Advertisement

Time from consent to cesarean delivery during labor

      Objective

      The objective of the study was to describe the time from consent to incision (consent time) when informed consent (IC) for cesarean delivery (CD) is obtained during labor and identify risk factors for especially short consent times.

      Study Design

      The study was a retrospective chart review of 90 cases of CD during labor. Medians and interquartile ranges for times were reported. Multivariable linear and logistic regressions were used to adjust for confounders. A Kaplan-Meier survival analysis was performed to compare consent time among women undergoing CD for fetal heart rate (FHR) indications with other women.

      Results

      The median consent time was 48 minutes (interquartile range, 25–72); 28.9% of patients delivered less than 30 minutes after consent. When adjusted for potential confounders, the odds of delivering less than 30 minutes after consent were 4.7 times higher (95% confidence interval, 1.4–15.2, P = .01) among women who underwent CD for FHR indications than for women who underwent CD for other indications.

      Conclusion

      This study demonstrates that when IC for CD is obtained during labor, consent time is brief, particularly among women undergoing CD for FHR indications. Although time is not necessarily a proxy for quality, an especially short consent time is likely to reduce the quality of the conversation and limit a patient's opportunity to understand and recall information. Additional research regarding the optimal approach to IC for CD in labor is needed; however, this study suggests that the practice of obtaining IC at the time the decision for CD is reached is unlikely to provide optimal time for a complete IC discussion.

      Key words

      For Editors' Commentary, see Contents
      See related editorial, page 166
      See Journal Club, page 274
      More than 1.2 million cesarean deliveries (CDs) are performed annually in the United States, with approximately half of those occurring after a woman is in labor.
      Centers for Disease Control and Prevention, National Center for Health Statistics.

      Podulka J, Stranges E, Steiner C. Hospitalizations related to childbirth, 2008. Statistical Brief no. 110. Available at: http://www.hcup-us.ahrq.gov/reports/statbriefs/sb110.jsp. Accessed May 10, 2012.

      Despite the high frequency of CDs in labor and the urgent or emergent nature of CD when it occurs during labor, in many settings, the informed consent (IC) process does not occur until the health care provider reaches the decision to proceed with CD. Chervenak and McCullough
      • Chervenak F.
      • McCullough L.
      An ethical framework for the informed consent process for trial of labor after cesarean delivery.
      described that the expected components of the IC process include providing patients the opportunity to absorb, retain, recall, and understand information and express a value-based preference. Accomplishing this process in the midst of labor is problematic because women are likely to be distracted by pain, fatigue, and/or fear. Additionally, because these decisions are time sensitive, women are often asked to consent in a short period of time, which may have an impact on the quality of the IC process.
      Although the practice of obtaining IC for CD during labor is common, the process has not previously been described in the literature.
      The data presented are a retrospective chart review of a representative sample of women who underwent CD during labor to describe the amount of time that women had from consent for CD to incision (consent time) and determine whether there were particular patients who were at high risk for especially short consent times. Data on the process of IC in labor may identify potential ways to improve this process and allow for a more authentic consent process.

      Materials and Methods

      After institutional review board approval, we performed a retrospective analysis of a representative sample of women who underwent CD in labor at an academic medical center with delivery volume of approximately 4000 deliveries annually and a CD rate commensurate with national statistics (approximately 25-30%). At the time of the study, the typical practice in this hospital was to obtain IC for CDs performed during labor when the decision for CD was reached, which is a common practice in many hospitals across the country.
      The IC for scheduled CDs was obtained in the outpatient setting during a routine preoperative visit. For antepartum patients, IC for CD was typically obtained at the time of admission. The medical records department was asked to provide a list of patients who underwent CD during the calendar year 2004, which was selected because it was the most recent year that neither author was in practice at this hospital. Women who underwent scheduled CDs (elective repeat CD, placental abnormalities, malpresentation, etc) were excluded. Only women undergoing urgent or emergent CD during labor were included. From the remaining patients, the first 100 patients to deliver during the calendar year were selected for chart review. Given the descriptive nature of this study, no sample-size calculation was indicated. The decision to review 100 charts was based on our estimation that a review of approximately 15% of unscheduled CDs would provide sufficient data to reliably estimate average consent time. Patients with missing or incomplete charts (missing time of CD consent) were excluded.
      Each patient's medical record was reviewed by a single author (K.S.), and data on maternal demographics, labor characteristics (use of oxytocin or fetal scalp electrode [FSE]), indication for CD (active phase arrest, fetal heart rate [FHR], arrest of descent or other), and timing of events were abstracted. Time data were abstracted and confirmed from multiple sources, including nursing and physician labor notes, nursing flow sheets, and operating room paperwork. The time documented on the consent form itself was the preferred source for time data; however, when this was unavailable, a timed nurse or doctor note stating when consent was obtained was also accepted. There were no cases of a significant discrepancy between reported timing of consent. Incision time was recorded on operating room flow sheets by nursing.
      Data were analyzed using Stata 12 (StataCorp, College Station, TX). Outliers in the greater than 97% percentile and less than 3% percentile of consent time were Wisorized to the 97th and third percentiles, respectively. Consent time and time from admission to incision were nonnormally distributed; therefore, log transformations of these variables (which were normally distributed) or nonparametric methods were used where appropriate. For portions of the analysis, consent time was divided into dichotomous categories (less than 15 minutes vs 15 minutes or longer or less than 30 minutes vs 30 minutes or longer).
      For bivariate analysis, the χ2 test was used to compare categorical predictors and dichotomous outcomes, and the Mann-Whitney rank-sum test was used to compare categorical predictors and continuous nonnormally distributed outcomes. Multivariable linear and logistic regression models were used to evaluate associations between key predictors and linear and dichotomous outcomes, respectively. Several variables (maternal age, race, parity, and obesity) were included in these models for face validity, although they were not significantly associated with consent time in bivariate analyses. Finally, a Kaplan-Meier survival analysis was performed to compare consent time among women undergoing CD for fetal indications vs other indications for CD.

      Results

      Of the 100 cases obtained from medical records, 9 charts were missing or incomplete. One patient was consented for cesarean section as a matter of routine for an antepartum admission prior to labor and was therefore excluded from analysis. The remaining 90 patients were included in the analysis (Figure 1). The most common indication for admission was labor induction (44%), and the most common indication for cesarean section was active-phase arrest (40%) (Table 1). Four women had a history of prior CD and 3 women had a twin pregnancy. Data for 4 women whose consent times were equal to or after incision time were analyzed as being consented 1 minute before delivery and 3 women who had more than 240 minute consent times were analyzed as having a 240 minute consent time.
      Figure thumbnail gr1
      Figure 1Study flow diagram
      Initial charts were identified by medical records, and 90 charts were included in the final analysis.
      Salmeen. Cesarean consent during labor. Am J Obstet Gynecol 2013.
      Table 1Characteristics of study population
      DemographicsValue
      Mean age, y (range)29 (18–45)
      Mean BMI, kg/m2 (range)32.6 (20.4–56.6)
       Normal weight (BMI <25), n (%)7 (7.8)
       Overweight (BMI 25.0-29.9), n (%)31 (34.4)
       Obese (BMI ≥30), n (%)52 (57.8)
      Mean gestational age, wks (range)39.7 (32–42.1)
      White race, n (%)63 (70)
      Nulliparous, n (%)70 (77.8)
      Indication for admission, n (%)
       Induction of labor40 (44)
       Spontaneous labor27 (30)
       Spontaneous rupture of membranes22 (24.4)
       Other1 (1.1)
      Indication for cesarean section, n (%)
       Active-phase arrest36 (40)
       Fetal heart rate32 (35.6)
       Arrest of descent20 (22.2)
       Other2 (2.2)
      Labor characteristics, n (%)
       Oxytocin used67 (74.4)
       Fetal scalp electrode used55 (61.1)
      BMI, body mass index.
      Salmeen. Cesarean consent during labor. Am J Obstet Gynecol 2013.
      Overall, the median consent time was 48 minutes, with an interquartile range of 25–72 minutes (Table 2). In a bivariate analysis, when assessed as a continuous outcome, consent time increased linearly with time from admission to incision (P = .002). Additionally, the use of oxytocin was significantly associated with increased consent time, with a median consent time of 53 minutes when oxytocin was used vs 28 minutes when no oxytocin was used (P = .01).
      Table 2Key outcomes
      VariableValue
      Median time from consent to incision (min), median time (IQR), P value
       Overall48 (25–72)
       FHR indication26 (12–51).001
       Non-FHR indication54 (40–79)
       Oxytocin used53 (36–79).01
       No oxytocin used28 (14–54)
       FSE used40 (22–64).03
       No FSE used54 (40–79)
      Adjusted change in consent time, % (95% CI)
      Adjusted for age, obesity, nonwhite race, parity, duration of admission to incision, and oxytocin use.
       FHR indication for CD56.1 ↓(8.3 ↓–104 ↓)
       Oxytocin used10.7 ↑(44.3.0 ↓–65.8 ↑)
       FSE used2.1 ↓(65.9 ↓–24.1 ↑)
      Consent time <30 minutes, %, P value
       Overall28.9
       FHR indication56.3.001
       Non-FHR indication13.8
       Oxytocin used20.9.004
       No oxytocin used52.2
       FSE used40.0.004
       No FSE used11.4
      Adjusted OR for <30 min consent time (95% CI)
      Adjusted for age, obesity, nonwhite race, parity, duration of admission to incision, and oxytocin use.
       FHR indication for CD4.7(1.4–15.2)
       Oxytocin used0.75(0.15–3.7)
       FSE used8.01(1.8–36.3)
      CD, cesarean delivery; CI, confidence interval; FHR, fetal heart rate; FSE, fetal scalp electrode; IQR, interquartile range; OR, odds ratio.
      Salmeen. Cesarean consent during labor. Am J Obstet Gynecol 2013.
      a Adjusted for age, obesity, nonwhite race, parity, duration of admission to incision, and oxytocin use.
      FSE use and FHR indication for CD were both associated with decreased consent time. The median consent times among labors with and without an FSE were 40 and 53 minutes, respectively (P = .03). The median consent time for women who had a FHR indication for CD was 26 minutes as compared with a median consent time of 54 minutes among women with non-FHR indications for CD (P < .001). Notably, among women who had a CD for FHR indications, 84% had an FSE as compared with 48% of women who had a CD for other reasons (P = .001). The median time from FSE placement to consent was 305 minutes. There were no significant associations between consent time and maternal age, nonwhite race, parity, gestational age, obesity, prior CD, indication for admission, or daytime delivery.
      In a multivariable model controlling for maternal age, nonwhite race, parity, obesity, and oxytocin use, only the associations between FHR indication for CD (P = .02) and duration from admission to incision (P = .002) remained significantly associated with consent time. As compared with non-FHR indications for CD, women who underwent CD for FHR indications had a 56.1% reduction in consent time (95% confidence interval [CI], 8.3-104%) when adjusted for demographics (which were included for face validity of the model, although they did not demonstrate significant associations in bivariate analysis), oxytocin use, and admission to incision time.
      When consent time was assessed in a dichotomous fashion, 28.9% had a consent time of less than 30 minutes and 10% of patients had a consent time of less than 15 minutes. Only women who had FHR indications for CD were delivered in less than 15 minutes. Among women who underwent CD for FHR indications, 56.3% of women had less than 30 minutes consent time as compared with 13.8% of women who underwent CD for non-FHR indications (P < .001). The odds of having a consent time of less than 30 minutes were 4.7 times higher (95% CI, 1.4-15.2, P = .01) among women who had a FHR indication for CD when adjusted for duration of admission to incision, age, nonwhite race, parity, obesity and the use of oxytocin. Using Kaplan-Meyer survival analysis, the proportion of women who remain undelivered at any given time is significantly lower among women who had a CD for FHR indications as compared with other women (P = .04) (Figure 2).
      Figure thumbnail gr2
      Figure 2Kaplan-Meier survival estimate for time to incision
      Survival curves for time from consent to incision for fetal vs nonfetal indications for cesarean delivery. A small percentage of patients in each category were not delivered after 120 minutes.
      Salmeen. Cesarean consent during labor. Am J Obstet Gynecol 2013.

      Comment

      This study demonstrates that when informed consent for CD in labor is obtained at the time the decision for CD is made, the length of time between the informed consent discussion and the procedure is less than 50 minutes for most patients. This brief interval from consent to incision may be insufficient to allow for an optimal IC process. Although the length of an IC conversation is not necessarily correlated with the quality of the conversation, and our study did not assess the quality of the IC processes that took place, it is unlikely that a high-quality IC process can occur in less than 50 minutes. The length of time required for adequate informed consent is likely to vary considerably by patient, and no published guidelines exist to suggest an optimal consent time. However, for comparison, IC discussions for scheduled CD in this institution took place at least 24 hours prior to planned delivery, were scheduled in 30 minute appointment slots, and IC was readdressed on the day of the procedure.
      In 2 published studies of general surgery populations, Fink et al
      • Fink A.S.
      • Prochazka A.V.
      • Henderson W.G.
      • et al.
      Predictors of comprehension during surgical informed consent.
      demonstrated that short consent conversations were associated with less comprehension and that the repeat-back technique is associated with improved comprehension but increased time in the consent process.
      • Fink A.S.
      • Prochazka A.V.
      • Henderson W.G.
      • et al.
      Enhancement of surgical informed consent by addition of repeat back: a multicenter, randomized controlled clinical trial.
      These studies and clinical experience support the conclusion that time is likely to be at least 1 key factor in the quality of an IC discussion.
      There are several studies that assessed the quality of the IC process for CD. Touqmatchi et al
      • Touqmatchi D.
      • Boret T.
      • Nicopoullous J.
      The quality of operative consenting against RCOG advice as standard.
      performed a chart audit evaluating documentation of risks of surgery, including 40 women undergoing CD. They demonstrated that major risks such as injury to the ureters and impact on future pregnancies were not documented in as many as 50% of charts. They concluded that “the time constraints of having to fill in the forms on the day of surgery” were a potential cause of the deficiency.
      • Touqmatchi D.
      • Boret T.
      • Nicopoullous J.
      The quality of operative consenting against RCOG advice as standard.
      In another study of 554 women undergoing CD, Odumosu et al
      • Odumosu M.
      • Pathak S.
      • Barnet-Lamb E.
      • Akin-Deko O.
      • Joshi V.
      • Selo-Ojeme D.
      Understanding and recollection of the risks associated with cesarean delivery during the consent process.
      investigated patients' recall of the risks of CD 24 hours after delivery. This study included women who underwent scheduled CD and were consented 1 week prior to surgery and women who underwent emergent CD and were consented during labor. Approximately 25% of patients had no recollection of the risks of CD, and women who underwent CD in labor comprised 78% of the no-recall group. The authors concluded that the odds of having no recall of the risks of CD were 4-fold higher among women who had a CD during labor as compared with a planned CD.
      • Odumosu M.
      • Pathak S.
      • Barnet-Lamb E.
      • Akin-Deko O.
      • Joshi V.
      • Selo-Ojeme D.
      Understanding and recollection of the risks associated with cesarean delivery during the consent process.
      In a third study, Akkad et al
      • Akkad A.
      • Jackson C.
      • Kenyon S.
      • Dixon-Woods M.
      • Taub N.
      • Habiba M.
      Patients' perceptions of written consent: questionnaire study.
      assessed patient understanding of the IC process among 732 patients undergoing obstetrical or gynecological procedures, including 242 women who had obstetrical procedures. This study noted that 25% of women undergoing emergency surgery stated they did not have sufficient time to think about the consent form, and 22% stated that they felt too ill to fully read the consent form.
      • Akkad A.
      • Jackson C.
      • Kenyon S.
      • Dixon-Woods M.
      • Taub N.
      • Habiba M.
      Patients' perceptions of written consent: questionnaire study.
      Each of these studies suggests that the IC process may be suboptimal when performed in an urgent or emergent setting. The data from our study suggest that limited time is available for IC when it is performed in the context of labor, and this may be an important contributing factor to the suboptimal consent process described in the studies mentioned in previous text.
      Several observations from our study warrant specific discussion. First, we did not identify increased consent time among women at especially high risk for CD including obese women or those with a prior CD. These may be patients among whom the potential for CD and the associated risks are sufficiently high that an IC discussion may have been warranted well before labor. It is also important to note that consent time was particularly short when the indication for CD was for fetal heart rate. There is obviously an important impetus to move quickly when there is a fetal heart rate indication for CD, and the fact that these women were delivered more quickly than other women is reassuring. However, the rapidity of the consent process in this context may have particularly important consequences, especially because CDs performed rapidly for fetal heart rate indications are more likely to be associated with poorer neonatal outcomes and higher rates of surgical complications.
      • Nielsen T.F.
      • Hokegard K.H.
      Cesarean section and intraoperative surgical complications.
      The opportunity to thoroughly discuss these potential risks may be even more important in this population than others; yet it is the population that had the least time for discussion.
      Another interesting observation is the association between FSE use and decreased consent time. Although it is not unexpected that FSE use was associated with decreased consent time, given the high correlation between FSE use and CD for FHR indications, it is notable that the median time from FSE placement to incision (305 minutes) was relatively long compared with the median consent times. This finding suggests that although there was concern for FHR hours prior to CD in some circumstances, providers did not initiate IC discussions at that time.
      Finally, we noted that increased time from admission to incision and oxytocin use were associated with increased consent time. Although it cannot be definitively concluded from the data, these observations suggest that when labor was longer with more opportunities for IC discussion, that providers may have been more likely to take additional time with the consent process.
      This study has several important limitations common to a retrospective chart review. Most importantly, we are limited in our assumptions. That is, although we want to draw conclusions about the quality of the IC process, the data available in this study do not provide information about women's perceptions or recollections of the consent process. It is possible that a high-quality IC discussion could be completed in a very short period of time. Additionally, our data do not reflect patient status at the time of consent (pain or anxiety level, fatigue, etc), which may also be important predictors of the quality of the IC discussion. Furthermore, we assumed that the women included in the study were a representative sample of women undergoing CD in labor; however, it is possible that there were systematic factors, such as providers on-call, that may have affected the timing of consent. Finally, we assumed that women had not previously had an informed consent discussion about CD prior to the time of signing their surgical consent. Although this is typically true based on our experience, it may not have been the case for all women included in this study.
      The importance of high-quality IC cannot be underestimated. Informed consent is the hallmark of the ethical principle of respect for autonomy. As described by Chervenak and McCullough,
      • Chervenak F.
      • McCullough L.
      An ethical framework for the informed consent process for trial of labor after cesarean delivery.
      to adhere to this ethical principle, physicians must allow patients the opportunity to absorb information about their care, understand this information, evaluate medically reasonable alternatives, and express a value-based preference. Furthermore, a suboptimal IC conversation may have important implications for patient's perceptions of the birth experience and satisfaction.
      Studies have demonstrated that emergency operative delivery has been associated with posttraumatic stress disorder, reduced patient satisfaction, and an increased risk for a negative birth experience.
      • Olde E.
      • van der Hart O.
      • Kleber R.
      • van Son M.
      Posttraumatic stress following childbirth: a review.
      In turn, these experiences are associated with avoidance of future child-bearing, request for future CD, depressive symptoms, and difficulty with infant bonding.
      • Waldenstrom U.
      • Hildingsson I.
      • Rubertsson C.
      • Radestad I.
      A negative birth experience: prevalence and risk factors in a national sample.
      Although there is no obstetrical literature to demonstrate that a high-quality IC discussion will prevent these complications, one study from the field of gastroenterology demonstrated that a thorough IC discussion was associated with reduced anxiety about procedures.
      • Betti S.
      • Sironi A.
      • Saino G.
      • Ricci C.
      • Bonavina L.
      Effect of the informed consent process on anxiety and comprehension of patients undergoing esophageal and gastrointestinal surgery.
      Finally, given the frequency of litigation in obstetrics and the idea that a considerable amount of litigation is associated with inappropriate patient expectations, high-quality IC discussions may help clarify expectations and reduce litigation, which has been cited as a potential factor in the rising rate of CD in the United States.
      • Zwecker P.
      • Azoulay L.
      • Abenhaim H.A.
      Effect of fear of litigation on obstetric care: a nationwide analysis on obstetric practice.
      As described by the American Congress of Obstetricians and Gynecologists, “a thorough informed consent process and strong documentation practices are the cornerstones of risk management. Effective informed consent is not just a form—it's a process to ensure that a patient makes the health care decision that is right for her.”

      American Congress of Obstetricians and Gynecologists. You asked, we answered: develop risk management tools during your residency. Available at: http://www.acog.org/About_ACOG/ACOG_Departments/Professional_Liability/YOU_ASKED__WE_ANSWERED__Develop_risk_management_tools_during_your_residency. Accessed March 15, 2013.

      Despite important limitations, we believe that this study reliably demonstrates that IC when obtained during labor is likely to be rushed and as a consequence may be suboptimal. This is likely to be particularly important for women with language, social, medical, and/or emotional barriers that may require additional time to achieve adequate informed consent, although our data do not explore this question.
      The optimal approach to informed consent for CD performed during labor requires further investigation. One approach may be standardized, universal informed consent discussions during routine prenatal care. Another approach may be to identify opportunities during labor, such as at the time of FSE placement or initiation of oxytocin therapy, when a high-quality IC discussion may be initiated, reducing the potential need for a rushed discussion. Yet another approach may be to provide access to Internet-based or written material at the beginning of pregnancy for a patient to explore at her own pace.
      Future research about the process of IC is needed to determine the ideal strategy to maximize the quality of the IC process while minimizing patient anxiety. Until such research is performed, we believe that the data presented here and elsewhere suggest that the practice of obtaining informed consent for CD at the time the decision for CD is reached is not likely to be optimal and should be replaced with alternative approaches.

      References

      1. Centers for Disease Control and Prevention, National Center for Health Statistics.
        National vital statistics reports. 2010; (Report no 59. Available at:) (Accessed May 10, 2012)
      2. Podulka J, Stranges E, Steiner C. Hospitalizations related to childbirth, 2008. Statistical Brief no. 110. Available at: http://www.hcup-us.ahrq.gov/reports/statbriefs/sb110.jsp. Accessed May 10, 2012.

        • Chervenak F.
        • McCullough L.
        An ethical framework for the informed consent process for trial of labor after cesarean delivery.
        Clin Perinatol. 2011; 38: 227-231
        • Fink A.S.
        • Prochazka A.V.
        • Henderson W.G.
        • et al.
        Predictors of comprehension during surgical informed consent.
        J Am Coll Surg. 2010; 210: 919-926
        • Fink A.S.
        • Prochazka A.V.
        • Henderson W.G.
        • et al.
        Enhancement of surgical informed consent by addition of repeat back: a multicenter, randomized controlled clinical trial.
        Ann Surg. 2010; 252: 27-36
        • Touqmatchi D.
        • Boret T.
        • Nicopoullous J.
        The quality of operative consenting against RCOG advice as standard.
        J Obstet Gynaecol. 2010; 30: 159-165
        • Odumosu M.
        • Pathak S.
        • Barnet-Lamb E.
        • Akin-Deko O.
        • Joshi V.
        • Selo-Ojeme D.
        Understanding and recollection of the risks associated with cesarean delivery during the consent process.
        Int J Gynaecol Obstet. 2012; 118: 153-155
        • Akkad A.
        • Jackson C.
        • Kenyon S.
        • Dixon-Woods M.
        • Taub N.
        • Habiba M.
        Patients' perceptions of written consent: questionnaire study.
        BMJ. 2006; 333: 528-529
        • Nielsen T.F.
        • Hokegard K.H.
        Cesarean section and intraoperative surgical complications.
        Acta Obstet Gynecol Scand. 1984; 63: 103-108
        • Olde E.
        • van der Hart O.
        • Kleber R.
        • van Son M.
        Posttraumatic stress following childbirth: a review.
        Clin Psychol Rev. 2006; 26: 1-16
        • Waldenstrom U.
        • Hildingsson I.
        • Rubertsson C.
        • Radestad I.
        A negative birth experience: prevalence and risk factors in a national sample.
        Birth. 2004; 31: 17-22
        • Betti S.
        • Sironi A.
        • Saino G.
        • Ricci C.
        • Bonavina L.
        Effect of the informed consent process on anxiety and comprehension of patients undergoing esophageal and gastrointestinal surgery.
        J Gastrointest Surg. 2011; 15: 922-927
        • Zwecker P.
        • Azoulay L.
        • Abenhaim H.A.
        Effect of fear of litigation on obstetric care: a nationwide analysis on obstetric practice.
        Am J Perinatol. 2011; 28: 277-284
      3. American Congress of Obstetricians and Gynecologists. You asked, we answered: develop risk management tools during your residency. Available at: http://www.acog.org/About_ACOG/ACOG_Departments/Professional_Liability/YOU_ASKED__WE_ANSWERED__Develop_risk_management_tools_during_your_residency. Accessed March 15, 2013.