Physician-scientists in obstetrics and gynecology: predictors of success in obtaining independent research funding.

BACKGROUND: Obstetrics and gynecology departments receive the smallest amount of National Institutes of Health research funding and have significantly lower application success rates compared to pediatric, internal medicine, and surgery departments. The development of mentored early career development training grants (K awards) has been one strategy implemented by the National Institutes of Health to help aspiring physician-scientists establish independent research careers. OBJECTIVE: The purpose of this study is to describe the cohort of obstetrics and gynecology physician-scientists who were K08, K12, and K23 recipients from 1988 through 2015 and to identify predictors of success in obtaining independent federal funding, as defined by acquisition of an R01, R21, R34, U01, U54, P01, or P50 award. We hypothesized that sex, subspecialty, type of K award, and dual MD/PhD would impact success rates. STUDY DESIGN: K08, K12, and K23 recipients from 1988 through 2015 were identified from the National Institutes of Health Research Portfolio Online Reporting Tools, the office of the National Institutes of Health Freedom of Information Act, and the website of the Reproductive Scientist Development Program. Data were stratified by sex, educational degree, subspecialty, and type of K award. Data were analyzed using the Pearson χ2 and Fisher exact tests. The Kaplan-Meier estimator was used to determine rates of conversion to independent funding over time. RESULTS: A total of 388 K recipients were identified. Women accounted for 66% of K awards while men accounted for 34%. Among K recipients, 82% were MDs, while 18% were MD/PhDs. K12 awards accounted for 82% of all K awards, while K08 and K23 awards accounted for 10% and 8%, respectively. Subspecialists in maternal-fetal medicine and reproductive endocrinology and infertility received the highest proportion of K awards, followed by generalists and gynecologic oncologists. Altogether, the 3 subspecialty groups accounted for 68% of all K awards. R01 awards made up the bulk of independent funding. Among recipients who received their first K award between 1988 and 2009, 63 of 288 (22%) were successful at obtaining an R01. Rates of R21 (n = 22), U01 (n = 15), U54 (n = 12), P01 (n = 5), R34 (n = 1), and P50 (n = 1) acquisition ranged from 0.35-7.6%. In all, 118 K scholars (41%) were successful at achieving independent funding of any type compared to 1219 of 7535 (16.2%) obstetrics and gynecology non-K scholars. K08 recipients received the largest proportion of R01 awards compared to K12 and K23 recipients (32% vs 20%; P = .12), while 21% of K12 recipients and 17% of K23 recipients achieved an R01. There were no differences in the rates of independent funding success among K12 programs. K23 recipients were more likely to obtain an R21 (22% vs 6%, P = .008) compared to K12 and K08 recipients. The mean time to R01 acquisition was 6.8 years, while the mean time to independent funding of any type was 6.4 years. There were no significant differences in independent funding success rates by sex, educational degree, or subspecialty, although generalists received the highest proportion of R01 awards (29%). CONCLUSION: Mentored early career development K programs enable aspiring obstetrics and gynecology physician-scientists to achieve higher rates of National Institutes of Health-based independent research funding compared to non-K recipients.

Prepregnancy body mass index and risk of preterm birth: association heterogeneity by preterm subgroups.

BACKGROUND: To evaluate the association between prepregnancy body mass index (BMI) is associated with early vs. late and medically-induced vs. spontaneous preterm birth (PTB) subtypes. METHODS: Using data from the Boston Birth Cohort, we examined associations of prepregnancy BMI with 189 early (<34 completed weeks) and 277 late (34-36 completed weeks) medically-induced PTBs and 320 early and 610 late spontaneous PTBs vs. 3281 term births (37-44 weeks) in multinomial regression. To assess for mediation by important pregnancy complications, we performed sequential models with and without hypertensive disorders of pregnancy, chorioamnionitis, and gestational diabetes. RESULTS: Prevalence of prepregnancy obesity (BMI ≥ 30.0 kg/m2) was 28% among mothers with medically-induced PTBs, 18% among mothers with spontaneous PTBs, and 18% among mothers with term births (p = <0.001). After adjustment for demographic and known risk factors for PTB, prepregnancy obesity was associated with higher odds of both early [OR 1.78 (1.19, 2.66)] and late [OR 1.49 (1.09, 2.04)] medically-induced PTB. These effect estimates were attenuated with inclusion of hypertensive disorders of pregnancy and gestational diabetes. For spontaneous deliveries, prepregnancy obesity was associated with decreased odds of PTB (0.76 [0.58, 0.98]) and underweight was nearly associated with increased odds of PTB (1.46 [0.99, 2.16]). CONCLUSION: Prepregnancy obesity is associated with higher risk of medically-induced, but not spontaneous PTB. Hypertensive disorders of pregnancy and gestational diabetes appear to partially explain the association between prepregnancy obesity and early and late medically-induced PTB.

The combined association of psychosocial stress and chronic hypertension with preeclampsia.

OBJECTIVE: This study aims to evaluate perceived lifetime stress, perceived stress during pregnancy, chronic hypertension, and their joint association with preeclampsia risk. STUDY DESIGN: This study includes 4314 women who delivered a singleton live birth at the Boston Medical Center from October 1998 through February 2008. Chronic hypertension was defined as hypertension diagnosed before pregnancy. Information regarding lifetime stress and perceived stress during pregnancy was collected by questionnaire. Preeclampsia was diagnosed by clinical criteria. RESULTS: Lifetime stress (odds ratio [OR], 2.1; 95% confidence interval [CI], 1.6-2.9), perceived stress during pregnancy (OR, 1.7; 95% CI, 1.3-2.2), and chronic hypertension (OR, 10.4; 95% CI, 7.5-14.4) were each associated with an increased risk of preeclampsia. Compared to normotensive pregnancy with low lifetime stress, both normotensive pregnancy with high lifetime stress (OR, 2.1; 95% CI, 1.6-2.9) and pregnancy with chronic hypertension and low lifetime stress (OR, 10.2; 95% CI, 7.0-14.9) showed an increased risk of preeclampsia, while pregnancy with high lifetime stress and chronic hypertension yielded the highest risk of preeclampsia (OR, 21.3; 95% CI, 10.2-44.3). The joint association of perceived stress during pregnancy and chronic hypertension with preeclampsia was very similar to that of the joint association of lifetime stress and chronic hypertension with preeclampsia. CONCLUSION: This finding indicates that high psychosocial stress and chronic hypertension can act in combination to increase the risk of preeclampsia up to 20-fold. This finding underscores the importance of efforts to prevent, screen, and manage chronic hypertension, along with those to reduce psychosocial stress, particularly among women with chronic hypertension.

Patient safety series: obstetric safety improvement and its reflection in reserved claims.

In reviewing outcomes that are associated with the implementation of a series of labor and delivery patient safety efforts from 2004-2009, we requested data on the number of related professional liability claims that were reserved by our insurance companies that are established with the specific objective of financing risks that emanate from their parent group or groups. While we restructured the manner in which we give care, required training modules, and provided simulations to our providers, our legal risk continued to be monitored independently and in parallel. Retrospective review of the number of cases for which money was held in reserve for claims demonstrated a 20% decrease per year, which was adjusted for delivery volume, over this time period. We believe that the improved care that resulted from our safety projects has led to this decreased legal risk.

Prepregnancy body size, gestational weight gain, and risk of preterm birth in African-American women.

BACKGROUND: Studies of the association of prepregnancy body mass index (BMI) and preterm birth have been inconclusive, and no studies have examined the effect of central adiposity on risk. There is also uncertainty about optimal gestational weight gain among Black women. METHODS: Using self-reported prospective data from the Black Women's Health Study, we investigated the relation of preterm birth to prepregnancy BMI (kg/m), waist circumference, and gestational weight gain among 7840 singletons born to black women, ages 21-44, during 1995-2003. We compared mothers of 1114 infants born 3 or more weeks early (597 spontaneous preterm births and 517 medically-indicated preterm births) with mothers of 6726 term infants. We used generalized estimating equation models to derive multivariable odds ratios (ORs) and 95% confidence intervals (CIs). RESULTS: Relative to normal weight women (BMI: 18.5-24.9), underweight women (BMI: <18.5) were at increased risk of both preterm birth subtypes; obese women (BMI > or = 30.0) were at increased risk of medically-indicated preterm birth and very early spontaneous preterm birth (<32 weeks' gestation). Waist circumference, a measure of central adiposity, was positively associated with medically-indicated preterm birth. Among obese women, average gestational weight gain in the second and third trimesters of <0.4 or >0.6 lbs/wk was associated with an increased risk of both preterm birth subtypes. CONCLUSIONS: Our data suggest that underweight increases risk of both preterm birth subtypes, while obesity increases risk of medically-indicated preterm birth and only a subgroup of spontaneous preterm births (<32 weeks' gestation). Central adiposity was an independent risk factor for medically-indicated preterm birth only. Among obese women, gestational weight gain within the range recommended by the 2009 Institute of Medicine report (0.4-0.6 lbs/week in the second and third trimesters) was associated with the lowest risk of preterm birth.

Maternal outcomes associated with planned primary cesarean births compared with planned vaginal births.

OBJECTIVE: To compare the outcomes and costs associated with primary cesarean births with no labor (planned cesareans) to vaginal and cesarean births with labor (planned vaginal). METHODS: Analysis was based on a Massachusetts data system linking 470,857 birth certificates, fetal death records, and birth-related hospital discharge records from 1998 and 2003. We examined a subset of 244,088 mothers with no prior cesarean and no documented prenatal risk. We then divided mothers into two groups: those with no labor and a primary cesarean (planned primary cesarean deliveries-3,334 women) and those with labor and either a vaginal birth or a cesarean delivery (planned vaginal-240,754 women). We compared maternal rehospitalization rates and analyzed costs and length of stay. RESULTS: Rehospitalizations in the first 30 days after giving birth were more likely in planned cesarean (19.2 in 1,000) when compared with planned vaginal births (7.5 in 1,000). After controlling for age, parity, and race or ethnicity, mothers with a planned primary cesarean were 2.3 (95% confidence interval [CI] 1.74-2.9) times more likely to require a rehospitalization in the first 30 days postpartum. The leading causes of rehospitalization after a planned cesarean were wound complications (6.6 in 1,000) (P<.001) and infection (3.3 in 1,000). The average initial hospital cost of a planned primary cesarean of US dollars 4,372 (95% C.I. US dollars 4,293-4,451) was 76% higher than the average for planned vaginal births of US dollars 2,487 (95% C.I. US dollars 2,481-2,493), and length of stay was 77% longer (4.3 days to 2.4 days). CONCLUSION: Clinicians should be aware of the increased risk for maternal rehospitalization after cesarean deliveries to low-risk mothers when counseling women about their choices. LEVEL OF EVIDENCE: II.

Genome-wide approach identifies a novel gene-maternal pre-pregnancy BMI interaction on preterm birth.

Preterm birth (PTB) contributes significantly to infant mortality and morbidity with lifelong impact. Few robust genetic factors of PTB have been identified. Such 'missing heritability' may be partly due to gene × environment interactions (G × E), which is largely unexplored. Here we conduct genome-wide G × E analyses of PTB in 1,733 African-American women (698 mothers of PTB; 1,035 of term birth) from the Boston Birth Cohort. We show that maternal COL24A1 variants have a significant genome-wide interaction with maternal pre-pregnancy overweight/obesity on PTB risk, with rs11161721 (PG × E=1.8 × 10-8; empirical PG × E=1.2 × 10-8) as the top hit. This interaction is replicated in African-American mothers (PG × E=0.01) from an independent cohort and in meta-analysis (PG × E=3.6 × 10-9), but is not replicated in Caucasians. In adipose tissue, rs11161721 is significantly associated with altered COL24A1 expression. Our findings may provide new insight into the aetiology of PTB and improve our ability to predict and prevent PTB.

Impact of labor induction, gestational age, and maternal age on cesarean delivery rates.

OBJECTIVE: To quantify the impact of labor induction and maternal age on cesarean delivery rates in nulliparous and multiparous women between 36 and 42 weeks' gestation. METHODS: We performed a retrospective cohort study on 14,409 women delivering at two teaching hospitals in metropolitan Boston during 1998 and 1999. Women who had contraindications to labor, including a prior cesarean delivery, were excluded. The risks for cesarean delivery by induction status, gestational age by completed week between 36 and 42 weeks, maternal age <35, 35-39, and >/=40 years, and stratified by parity, were calculated by logistic regression. RESULTS: In nulliparas, labor induction was associated with an increase in cesarean delivery from 13.7% to 24.7% (adjusted odds ratio [OR] 1.70; 95% confidence interval [CI] 1.48, 1.95]). In multiparas, induction was associated with an increase from 2.4% to 4.5% (OR 1.49; 95% CI 1.10, 2.00). Other variables that placed a nulliparous woman at increased risk for cesarean delivery included maternal age of at least 35 years and gestational ages over 40 weeks. For multiparas, only maternal age 40 years or older and gestational age of 41 weeks were associated with an increase in cesarean deliveries. CONCLUSION: Induction of labor, older maternal age, and gestational age over 40 weeks each independently increase the risk for cesarean delivery in both nulliparous and multiparous women. Although the relative risk from induction is similar in nulliparas and multiparas, the absolute magnitude of the increase is much greater in nulliparas (11% versus 2.1%).

Persistent fetal occiput posterior position: obstetric outcomes.

OBJECTIVE: To evaluate the obstetric outcomes associated with persistent occiput posterior position of the fetal head in term laboring patients. METHODS: We performed a cohort study of 6434 consecutive, term, vertex, laboring nulliparous and multiparous patients, comparing those who delivered infants in the occiput posterior position with those who delivered in the occiput anterior position. We examined maternal demographics, labor and delivery characteristics, and maternal and neonatal outcomes. RESULTS: The prevalence of persistent occiput posterior position at delivery was 5.5% overall, 7.2% in nulliparas, and 4.0% in multiparas (P <.001). Persistent occiput posterior position was associated with shorter maternal stature and prior cesarean delivery. During labor and delivery, the occiput posterior position was associated with prolonged first and second stages of labor, oxytocin augmentation, use of epidural analgesia, chorioamnionitis, assisted vaginal delivery, third and fourth degree perineal lacerations, cesarean delivery, excessive blood loss, and postpartum infection. Newborns had lower 1-minute Apgar scores, but showed no differences in 5-minute Apgar scores, gestational age, or birth weight. CONCLUSION: Persistent occiput posterior position is associated with a higher rate of complications during labor and delivery. In our population, the chances that a laboring woman with persistent occiput posterior position will have a spontaneous vaginal delivery are only 26% for nulliparas and 57% for multiparas.

Should older women have antepartum testing to prevent unexplained stillbirth?

OBJECTIVE: Older women are at an increased risk for unexplained stillbirth late in pregnancy. The purpose of this study was to compare 3 strategies for the prevention of unexplained fetal death in women aged 35 years and older. We compared usual care (no antepartum testing or induction before 41 weeks), weekly testing at 37 weeks with induction after a positive test, and no testing with induction at 41 weeks. METHOD: We used a Markov model to quantify the risks and benefits of each strategy in terms of the number of antepartum tests, inductions, and additional cesarean deliveries per fetal death averted. Probability data used in the model were derived from obstetrical databases and the literature. RESULTS: Without a strategy of antepartum surveillance between 37 and 41 weeks, women aged 35 years and older would experience 5.2 unexplained fetal deaths per 1,000 pregnancies. For nulliparous women 35 and older, weekly antepartum testing initiated at 37 weeks would avert 3.9 fetal deaths per 1,000 pregnancies but would require 863 antepartum tests, 71 inductions, and 14 additional cesarean deliveries per fetal death averted. A strategy of no testing but induction at 41 weeks would avert 0.9 fetal deaths per 1,000 pregnancies and require 469 inductions and 219 additional cesareans per fetal death averted. CONCLUSION: A strategy of antepartum testing in older women would reduce the number of unexplained stillbirths at term and would result in fewer inductions and cesareans per fetal death averted than a strategy of no antepartum testing but induction at 41 weeks.