Risk of Instrumental Delivery in Maternal Obesity: Estimates With Measures of Effect Size

Background: Obesity, defined as a body mass index ≥30 kg/m2, is epidemic in the United States and is associated with increased risks of adverse events. Studies have examined the role of maternal obesity on the incidence of instrumental vaginal delivery, but the results are divided. However, these analyses used frequentist tests that risk false discovery. The purpose of this retrospective study was to quantify the association of maternal obesity to the need for instrumental delivery with measures of effect size. Measures of effect size allow assessment of the impact of clinical risk factors on outcomes of interest. Methods: All parturients aged ≥18 years in active labor at our facility from January 2018 to May 2019 were entered into this study. Patient demographics, previously reported comorbidities, and obstetric parameters were collected and analyzed to determine the clinical impact of maternal obesity on the incidence of instrumental delivery. One effect size measure, risk differences, was used to quantify the clinical effect of maternal obesity on the need for instrumental delivery. A generalized linear model was used to standardize the measures of effect size of previously reported comorbidities, including maternal obesity, and to determine their association with the need for instrumental delivery. Results: The incidences of chronic and gestational hypertension, preeclampsia, chronic diabetes, and reactive airway disease were higher in parturients with maternal obesity. Risk differences due to maternal obesity were observed in parturients presenting with shoulder dystocia but not in those who underwent oxytocin induction or in nulliparous parturients. Following regression analysis, maternal obesity did not clinically impact the need for instrumental delivery. Conclusion: These findings suggest that maternal obesity did not have a clinical impact on the need for instrumental delivery.


INTRODUCTION
6][17] The purpose of this observational study was to quantify the magnitude and precision of the association of maternal obesity to the need for instrumental delivery.

METHODS
Following institutional board approval, all parturients aged ࣙ18 years in active labor from January 2018 to May 2019 at Ochsner Health in New Orleans were entered into this study.Patient demographics, previously reported comorbidities, and obstetric parameters were collected to examine the role of maternal obesity on the mode of delivery (instrumental vaginal delivery vs normal spontaneous vaginal delivery) in parturients during active labor.

Statistics
Categorical variables are presented as counts and percentages with confidence intervals (CI), and differences between BMI groups (BMI ࣙ30 kg/m 2 vs BMI <30 kg/m 2 ) were assessed with chi-square tests.Continuous variables with skewed distributions are presented as medians with 25% to 75% interquartile ranges [IQR], and differences

Chai, M
between BMI groups were assessed with the Wilcoxon ranksum test.9][20][21][22][23] When the ranges of these CIs do not contain zero, a clinical impact is suggested.
A generalized linear model 24 was constructed to quantify the association of adjusted risk differences of previously reported comorbidities to the need for instrumental delivery.Clinical impacts of the previously reported comorbidities in this model are expressed as standardized risk differences with 95% confidence limits (CL). 24Frequentist tests are shown for comparison, with P values set for significance at <0.005 to minimize the risk of false discovery rates or of declaring associations significant by chance alone. 14,25,26he statistical program JMP Pro 17.2 (SAS Institute Inc) was used for this study.

RESULTS
Demographics, comorbidities, and measured obstetric parameters for 1,183 parturients undergoing active labor are shown in Table 1.When delivery BMI was partitioned into obese (BMI ࣙ30 kg/m 2 ) and nonobese (BMI <30 kg/m 2 ) groups, 592/1,183 or 50.0%(CI 47.2%-52.9%) of the parturients were observed to be obese.Risk differences were observed in the obese parturients with histories of chronic hypertension, gestational hypertension, preeclampsia, chronic diabetes, and reactive airway disease.Risk differences due to maternal obesity were observed in parturients presenting with shoulder dystocia, but not for those who underwent oxytocin induction or for nulliparous parturients.No risk differences were observed between the 2 BMI groups by the type of occiput presentation or in the mode of delivery.
Admitting diagnoses by BMI group are shown in Table 2. Risk differences due to obesity were only observed in parturients scheduled for labor induction or who presented with premature rupture of membranes.
Eighty-two cases of shoulder dystocia were observed in this study.The clinical impact of maternal obesity on the association of shoulder dystocia to the mode of delivery is shown in Table 3(A).Risk differences were calculated, and no clinical impact of obesity was observed in parturients receiving instrumental delivery.
Seventy-seven percent of patients (CI 75%-79%) received oxytocin, either with scheduled admission or initiated during labor.The clinical impact of maternal obesity on the association of oxytocin use to mode of delivery is shown in Table 3(B).No clinical risk differences in the need for instrumental delivery were observed in obese parturients compared to nonobese parturients.
Risk differences were calculated for the role of maternal obesity on the association of nulliparity to the mode of delivery, and the results of this association are shown in Table 3(C).No clinical risk differences of maternal obesity were observed with the association of nulliparity to the mode of delivery.
Risk differences were standardized for the previously reported comorbidities associated with maternal obesity, and the results of this generalized linear regression analysis are shown in Table 4. None of the comorbidities-including maternal obesity-was associated with the need for instrumental delivery.

Shoulder Dystocia and Maternal Obesity
Although a higher incidence of parturients presenting with shoulder dystocia was observed in the obese group, the incidence of instrumental delivery in shoulder dystocia cases was the same for both BMI groups, as were the incidences of instrumental delivery in those who underwent oxytocin induction and in nulliparous parturients.Although Scott-Pillai et al 2 and Robinson et al 11 observed higher associations of shoulder dystocia with increasing obesity, both groups reported a lower need for instrumental delivery in their obese parturients.Although Usha Kiran and colleagues 12 reported a very low incidence of shoulder dystocia, they observed no differences in the need for assisted vaginal delivery in their obese parturients.

Instrumental Delivery and Maternal Obesity
Studies have also examined the role of maternal obesity on the incidence of instrumental delivery, but these results were divided.Blomberg, 13 Cedergren, 6,7 Nohr et al, 8 Schrauwers and Dekker, 9 and Sydsjö et al 10 reported higher associations of instrumental delivery with maternal obesity.In contrast, Pettersen-Dahl et al, 1 Scott-Pillai et al, 2 Foo et al, 3 Roman et al, 4 Robinson et al, 11 and Sarkar et al 5 did not observe an increased association of instrumental delivery with maternal obesity.The reasons for these decreasing associations are not clear, but Pettersen-Dahl et al suggested that the obstetricians may have decided to perform early cesarean deliveries rather than proceed with instrumental delivery. 1Scott-Pillai et al also observed increased rates of cesarean delivery, whether elective or emergency, in their obese parturients with corresponding lower rates of normal or instrumental deliveries. 2 Finally, Roman et al observed an association of cesarean delivery with shoulder dystocia and a resultant lower need for instrumental delivery in their obese parturient population. 4

Admitting Diagnoses and Maternal Obesity
The primary admitting diagnoses in this study were similar to those reported in other studies, [1][2][3][4] except for a higher incidence of parturients admitted for scheduled oxytocin induction in this study.Pettersen-Dahl and colleagues reported admitting incidences for oxytocin induction of 47.3% for obese parturients and 17.1% for nonobese parturients. 1 Jensen and colleagues reported admitting oxytocin induction incidences of 12.4% in their nonobese group and 14.1% and 23.4% in their 2 obese groups. 27These incidences 1,27 are lower than our observations (Table 2), and the differences in admitting oxytocin induction incidences may reflect distinctions in practice patterns.However, the risk differences Notes: Data are presented as n (%) unless otherwise indicated.Risk differences due to maternal obesity are in bold.CI, 95% confidence interval.IQR, 25%-75% interquartile range.
analysis in Table 3(B) shows that oxytocin use was not associated with the need for instrumental delivery.

Previously Reported Comorbidities and Maternal Obesity
We performed generalized linear regression analysis on previously reported comorbidities, including maternal obesity, to standardize their interplay with the need for instrumental delivery.None of the comorbidities, including maternal obesity, was associated with the need for instrumental delivery.In the comparative studies, [1][2][3][4][5][6][7][8][9][10][11][12][13] odds ratios were frequently used as a measure of effect size, but odds ratios have the potential to overexpress relative risk.][17] Additionally, odds ratios cannot be calculated when no events are recorded in any comparative group.Finally, few clinicians are truly facile with this statistic; odds ratios do not account for the incidence of the outcome in question and do not translate well during patient discussions at the bedside, 28 whereas measures of effect size, such as risk differences, can be clearly understood by physicians, patients, and their families. 15,28

Limitations and Strengths
A limitation of retrospective studies is the completeness of medical record data. 29However, a strength of this study is the near 100% data collection (1.1% missing data) because of the development of electronic medical records.Another limitation of this study is the lack of data on early cesarean deliveries because we did not study decisions for this surgical option.
A strength of this study is the use of measures of effect size.1][22][23] Effect sizes provide insight into the numeric relationship between binary variables associated with an outcome of interest.This effect size is referred to as the risk difference.Incorporating  these measures can enhance the depth of analysis and interpretation in clinical settings.When medical researchers only use frequentist tests in investigation of their findings, statistically significant results may or may not have clinical importance, and moreover, there is a 29.6% chance that their findings are not reproducable. 30An additional measure to improve the value of statistical significance not due to chance is to reset the P values of frequentist statistics to the now recommended value of <0.005, 26 which reduces the potential rate of false discovery from 29.6% to <6.9%. 14nother limitation of this study is potential bias due to confounding.However, a robust statistical method was used to standardize the interplay of previously reported predictors against the outcome in question.Also, the magnitude and precision of a predictor of interest on the outcome can be realized.Generalized linear modeling provides a useful tool to standardize measures of effect size, as the standardization expresses information about the clinical magnitude and the precision of that magnitude of the association rather than the inferences based on frequentist testing. 31easures of effect size provide robust clinical evidence in support of clinical decisions. 18,19Whether the effect size is clinically relevant depends upon the importance of the risk factor and the risk differences with the CI ranges of the 2 outcome groups.Finally, standardized risk differences hold an advantage over odds ratios as they express information about the occurrence of the outcome in addition to information about the magnitude and precision of the association. 31

CONCLUSION
Although risk differences for chronic hypertension, gestational hypertension, preeclampsia, chronic diabetes, and reactive airway disease were increased in parturients with maternal obesity, these parameters, once standardized, did not play a clinical role on the need for instrumental delivery.Although a higher incidence of parturients presented with shoulder dystocia in the obese group, the incidence of instrumental delivery was not clinically different in the nonobese and obese groups, nor was any difference seen in those who underwent oxytocin induction or in nulliparous parturients.Maternal obesity did not play a role in the need for instrumental delivery.The use of measures of effect size provides clinical assessments of outcomes rather than inferences based on statistical significance values obtained from frequentist statistics.

Table 4 . Adjusted Risk Differences of Previously Reported Comorbidities and the Risk of Instrumental Delivery
26tes: Whole model chi-square=7.4,degrees of freedom=8, P=0.3922.Goodness of fit statistic: log-ratio chi-square=1182, P=0.4292.P values <0.005 are statistically significant.26BMI,body mass index; CL, confidence limit; Std Error, standard error of the risk difference.