Clinical decision making is an integral part of healthcare today. However, clinical decision making within the midwifery profession differs from many other areas in healthcare (Ockenden, 2020). Midwives support and advocate for primarily healthy women throughout a life-altering transition. They are upheld to a professional standard represented by the UK Nursing and Midwifery Council (NMC, 2018) code to provide safe, evidence-based, compassionate and individualised care in partnership with women, promoting choice and shared decision-making (Ménage, 2016; National Institute for Health and Care Excellence (NICE), 2019). Midwives are responsible for the care of the woman and her baby, more commonly known as the woman—baby dyad. The Association for Improvements in the Maternity Services (2012) believe this is crucial for the health of women and their families.
Clinical decision making is not static, and recommendations for best practice are continuously changing as innovative research and investigations provide up-to-date evidence to enable practitioners to give individualised care and minimise perinatal morbidity and mortality. Reports such as Ockenden (2022) provide thought-provoking case studies that enable midwives to reflect on their own practice cases, to learn from their mistakes and strengths, and change how they approach clinical decision making for individual clinical scenarios to improve outcomes for women and their babies. The critical reflection presented in this article reviews the clinical decision making process within a personal midwifery case-decision, focusing on the hypothetico-deductive and dual processing models involving a midwife, student midwife and woman, Lucy (pseudonym) (Elstein and Schwarz, 2002; Jefford et al, 2011; Kahneman, 2011).
The process of decision making has been studied within psychology, economics, law and healthcare (Dowding et al, 2012). Decision making, judgements and critical thinking are often thought to be interlinked. Critical thinking is said to be a prerequisite for ‘good’ judgements and decision making, where judgements are ‘an assessment between alternatives’ while a decision is ‘a choice between alternatives’ (Dowding and Thompson, 2003). Alternatively, ‘clinical decision making is a contextual, continuous and evolving process, where data [are] gathered, interpreted and evaluated in order to select an evidence-based choice of action’ (Tiffen et al, 2014).
The proposed definitions of clinical decision making are extrapolated from medical emergencies, diagnosis and critical events. These bear some relevance to midwifery practice but fail to fully encapsulate the process of midwifery clinical decision making, as they focus solely on the nursing and medicine professions (Thompson, 1999; Croskerry, 2002; Tiffen et al, 2014). Daemers et al's (2017) qualitative study demonstrates the complexity of clinical decision making in midwifery; by using a case decision, the present article aims to encourage midwives and other healthcare professionals to reflect on their own clinical decision making and the factors that influence it. Daemers et al (2017) assert that each midwife appears to have their own individual contributing factors to decision making, alongside generic factors that connect most midwives clinically. These include:
- The midwife's philosophy of childbirth
- Local clinical guidelines
- Perception of the importance of women-centred care and shared decision making
- Collaboration with the multidisciplinary team
- Attitudes towards physiology of childbirth
- Human factors.
It seems futile to define clinical decision making in midwifery because of the spectrum it encompasses. However, developing an understanding of one's own clinical decision making by encouraging self-reflection and analysis, and continuing to question the clinical decision making process, develops the skills required for clinical decision making in midwifery, improving the safety, competency and reasoning of the care that midwives provide (Cioffi, 1998).
What is decision making in midwifery?
Midwives often make clinical decisions with missing or ambiguous information, requiring skill and a degree of managed risk consideration (NHS, 2022) or trade off, as per Sherif et al's (1965) degrees of latitude and what is deemed acceptable. This must be done while ensuring that decisions are evidence-based and within the expectations of the code (Rew, 2000; Royal College of Obstetricians and Gynaecologists (RCOG), 2014; NMC, 2018). The quality of a midwife's decision making can directly affect the safety and quality of the care provided (Jefford, 2012) and, as highlighted by the Ockenden (2020) report, poor clinical decision making can consequently lead to increases in mortality and morbidity for both mother and baby.
Clinical governance enables midwives and other health professionals to be held accountable for their decisions. Therefore, these professionals need be able to provide rationale with evidence-based practice (Department of Health, 2010; NMC, 2018). A challenge that midwives face is the rapidly evolving role of the midwife and the prerequisite of decision making that comes with it (Mong-Chue, 2000; Bradfield et al, 2018). Midwifery, like most medical and health professions, is not an exact science, which has the potential to lead to mistakes linked to human factors (Mead and Sullivan, and be reported negatively or controversially in the media (Lay, 2022).
Having the ability to critically analyse the ‘four dimensions of critical thinking’, proposed by Mong-Chue (2000), enables midwives to improve care by seeking the views of women and colleagues, reflecting on one's own practice, critically appraising evidence before its application and by being mindful of the influence that the environment can have on thinking. Shared decision making is an integral part of the midwife's daily decision making process, encompassing the knowledge, experience, evidence and clinical expertise of professionals and the values, preferences, beliefs and experiences of women in partnership and collaboratively (Coulter and Collins, 2011; NICE, 2019). Fundamentally, it is key in woman-centred care that aims to empower women to feel more knowledgeable, better informed, risk aware and clearer about their values and beliefs when making decisions about their care provision (Barry and Edgeman-Levitan, 2012; Stacey et al, 2017).
Fetal growth surveillance evidence-base
The tripartite case-decision (Box 1) concerns fetal growth surveillance, which is of high priority during the antenatal period, not only to benefit the health and care of the woman—baby dyad, but to promote public health with the ever-growing population (Department of Health, 2010; RCOG, 2014). Current guidelines warrant the measurement of the symphysis-fundal height at every antenatal appointment past 24 weeks gestation for low-risk mothers (RCOG, 2014; O'Donnell et al, 2020; NICE, 2021). Screening for small for gestational age and intrauterine growth restriction using the symphysis-fundal hright method is highly debated in the literature (Belizán et al, 1978; Persson et al, 1986; Haram et al, 2006). NICE (2021) developed monitoring fetal growth guidance following a systematic review of 11 retrospective cohort studies, six prospective cohort studies, one nested case-control study and one population-based study. The committee concluded that ultrasound scans are not a sensitive screening tool to determine small for gestational age, but become more reliable closer to birth at term gestation and remain more sensitive than symphysis-fundal height measurements alone.
Box 1.Tripartite case decisionLucy (pseudonym) was a gravida 1 parity 0, low-risk, 34 week pregnant woman, who had consented to being case-loaded by a third-year student midwife and supervised by a registered midwife. The student midwife attended Lucy's home for a routine antenatal appointment, where Lucy reported feeling well in herself, with all findings within normal parameters. The symphysis-fundal height measurement was 34cm. As a result of the nature of a home visit as a student midwife, there was no access to the electronic records of Lucy's previous appointments at her home.The student midwife arranged the 36 week appointment with Lucy prior to returning to the clinic to document the appointment outcome. On reviewing the intergrowth chart with the 34 week measurement, it was clear there was reduced growth velocity. The student midwife and registered midwife discussed this, and the student midwife highlighted that, to be in line with trust guidelines, Lucy should be referred for an urgent growth ultrasound scan to confirm appropriate growth. The registered midwife agreed with the student midwife, and the student midwife telephoned Lucy, highlighting the recommendation for a growth ultrasound after a ‘symphysis-fundal height reading that displays a reduced growth velocity or static growth’ (O'Donnell et al, 2020). Lucy reviewed her options in light of the new information and followed the recommendation for the student midwife to send a referral for a growth ultrasound.
Despite symphysis-fundal height measurement having poor sensitivity, it is easy to perform with little resource implications and with no adverse effects during the procedure (NICE, 2021). Realistically, it would be resource intensive to offer all women an ultrasound scan and the committee concluded it is appropriate to offer the symphysis-fundal height measurement for low-risk women at each antenatal appointment post-24 weeks and plotted on a growth chart, as recommended in the saving babies’ lives bundle (version 2) (NICE, 2021).
The intergrowth-21st project aimed to produce international standard growth charts, not exclusively for fetal growth, that could be used to monitor and evaluate maternal and fetal wellbeing (Oxford University, 2022). Gardosi et al (2018) reviewed this ‘one size fits all’ approach to fetal growth monitoring, finding significant variation in populations and individuals, which makes the project's approach redundant. Since 2018, just under 80% of UK hospitals have implemented the growth assessment protocol derived from the perinatal institute (Gardosi et al, 2018). Part of the protocol is using GROW software that produces growth charts, which are customised for constitutional variation and optimised by excluding pathological factors; this improves identification of abnormal growth while reducing false positive diagnoses (Gardosi et al, 2018; Perinatal Institute, 2020). Gardosi et al (2018) found that after implementation of the protocol in England and Scotland, stillbirth rates dropped 19% and 20% respectively from 2009—2016. While the intergrowth-21st growth charts are deemed to be flawed by some (Gardosi et al, 2018), its ‘one size fits all’ ethos has led to the development of the GROW software that is now recommended in RCOG (2014) and NICE (2021) guidance for fetal growth monitoring.
GROW relies on consistent and accurate symphysis-fundal height measurements, performed by a midwife, which is subjective when assessing fetal growth. Lucy (Box 1) was recommended a growth ultrasound scan by the student midwife because of the reduced growth velocity displayed on her GROW chart once the measurement was plotted, as per NICE (2021). Lucy reviewed this information in collaboration with the student midwife and registered midwife, resulting in the tripartite decision to be referred for a growth ultrasound scan.
Decision-making theories
Table 1 describes the three decision-making theories that are commonly used to aid categorisation of the available decision-making models, as applied to the case decision (Dowding and Thompson, 2003; Standing, 2008).
Table 1. Decision-making theories
| Approach | Definition | Models | Application to case decision |
|---|---|---|---|
| Normative | Focus on how rational individuals make decisions with the aim of determining how decisions ought to be made in an ideal and optimal world. Decisions are supported by clear or probable evidence. |
|
N/A |
| Descriptive | How individuals make judgements and decisions in the real world, focusing on actual conditions, contexts, ecologies and environments in which they are made. They place no focus on whether the individual is logical or rational. |
|
Heuristics/intuition played a part as once the student midwife and midwife reviewed the GROW chart, both knew to recommend a growth ultrasound scan from cues gathered during assessment. The pattern of reduced growth velocity and decision to send for an ulstrasound is reoccurring in the antenatal period and so has become the ‘rule of thumb’ for most midwifery professionals. |
| Prescriptive | These theories set out to improve the judgements and decisions of individuals by thinking about how they make decisions. |
|
The intuition to recommend an ultrasound came from local trust and NICE guidelines on reduced growth velocity. The intergrowth chart could be perceived as a decision-making tool in itself, by providing information for assessing trends in an individual's growth compared with percentiles of ‘normal’ growth. |
Sources: Dowding and Thompson, 2003; Shaban, 2005; Standing, 2008
Dual-processing theory and the hypothetico-deductive model
The term ‘dual processing’ has been applied to many cognitive studies, such as learning, reasoning, conceptual thinking, decision making and social cognition. The fundamentals of the theory are that of two mental processes, corresponding to intuitive and deliberate thinking (Evans, 2007). These two processes, with respect to decision making, are commonly labelled system 1 (S1) and system 2 (S2) thinking (Evans, 2007; Kahneman and Klein, 2009) (Table 2).
Table 2. Cognitive processes
| Process | Explanation |
|---|---|
| S1 | S1 thinking is automatic, involuntary, pragmatic, unconscious and cognitively effortless |
| S2 | S2 thinking is slow, conscious, controlled, voluntary and cognitively effortful |
| Intuition | The common themes to define intuition are that it is not rational, it just happens and cannot be explained |
| Hypothetico-deductive model | A structured, rational and evidence-based process |
Sources: Shaban, 2005; Evans, 2007; Kahneman and Klein, 2009
S1 corresponds with intuitive/experimental approaches to decision making (Jefford et al, 2011; Smith, 2016). Intuition is an accredited process for clinical decision making, yet it poses a challenge to define for both the midwife and the pregnant woman. Despite a lack of agreement in the literature on a succinct definition of intuition, there are common themes throughout proposed definitions. Kahneman and Klein (2009) proposed different types of intuition, such as skilled intuition, defined by Simon (1992) as ‘the situation has provided a cue: this cue has given the expert access to information stored in memory, and the information provides the answer. Intuition is nothing more and nothing less than recognition’. In the case decision, the cue provided was the plotting of the symphysis-fundal height measurement and the memory it accessed for both the midwife and student midwife, being the last time this situation had arisen, prompted the immediate clinical decision that Lucy should be referred for a growth ultrasound as per local trust and NICE (2021) guidance (O'Donnell et al, 2020). This idea feeds into pattern recognition and intuitive cognition's role in S1, and challenges the hypothetico-deductive model. S1, intuition and pattern recognition do not follow the structured, rational, evidence-based process that the hypothetico-deductive model suggests clinical decision making uses (Elstein and Schwarz, 2002).
In Lucy's case, S2 and the hypothetico-deductive model were somewhat harmonious. The hypothetico-deductive model of decision making is the dominant approach in the medical profession because of its empirical and scientific basis. It is a method of deciding the best alternative from those available based on the rational and empirical data presented (Thompson and Dowding, 2009; Jefford et al, 2011).
As the nature of clinical decision making is multifaceted, it can make the conceptualisation of a given situation and its appropriate action challenging. Therefore, hypothesis generation is a psychological necessity to transform an unstructured problem into a structured one with a set of possible solutions. It provides an efficient and effective reasoning strategy for clinicians to follow in a logical format to support the documentation of such narratives (Elstein, 1994). The hypothetico-deductive model has been adapted from the ‘4 stages of reasoning’ (Table 3) first proposed by Elstein and colleagues in 1978 (Thompson and Dowding, 2009) and applied to Lucy's case decision.
Table 3. Four stages of reasoning
| Stage | Explanation | Application to Lucy's case |
|---|---|---|
| Cue acquisition | Gathering clinical information from history taking, reviewing previous medical documentation and discussing signs and symptoms with the patient. | The cues gained from Lucy's appointment were her gestation, presence or absence of fetal movements, emotional and physical wellbeing, the symphysis-fundal height measurement, previous symphysis-fundal height measurements, presence of the fetal heart, blood pressure, urinalysis and obstetric history. |
| Hypothesis generation | Generating potential explanations for the clinical picture presented by cues. The clinician may make an initial hypothesis, which is related to the data gathered and held in short-term memory. | Lucy presented with no confounding concerns about her pregnancy. She was experiencing fetal movements as normal and the clinical information gathered was within normal parameters, except for the trend of symphysis-fundal height measurement on the GROW chart. Initial hypothesis generation was a reduced growth velocity of the fetus based on plotting and interpretation of the intergrowth chart. By extension, a growth ultrasound scan was the recommendation under current local and NICE guidance. |
| Cue interpretation | Interpretating cues from data gathering and using that information to confirm, refute or class as not applicable to the initial hypothesis generated from the first two stages. | Data gathering was a growth ultrasound scan, which provided more information when interpretating if the fetus was growing in accordance with its natural trend. Based on this cue, the hypothesis of reduced growth velocity was re-evaluated and refuted. |
| Hypothesis evaluation | Weighing up pros and cons of each hypothesis with relation to previous stages, with the aim of choosing a favoured explanation by the majority of evidence. | The initial hypothesis is flawed, stemming from the inaccuracies of the symphysis-fundal height measurement. The growth ultrasound scan overpowered all other cues, and so the final hypothesis was that the fetus was developing normally. Consequently, no additional action was required. |
Adapted from: Thompson and Dowding (2009)
The conscious, slow, voluntary and cognitively demanding process of the hypothetico-deductive model shadows that of S2 thinking. As the outline in Table 3 demonstrates, this cognitive method can be extrapolated to Lucy's case, as well as S1. Kahneman (2011) indicates that occasionally there is overlap between S1 and S2 thinking. In Lucy's case, there was a clear overlap of the two mental processes. S1 thinking was used by both the midwife and student midwife when the information from the plotted intergrowth chart was assessed. This was based on their intuition, clinical experiences and recognition of reoccurring clinical scenarios. It was cognitively effortless and fast. However, to get this outcome, cue acquisition and interpretation had to first take place, following a more S2 approach (Jefford et al, 2011).
Lucy's approach to decision making encapsulates a multitude of stages, akin to the midwife. For many women, their background, knowledge and experience with making midwifery decisions may be limited, compared to that of the midwife and/or obstetrician caring for them. This can make midwifery decision making difficult and overwhelming for women. Murugesu et al (2021) begins to expose this by proposing a 5-stage model for women making shared decisions:
- Understanding pregnancy stages
- Understanding consequences, risks and benefits
- Identifying their own preferences and values
- Participating in decision-making conversations with a midwifery professional
- Making a decision.
These stages set out the theory of how women process decision making within midwifery. Many of these stages rely on appropriate information gathering from trustworthy sources (Stacey et al, 2017), which is a known aspect of the midwife's role (NMC, 2018). Lucy trusted the student midwife and midwife's professional opinion and recommendation, and they had explained the rationale for the referral, the guidance behind reduced growth velocity and the evidence as to why that guidance is in place. This enabled Lucy to feel in control of her pregnancy decisions by equipping her with the tools to conceptualise the situation, weigh up the risks and benefits and understand the evidence base before making a decision (Stacey et al, 2017; Murugesu et al, 2021).
Limitations of the dual processing and hypothetico-deductive model
Although the hypothetico-deductive model of clinical decision making has often been shown to be successful, it does have several cognitive biases. The most documented bias is known as anchoring (Harbison, 2001). This is when the decision maker continues to favour (anchor to) their initial hypothesis, despite a flow of contradictory evidence. The hypothetico-deductive model abandons a crucial aspect of midwifery care: shared decision making. It makes following the trajectory of woman-centred care and shared decision making difficult because of disregard for the values, beliefs and experiences of the people the clinical decisions are being made about (Barber, 2012). Furthermore, this model is deprived of the use of intuition and heuristics that other successful clinical decision-making models expose and relate to (Thompson and Dowding, 2009).
The notion that S1 thinking is reliant on mental operations, such as mental shortcuts (heuristics) and biases from associative memory/experience, opens S1 thinking up to errors in judgement, such as how with the hypothetico-deductive model, a decision-maker can anchor themselves to their initial judgement (Harbison, 2001). The midwife and student midwife anchored to their decision to refer Lucy for a growth ultrasound scan because of their intuition and memory of local trust and NICE guidance, which could be perceived as rote learning. In this scenario, anchoring to a hypothesis did not pose detriment to the decision. However, this mental bias can prevent good and clear clinical decision making in other situations, leading to errors and detriment to those being cared for (Kahneman, 2011). As a result of the nature of intuition, the decision maker relies heavily on their contextual knowledge, which is limited to their experiences and environments (Shaban, 2005). It can be concluded that S1 fails to advocate for decisions to be made from the evidence base and so logical justification is absent (Barber, 2012). This poses its own difficulty when making contemporaneous documentation of the decision-making process, an aspect of midwifery decision making that is highly important (NMC, 2018).
S2 thinking has its own limitations, requiring significant cognitive effort and so is not sustainable for long periods of time (Kahneman, 2011). This is prudent to the midwifery profession because of the long working hours and risk of burnout. The dual processing model focuses on an explanation as to how clinical decisions are made, but does not provide any mechanism for how clinical decision makers can transfer appropriately between intuitive and analytical cognition to reach a good clinical judgement (Jefford et al, 2011). To use intuition and analytical cognition in conjunction with one another relies on the professional having awareness of the situation and their own cognitive biases. This means the professional needs to have the ability to assess a situation with a helicopter perspective and consider how their experiences influence a given situation.
Conclusions
Using a clinical tripartite case-decision alongside a decision making model provided the opportunity to reflect on current practice and the knowledge base, and identify areas for further development as the student midwife entered their first post as a newly qualified midwife. The importance of the woman's voice in clinical decision making has strengthened the need for collaborative and shared approaches in care provision, incorporating all professionals involved in the woman's narrative. Having the opportunity to delve deeper into clinical decision making has enabled the student midwife to question the clinical decisions made intuitively, resulting in a greater awareness of unconscious biases when discussing decisions with women or colleagues. It is essential for midwives to continue to question their clinical decision-making process, and their inclusion of the women and her family in this process, to enhance skills, safety, competency and reasoning of the care they provide. BJM
Key points
- Midwives routinely make clinical decisions in partnership with women and their families.
- The quality of a midwife's decision making can directly affect the safety and quality of the care provided.
- Clinical decision making has often been shown to be successful, yet it does come with several cognitive biases.
- The midwife's role is to support women to feel in control of their pregnancy decisions, assisting with weighing up the risks and benefits and understanding the evidence base required for decision making.
- Midwives are held accountable for their decisions; therefore, they need to be able to provide a rationale using evidence-based practice.
CPD reflective questions
- How do you perceive the accuracy of symphysis-fundal height measurements compared to the evidence base, and how does this impact your clinical decision making?
- What human factors could affect how you make clinical decisions?
- How often do you use intuition/pattern recognition when making clinical decisions and how does this impact your confidence in the decisions made?
- What model of clinical decision making relates best to your practice and why?
- How often do women in your care have a strong voice in the decision-making process? Identify three cases where this is true and three cases where it is not, and compare the contexts to see what learning you can take from this.