Developing and pre-testing a decision board to facilitate informed choice about delivery approach in uncomplicated pregnancy

The rate of deliveries by Caesarean section (CS) is increasing internationally [1–4]. Many factors have influenced this trend including changing clinical indications and maternal characteristics [5, 6], as well as physician and maternal preferences. Physician preference for caesarean delivery has varied by country (UK 15-17% [7], Australia and New Zealand 11% [8], Ireland 7% [9]) and is generally attributed to fear of perineal trauma [8]. Maternal request for CS has also been widely documented [10–14]. According to a recent estimate between 4% and 11% of Caesarean deliveries worldwide are performed following maternal request in the absence of medical indication [15].

Concern about the rising rate of CS is based predominantly on an increase in maternal mortality and morbidity compared to vaginal delivery (VD), consequences for subsequent pregnancies and deliveries, neonatal respiratory morbidity [15], and cost implications [16]. In a recent study of over 100,000 pregnant women in eight Latin American countries [17] caesarean delivery was associated with a significantly increased risk of maternal morbidity for both intrapartum (OR 2.0, confidence interval 1.6 to 2.5) and elective cases (OR 2.3, confidence interval 1.7 to 3.1). Much of the on-going discussion has therefore focused on elective and 'on-demand' surgeries [18–21] and debates have been heated [22, 23]; this is an emotive subject that touches on professional integrity, consumerism of patients, and ethics [24–26]. To address these concerns the UK National Institute for Clinical Excellence (NICE) released a detailed guideline which states that maternal request is not, on its own, an indication for CS [27]. The guideline emphasizes that clinicians should explore reasons for the request and discuss associated benefits and risks. A statement released by the Society of Obstetricians and Gynaecologists of Canada (SOGC) similarly confirms that the society 'does not promote Caesarean sections on demand,' but reiterates SOGC's commitment to patient choice [28]. The American College of Obstetricians and Gynecologists has recommended that while the principle of patient autonomy supports the right to undergo elective CS, the physician must ensure that each patient is fully informed regarding the risks and benefits of delivery options [29].

Despite a consistent international emphasis on informed decision-making, there remains a particular lack of clarity about how information related to delivery approach in uncomplicated pregnancies should be provided. This is concerning given the scarcity of medical literature about the relative risks and benefits of CS versus VD, and it is particularly unclear how to inform women about risks that are small [30]. The decision board may be an efficacious way to address this gap. Decision boards are decisional aids that provide a standardized base of written and graphic evidence-based information about treatment choices and associated risks/benefits [31]. The decision board is administered by care providers during consultations to support and guide rather than substitute for health-related counselling, thereby helping to ensure that both patient and care provider are informed about the broad range of factors that impact health-related decisions.

A number of trials have evaluated decision boards in varied clinical settings [32]. In their systematic review of 17 randomized trials, O'Connor and colleagues reported that decision aids were associated with higher knowledge scores and increased patient participation in decision-making, but had a variable effect on the decisions that were made [32]. In a study evaluating a decision board for women deciding between mastectomy and lumpectomy 91% of clinicians also reported that the tool had been a helpful part of the consultation [33].

The purpose of this article is to describe the development and pre-testing of a decision board that can be used in practice to guide discussion and promote informed and shared decision-making between health care providers and women who have the opportunity to choose between delivery options (VD or elective CS) in uncomplicated pregnancy. Before administering the decision board to pregnant women, we wished to determine whether women of child-bearing age who might face the decision in the future found the tool useful and informative, how the decision board would impact their decision-making, and what additional information they would want to receive before making a 'real' decision about method of delivery. Our overall aim was not to enter into the debate about the availability of on-demand CS but rather to develop a vehicle that communicates information in as unbiased a way as possible, and in a way that women can understand and use in discussion with their health care provider.

Letters inviting participation in the study were mailed to 90 women from two family practices. Forty women, aged 20 to 40 (median age 27 years) agreed to participate. None had delivered a baby and none, to the best of their knowledge, were pregnant at the time of the study. Despite efforts to ensure varied educational levels most women (n = 36, 90%) had completed post-secondary education.

Our study is the first that we are aware of to develop and examine a decision aid to facilitate decision-making between elective CS and VD in women with uncomplicated pregnancies. Women of childbearing age who were not pregnant and had never given birth responded positively. Most thought that they had learned something new and that risk-related information would impact their decision-making regarding delivery approach. Their decisions changed little after using the decision board but women were more confident in their choices, stating that the process confirmed and added to the information they had pre-intervention. The enhanced confidence was most evident in two women who were initially undecided but chose VD after reviewing the decision board, citing risk factors associated with CS.

Our findings are similar to those reported in trials that have evaluated decision aids in a variety of clinical settings: decision aids improved knowledge and decreased decisional conflict [32]. Decision aids for women choosing delivery approach following a previous CS have been evaluated in randomised trials [36, 37]. An Australian trial examined a decision aid booklet, which included risks and benefits of each approach [36]. Women randomised to receive the booklet at 28 weeks gestation (n = 115) demonstrated a significant increase in focused knowledge scores (p < 0.001) and a significant decrease in decisional conflict (p < 0.05) compared with women in the control group (n = 112). Although preference for trial of labour was similar between groups at 36 weeks, women in the intervention group were less likely to be "unsure" of their decision [36]. The second, larger UK trial [37] compared usual practice with two forms of decision aid: (1) a computer-based information program relating possible health outcomes and probabilities associated with VD, elective CS and emergency CS, and (2) a separate program that asked women to value possible health outcomes using a visual analogue scale, but omitted probabilities. Women in both intervention groups had higher knowledge scores and lower decisional conflict than those in usual care, and although there was no statistical difference in mode of delivery between groups, researchers postulated that the higher rate of vaginal birth in the group that attached values to the outcomes was clinically important. Another trial evaluating a decision aid (consisting of a booklet plus an audio-CD and worksheet) about management options for women with breech presentation at term also found that the decision aid improved knowledge and reduced decisional conflict [38], although there was no statistically significant difference between groups in the proportion of women undergoing external cephalic version.

Our study contributes to this expanding body of knowledge by addressing choice of delivery approach in women who have not had a prior CS and who have no known complications, that is, those women who may seek a CS 'on demand'. Participants reported a significant decrease in decisional conflict and a significant increase in the perceived effectiveness of their decisions after reviewing the decision board (p = 0.001). The interpretation of these findings, however, merits caution. Our sample size was small (n = 40) and most women were highly educated. Participants were representative of an academic GP's practice (DL) but may not be representative of the larger population of women in our city between the ages of 18 and 40. Moreover, since on-demand CS is not currently a universal option we wished to avoid misleading pregnant women by raising the expectation that CS would be available to all of them. We therefore chose to pre-test the decision board with non-pregnant women to obtain feedback related to the tool's comprehensiveness and feasibility. To enhance the transferability of our findings we recruited participants of childbearing age. These women were excellent critics of the decision aid, were vocal in their praise and criticism of the decision board, and voiced numerous beliefs/perceptions that have been previously reported by pregnant women [38–40].

Despite potential criticism about the simplistic descriptions we used of the interventions and limited number of possible risks associated with the interventions, our findings support the usefulness and feasibility of the decision board approach and constitute an integral step in our overall aim to develop a vehicle for communicating risk-based information related to delivery approach (VD, elective CS). Women appreciated the systematic presentation of information and complementary ways of formatting the data. They provided equally important feedback, however, about information that they thought had been omitted from the decision board, including risks to the baby, realistic detail about the actual experience of undergoing VD and CS, the associated pain and pain management options, the risk of perineal tears and/or episiotomy with VD, and the recovery period following both VD and CS. This feedback will be critical to future revisions of our decision board, and women's suggestions will be incorporated to reflect the most current evidence as we work to create and evaluate a more comprehensive tool [41]. Additional detail will more accurately reflect what is meant by bladder injury and infection, while pain control options associated with VD and CS will be more comprehensively described. Attention will be also paid to the descriptions of delivery approaches in Figure 1, which several women believed were overly simplified, and to the textual descriptions of VD and CS which may have biased the decision-making process [42]. Perhaps most importantly, potential sequelae of choices that are made will be added to guide discussion with health care providers, and to ensure that women understand that attempted VD may result in significant medical interventions (such as labour induction, epidural), instrumental delivery (forceps or vacuum) or emergency CS. Current research suggests that the risk of emergency CS is less than 5% for multiparous women in spontaneous labour at term but may be as high as 35% for primiparous women who undergo induced labour at 42 weeks gestation [43]. Augmentations to the decision board will promote understanding that the health and safety of mother and baby will ultimately dictate delivery approach.

In the next phase of this research program it will be important to evaluate the revised decision board with care providers and women who are pregnant. Pregnant women are likely more motivated to seek focussed information and may therefore differ in their views of what is relevant, useful, and important. Repeated series designs should inform the impact of the decision board at varied intervals throughout pregnancy, while focus group interviews will provide a means to help ensure that content is both comprehensive and relevant given the varied circumstances and needs of pregnant women. As decision aids are meant to support medical consultation, the decision board will need to be tested in clinical practice to incorporate the two-way discussion that is critical to shared patient/health care provider decision-making [30].

As debate continues about the ethics of offering on-demand CS, the question for health care providers remains how to best counsel patients who request elective CS. At the time we were developing the decision board, the evidence presented in the NICE guidelines [27] represented the most rigorously and independently evaluated risks of CS and VD. Clearly, however, there is much work still to be done including the need for well-designed studies comparing short and long-term outcomes of VD and elective CS [22]. As a dynamic tool reflecting the current state of evidence and the needs of stakeholders (health care professionals and their patients), a decision board may best be described as a work in progress.

While clearly a controversial issue, the choice to undergo elective CS is being exercised to varying degrees by women in Canada and internationally [15, 17, 44–47]. This study represents a preliminary but integral step to help ensure women considering delivery approaches in uncomplicated pregnancies are fully informed. Our decision board affords a standardized means to organize data that non-pregnant women found useful and future iterations based upon our findings may guide discussions between patients and health care providers.