the misclassification that occurs in studies, which rely solely on birth data for ascertaining congenital anomaly status. The general problem of misclassification of determinant and/or outcome status is not unique to research using databases and is well understood. Studies using the information on congenital anomalies at birth and the study by Suárez-Idueta et al.2 will yield reasonable and useful estimates of association if misclassification of determinant/outcome status is non-differential (with estimates biased towards the null). On the other hand, the need to obtain accurate quantitative information on confounders represents a more crucial requirement as the distortion in effect estimates given residual confounding is less predictable (see below). Attempts at including data such as those on socioeconomic, environmental and behavioural factors in the Stockholm-Gotland cohort are therefore a prudent step, and database managers elsewhere would be wise to follow suit. Large databases tend to include some transcription and other related errors, and implausible values for all relevant data elements need to be addressed at the outset. Additionally, the idiosyncrasies of specific databases need to be understood if the results are to be interpreted correctly. Figure 2 presents the frequency distribution of birthweights obtained from birth databases in Sweden and the United States and shows end-digit preference in Sweden and end-digit and ounce preference in the United States. Although the phenomena illustrated in this example are trivial and will not impact the results of most studies, an awareness of such issues is required for the occasional situation when such quirks can lead to problems. One potential problem with database research arises in connection with attempts to answer causal questions non-experimentally. Whereas research has traditionally required pre-specification of the object of study and ad hoc collection of data on the determinant, confounders and outcome status, database studies use data collected for a purpose not directly related to the object of any particular research study. Thus, databases may contain accurate information on the determinant and outcome, and only include poor quality or absent information on key confounders. Unlike non-differential misclassification of determinant/ outcome status, which typically has predictable consequences, misclassification of confounder status poses a greater threat to validity. Although recent developments in the area of quantitative bias analysis permit an assessment of threats to validity due to unaddressed confounding, it is debatable whether researchers should attempt to answer causal questions when the database does not contain accurate information on key confounders. Two other inter-related issues involving database research require mentioning, namely, ethics/confidentiality and data access. Ethics and confidentiality issues, while extremely important, are currently managed variably, with some countries and institutions allowing free access to anonymized data repositories, others providing more restricted access, and the scientific community increasingly advocating for data sharing along with publication. Data access issues have to be resolved by balancing the public good that could follow increased access with the risk of confidentiality being breached because of unconstrained access. Databases and linked databases (which are literally matrices, hence the title) are ushering in an era where epidemiologists and other professionals will be required to work with vast quantities of detailed clinical and related health data. Although many routine, big data analysis tasks will likely be automated in the future, the automated algorithms and new studies will require a multi-disciplinary integration of substantive and methodologic inputs. The adage that 'knowledge is power' requires data to be first processed into information and information to be distilled into knowledge. Neda Razaz is an Assistant Professor in the Clinical Epidemiology Division, at Karolinska Institutet in Stockholm, Sweden. Her research program aims to understand the role of maternal and paternal chronic illness during pregnancy, neurodevelopment, and other long-term outcomes in childhood and early adulthood. Dr Razaz serves as a junior editor of Paediatric and Perinatal Epidemiology. Sid John is a Staff Researcher in the Department of Obstetrics and Gynaecology at the University of British Columbia, Vancouver, Canada. He has a background in biological systems engineering, and his current research is on gestational diabetes and fetal growth. K.S. Joseph is a Professor in the Department of Obstetrics and Gynaecology at the University of British Columbia, Vancouver, Canada, and his interests include maternal, fetal and infant health and health services. Dr Joseph serves on the editorial board of Paediatric and Perinatal Epidemiology.

Despite recognition regarding the need to balance medical and social risk, there is little evidence in the literature regarding the appropriate delivery hospital for women with a previous cesarean delivery (PCD). We compared severe maternal morbidity/mortality (SMM) and serious neonatal morbidity/mortality (NMM) among deliveries to women with a PCD by hospital in Canada.

Linked article: This is a mini commentary on PWG Tennant et al., pp. 82–89 in this issue. To view this article visit https://doi.org/10.1111/1471-0528.16906

Comparative statistics routinely published by the Organization for Economic Cooperation and Development have regularly characterized Canadian rates of maternal trauma as being the highest among member countries both among spontaneous vaginal deliveries (SVDs) and among operative vaginal deliveries (OVDs). We sought to quantify the associations between hospital-level rates of maternal trauma among SVDs and OVDs.

Background: Operative vaginal delivery (OVD) is considered safe if carried out by trained personnel. However, opportunities for training in OVD have declined and, given these shifts in practice, the safety of OVD is unknown. We estimated incidence rates of trauma following OVD in Canada, and quantified variation in trauma rates by instrument, region, level of obstetric care and institutional OVD volume. Methods: We conducted a cohort study of all singleton, term deliveries in Canada between April 2013 and March 2019, excluding Quebec. Our main outcome measures were maternal trauma (e.g., obstetric anal sphincter injury, high vaginal lacerations) and neonatal trauma (e.g., subgaleal hemorrhage, brachial plexus injury). We calculated adjusted and stabilized rates of trauma using mixed-effects logistic regression. Results: Of 1 326 191 deliveries, 38 500 (2.9%) were attempted forceps deliveries and 110 987 (8.4%) were attempted vacuum deliveries. The maternal trauma rate following forceps delivery was 25.3% (95% confidence interval [CI] 24.8%-25.7%) and the neonatal trauma rate was 9.6 (95% CI 8.6-10.6) per 1000 live births. Maternal and neonatal trauma rates following vacuum delivery were 13.2% (95% CI 13.0%-13.4%) and 9.6 (95% CI 9.0-10.2) per 1000 live births, respectively. Maternal trauma rates remained higher with forceps than with vacuum after adjustment for confounders (adjusted rate ratio 1.70, 95% CI 1.65-1.75) and varied by region, but not by level of obstetric care. Interpretation: In Canada, rates of trauma following OVD are higher than previously reported, irrespective of region, level of obstetric care and volume of OVD among hospitals. These results support a reassessment of OVD safety in Canada.

(Abstracted from Paediatr Perinat Epidemiol 2022;36:577–587) Time of delivery in low-risk pregnancies is an important consideration for clinicians. They must balance the risks associated with early delivery, including respiratory and other neonatal complications, versus those associated with later delivery, including stillbirth, meconium aspiration syndrome and maternal complications.