Objective
The objective of the study was to investigate the relationship between physician coping skills, need for cognition, tolerance of ambiguity, and anxiety and their patients’ mode of delivery.
Study Design
Ninety-four obstetricians were surveyed using 5 standardized psychometric scales: Reflective Coping, Proactive Coping, Multiple Stimulus Types Ambiguity Tolerance-II, Need for Cognition (measures learner motivation and engagement in cognitive efforts), and State-Trait Anxiety Inventory. Psychometric test scores were analyzed by quartile. Data regarding route of delivery were collected on 3488 nulliparous, term, cephalic, nonanomalous singleton deliveries performed by participating physicians. χ 2 tests and random-effects logistic regression analyses were used to examine the relationship between the obstetrician cognitive traits and type of delivery.
Results
Sixty-one percent of the women were delivered by spontaneous vaginal delivery, 15.5% by operative vaginal delivery, and 23.5% by cesarean delivery. Random-effects multivariable logistic regression adjusting for patient characteristics demonstrated that physicians with the most reflective coping (ie, highest quartile) were significantly less likely (adjusted odds ratio, 0.70; 95% confidence interval, 0.50–0.98) to perform operative vaginal delivery. However, lower anxiety and higher ambiguity tolerance were associated with an increased risk of chorioamnionitis and postpartum hemorrhage, respectively. There were no identified differences in adverse neonatal outcomes by physician cognitive or affective traits.
Conclusion
There is a decreased risk of operative vaginal delivery for patients delivered by providers with better adaptive decision-making traits. Other cognitive and affective traits were associated with a greater chance of chorioamnionitis and hemorrhage. Further work is required to elucidate whether training in these cognitive and affective traits can alter obstetric outcomes.
Provider cognitive and affective traits are increasingly recognized as important areas of investigation in patient safety and medical education. Cognitive traits, such as physicians’ learner motivation, cognitive biases, analytic skills, and cognitive efficiency, likely affect clinical decisions. Affective traits, on the other hand, include an individual’s stable predisposition toward types of emotional responses. Emerging evidence suggests that provider factors do influence obstetrical outcomes. Examples include the provider’s influence on a patient’s decision to undergo a trial of labor after cesarean delivery and the decreased likelihood of obstetric interventions when a provider is working within a night float system.
However, despite growing knowledge that cognitive and affective traits may influence clinical decisions, we know little about the relationship between cognitive processes of obstetricians and actual clinical outcomes. Using hypothetical scenarios, Chiossi et al found no relationship between provider affective traits, such as depression and anxiety symptoms, and expedited delivery decisions in the setting of nonreassuring fetal status.
In a small study, Dunphy et al assessed cognitive and affective traits among obstetricians at a single center in Canada. Providers underwent testing with 6 scales of cognition, metacognition, and affect. Physicians with greater self-efficacy, better coping, and lower trait anxiety were more likely to supervise spontaneous vaginal deliveries and less likely to perform midpelvic operative vaginal deliveries among nulliparous women. Their findings suggest that in clinically unpredictable situations such as intrapartum care, physician cognitive and affective characteristics may be associated with obstetrical decision making. However, this study included only 12 physicians and its findings have not been replicated.
To our knowledge, no studies have investigated the relationship between obstetrician cognitive and affective traits and perinatal outcomes in a large American cohort. Thus, we designed this study to investigate physician cognitive and affective traits in a large, urban women’s hospital that performs approximately 12,000 deliveries per year. The primary goal was to investigate the relationship between physician cognitive and affective traits and their patients’ mode of delivery. We hypothesized that in this population, better coping skills and less physician anxiety would be associated with more spontaneous vaginal deliveries.
Materials and Methods
This is an observational study examining the relationship between a physician’s cognitive and affective traits and their patient’s obstetric outcomes. Providers of obstetric care at a single academic institution were surveyed using 5 standardized psychometric measures. Their patient outcomes were then retrospectively reviewed to identify differences in obstetrical outcomes based on provider cognitive characteristics. Institutional review board approval from Northwestern University was obtained.
Eligible participants included all attending obstetricians currently practicing at the institution of study. General obstetricians and maternal-fetal medicine specialists were included. Resident physicians and midwives were ineligible because attending obstetricians are primarily responsible for the mode of delivery decisions. These obstetricians were invited to complete a written survey that included demographic characteristics and five established, validated scales. These 5 scales were: Reflective Coping (RC), Proactive Coping (PC), Multiple Stimulus Types Ambiguity Tolerance-II (MSTAT-II), Need for Cognition (NFC) and the trait component of the State-Trait Anxiety Inventory (STAI).
The RC, PC, MSTAT-II, and NFC largely measure cognitive traits, whereas the STAI is a measure of an affective trait. It is conceptually plausible that the specific traits measured by these scales may be associated with physician behavior. Also, the prior work of Dunphy et al demonstrated several relationships between these scales and obstetrical outcomes in their population; 1 goal herein was to expand upon their prior work by applying the scales used in that study to the larger population of patients and providers at our institution. Thus, these scales were selected on the basis of their high construct validity, the theoretical basis for relationship to medical decision making, and use in prior work. The details of each instrument can be found in Table 1 .
| Instrument | Construct | Number of items |
|---|---|---|
| RC scale of the Proactive Coping Inventory |
| 11 |
| PC scale of the Proactive Coping Inventory |
| 14 |
| MSTAT-II |
| 13 |
| NFC |
| 18 |
| STAI, trait component |
| 20 |
Patients eligible for study inclusion were nulliparas with term, cephalic, singleton gestations who were cared for by a physician who completed the survey. The population was limited to nulliparas because prior data suggest the relationship between physician traits and mode of delivery was limited to the nulliparous population. Women with fetuses with major anomalies or who had a fetal demise in utero were excluded. Women who were otherwise ineligible for vaginal delivery, such as those with placenta previa or prior uterine surgery, were additionally excluded.
The institution cesarean rate for the eligible population is approximately 22.5%. Based on prior literature, we estimated that the highest quartile of cognitive traits would be associated with a 20% reduction in cesarean deliveries. Thus, power analysis yielded a sample size of 3400 deliveries for 80% power to detect a difference in a cesarean rate from 22.5% to 18.0%. All deliveries by participating physicians from January 2012 to June 2013 were reviewed for eligibility; of these, charts for the 3488 eligible participants were abstracted.
The primary outcome in a univariable analysis was mode of delivery (cesarean delivery, operative vaginal delivery, or spontaneous vaginal delivery). Operative vaginal delivery included forceps- and vacuum-assisted vaginal deliveries.
Secondary outcomes included the frequency of maternal chorioamnionitis (defined as a temperature >100.3°F without an identified etiology other than intrauterine infection), postpartum hemorrhage (defined as estimated blood loss >500 mL for a vaginal delivery and >1000 mL for a cesarean delivery), major genital tract lacerations (defined as a third- or fourth-degree laceration), and episiotomies. Neonatal outcomes included 5-minute Apgar score less than 4, cord umbilical artery pH less than 7.0, neonatal sepsis, neonatal seizures, and neonatal hypoxic ischemic encephalopathy, and admission to the neonatal intensive care unit.
We described patient characteristics, stratified by mode of delivery, using χ 2 and analysis of variance tests. The psychometric scales were scored using established scoring techniques, and scores were categorized by quartile. For the RC, PC, MSTAT-II, and NFC, the fourth quartile (highest scores) represented the most advantageous coping/cognitive traits and was used as the referent. For the STAI, lower scores indicate the least trait anxiety, and thus, the lowest quartile scores were used as the referent group and here are referred to as the fourth quartile. Delivery mode was investigated by cognitive and affective quartile and compared using χ 2 tests.
Based on findings in the χ 2 tests, which revealed a difference specifically in the frequency of operative vaginal delivery, the delivery mode was then dichotomized into operative vaginal delivery vs all other delivery methods (spontaneous vaginal delivery and cesarean delivery). Univariable logistic regression and hierarchical random-effects multivariable logistic regression analyses were utilized to examine relationships between physician cognitive and affective traits and risk of operative vaginal delivery. The regressions adjusted for potential confounders, including patient characteristics identified to be significantly associated with mode of delivery (age, ethnicity, body mass index, insurance status, gestational age, and induction of labor) and included the delivering physicians as random-effects terms that accounted for the effect of nonindependence because of clustering by physician. Lastly, we investigated the secondary outcomes by quartile of physician cognitive and affective traits. Statistical analyses were undertaken using STATA version 11 (StataCorp, College Station, TX). All analyses were 2 tailed and P < .05 was used to define statistical significance.
Results
Ninety-four of the eligible 115 obstetricians (82%) signed written, informed consent and completed the survey. During the period of the study, 3488 of their patients met inclusion criteria. The patients’ mean age was 31.2 years, the mean body mass index was 29.7 kg/m 2 , and the mean gestational age at delivery was 39.5 weeks. The population was largely white (67.1%), and most (90.7%) were privately insured. Sixty-one percent delivered by spontaneous vaginal delivery, 15.5% by operative vaginal delivery, and 23.5% by cesarean section. Patient characteristics differed significantly based on delivery mode ( Table 2 ).
| Characteristic | Spontaneous vaginal delivery | Operative vaginal delivery | Cesarean delivery | P value |
|---|---|---|---|---|
| Age, mean (SD) | 30.67 (4.7) | 31.83 (4.5) | 32.26 (5.0) | < .01 |
| Body mass index, mean (SD) | 28.99 (4.8) | 29.12 (4.8) | 32.12 (5.9) | < .01 |
| Ethnicity | .002 | |||
| White | 1450 (61.9%) | 377 (16.1%) | 514 (22.0%) | |
| African American | 200 (62.5%) | 33 (10.3%) | 87 (27.2%) | |
| Hispanic | 209 (49.2%) | 47 (13.3%) | 97 (27.5%) | |
| Asian | 246 (57.8%) | 78 (18.3%) | 102 (23.9%) | |
| Other/unknown | 22 (45.8%) | 7 (14.6%) | 19 (39.6%) | |
| Clinic status | .009 | |||
| Private | 1903 (60.2%) | 505 (16%) | 754 (23.9%) | |
| Clinic | 224 (68.7%) | 37 (11.4%) | 65 (19.9%) | |
| Gestational age, wks (SD) | 39.47 (1.0) | 39.60 (1.1) | 39.84 (1.1) | < .01 |
| Induction of labor | < .01 | |||
| No induction | 1760 (64.3%) | 437 (16%) | 542 (19.8%) | |
| Induction of labor | 367 (49.0%) | 105 (14%) | 277 (37%) |
Obstetrician cognitive and affective traits were found to be significantly associated with operative vaginal delivery ( Table 3 ). Physicians with high reflective coping were less likely to deliver patients by operative vaginal delivery (12.5% vs 16.5%; P = .015). Similarly, physicians with a better tolerance of ambiguity were less likely to deliver patients by operative vaginal delivery (11.8% vs 16.4%; P = .006). There were no differences in delivery method with respect to obstetricians’ proactive coping, need for cognition, or trait anxiety.
| Characteristic | Spontaneous vaginal delivery, n (%) | Operative vaginal delivery, n (%) | Cesarean delivery, n (%) | P value |
|---|---|---|---|---|
| Reflective coping a | ||||
| Q4 b | 520 (62.1) | 105 (12.5) | 213 (25.4) | .015 |
| Q1-3 c | 1607 (60.6) | 437 (16.5) | 606 (22.9) | |
| Proactive coping d | ||||
| Q4 | 504 (62.9) | 109 (13.6) | 188 (23.5) | .21 |
| Q1-3 | 1623 (60.4) | 433 (16.1) | 631 (23.5) | |
| Tolerance of ambiguity e | ||||
| Q4 | 437 (65.3) | 79 (11.8) | 153 (22.9) | .006 |
| Q1-3 | 1690 (59.9) | 463 (16.4) | 666 (23.6) | |
| Need for cognition f | ||||
| Q4 | 373 (62.7) | 91 (15.3) | 131 (22.0) | .60 |
| Q1-3 | 1754 (60.6) | 451 (15.6) | 688 (23.8) | |
| Trait anxiety g | ||||
| Q4 | 624 (62.6) | 155 (15.6) | 218 (21.9) | 0.34 |
| Q1-3 | 1503 (60.3) | 387 (15.5) | 601 (24.1) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
