Understanding delayed access to antenatal care: a qualitative study Rosalind Haddrill

The research question

The previous chapters have given an introduction to the subject of late initiation of antenatal care (late booking). Chapter one has considered the reasons why late booking matters and who the women are who book late. Chapter two has illustrated the complexity of beliefs and behaviours in relation to antenatal care initiation and the limitations of existing research. The original aim of this study reflected a pragmatic health services research approach: to develop a better understanding of late booking in a UK setting, in order to influence future practice. This was in response to concerns about health inequalities in relation to pregnancy, and particularly the links between poor access to care and adverse outcomes. As with the literature review, the research question evolved from this aim and was refined throughout the initial phases of literature searching and data collection, alongside the multi-professional research supervision. The process of formulating the research question again echoes Eaking and Mykhalovskiy’s (2003) view of the question as a compass rather than an anchor, opening up rather than restricting possible avenues of enquiry, framing the research in an answerable way, whilst aiming to avoid assumptions and pre-conceived ‘common sense’ answers (about late booking) (Green and Thorogood 2014). It is concerned with women’s experiences of late booking but also their beliefs in relation to this, and the particular significance of barriers to the early initiation of care. As such the overall research question has become “what perceptions and beliefs do late booking women express about the delayed initiation of antenatal care?”
The research question echoes chapter one, which considered that numerous studies and reports acknowledge that the statistics and demographic information around late booking are only one part of the story. The need to see beyond these, to understand women’s attitudes towards antenatal care and its initiation, in order to improve early access, has been expressed by both researchers and policy makers. This requires the application of research methods which allow a detailed examination of experiences and beliefs in relation to pregnancy and care. The challenges of researching a group of women who are potentially more vulnerable, and may be less articulate and confident, are considerable.
As chapter two has illustrated, a wide range of research methods can be used to investigate late booking, including both quantitative and qualitative approaches. Malterud (2001a,b) argues that such approaches are complementary rather than opposite, that they share underlying principles; it is the researcher’s role to choose “the path of inquiry that will most adequately provide valid accounts of the actual study field” (2001a: 399). The limitations of particular methods in achieving such ‘valid accounts’, particularly in relation to experiences and beliefs, are also acknowledged in chapter two. The use of quantitative methods, for example structured interviews and questionnaires, has many advantages, particularly the ability to survey large numbers of participants and cover a wide range of subjects. However, as some authors have reflected, such methods may be less personal, more superficial, and may fail to recognise the context of women’s lives. As a result they may generate results which are less honest and open, and which are unable to reflect the depth and detail of women’s views.
Both chapters one and two have considered the need for further research. However what is also evident from the literature review is the lack of a suitable template for the study of late booking in the UK. Unlike many of the US studies, there are no standard survey instruments available or appropriate to assess the adequacy or initiation of antenatal care. This necessitates a customised, purposive design rather than a standardised approach to answering the research question.
The title ‘qualitative research’ encompasses many forms of social inquiry, widely used in healthcare research, which focus on the way people make sense of their experiences and the world in which they live. In simple terms, whereas quantitative research focuses on questions such as ‘how often?’, qualitative research explores patterns of behaviour and subjective meaning, considering questions such as ‘why?’ and ‘how?’ (Pope and Mays 1995). They are largely interpretive approaches exploring behavioural processes and feelings, which allow for an in-depth exploration of attitudes, values and beliefs. As Holloway and Wheeler suggest, “qualitative research is a coherent way of researching human thought, perception and behaviour” (2010: 28).
Qualitative research has the ability to produce rich insights into human behaviour, social processes and complex issues, and as such it has been argued that it offers the potential for greater depth and validity than quantitative approaches (Murphy et al 1998, Bowling 2009). As Powell Kennedy (2011) argues, quantitative methods such as randomised controlled trials (RCTs), with their tight control over subjectivity and context, do not fully reflect the world of the individual and what is meaningful and relevant to them. Jenkinson (1997) acknowledges that qualitative methods have the capability to minimise the disparity between what people say they do and what they actually do.
The boundaries of qualitative research are sometimes less than clear: it has been described in terms of not being quantitative research (Pope and Mays 2006), but Bryman (2008) is also clear that it is not “quantitative research with the numbers missing” (p438). Qualitative researchers choose a variety of approaches and methods to achieve their aims. Denzin and Lincoln (2003) call these a “complex, interconnected family of terms, concepts and assumptions” (p3), and debates continue about the nature, purpose, status and methods of qualitative research (Murphy et al 1998, Paley and Lilford 2011). It is an approach, usually inductive in nature, which is not static but developmental and dynamic in character, with a strong focus on the process of the research as well as the findings. A flexible and emergent design, which evolves over the course of the research, reflects the desire for research to represent the realities and viewpoints of the participants, which may not be known at the outset of a study (Lincoln and Guba 1985). This is in contrast to the more rigid quantitative ‘testing’ of pre-determined ideas from the researcher’s perspective. As such, qualitative methods can access ‘challenging’ areas of social life that may not be open or amenable to quantitative research (Pope and Mays 2006). They have shown demonstrable advantages over quantitative approaches in situations (such as this) where there is little pre-existing knowledge and issues are sensitive and complex (Bowling 2009).
Qualitative methods for the study of health and health services are widely used and accepted as a legitimate method of attaining knowledge about practice issues (Hunter 2008). Their use reflects the dramatic changes in the ways in which health is understood and managed that have occurred in recent decades, beyond that which can be controlled, measured and counted (Malterud 2001a). As Green and Thorogood (2014) suggest, qualitative healthcare research is

“conducted to understand more about a phenomenon rather than measure it, and to investigate health, illness or health services from the perspective of the communities and individuals affected, or the professionals who provide health services for them” (p6)

It reflects the subtlety of beliefs and behaviours in health services and the influence of these. As Malterud (2001a) observes, “clinical knowledge consists of interpretive action and interaction – factors that involve communication, opinions and experiences” (p397). Qualitative approaches are specifically related to the consideration of these factors, in this study in the clinical context of antenatal care. They have the potential to focus on feelings, experiences and thoughts, but also consider change and conflict, for example issues relating to access and participation in services, and to evaluate changes in health service provision from the viewpoint of both patients and health professionals (Pope and Mays 2006, Holloway and Wheeler 2010).
Sword (1999) acknowledges that antenatal care utilisation behaviour is multifaceted and multidimensional, and that research must address this, but has failed to do so:

“to date, knowledge about barriers to prenatal care has evolved largely through positivist research approaches… based on what service providers and researchers perceive as important factors… the use of traditional methodologies also may oversimplify the notion of barriers because they do not allow for the exploration of psychosocial processes that determine use.. qualitative methods are appropriate… because they elucidate personal perspectives and contextual meanings”(p1174-5).

She argues that unlike other research methodologies qualitative research is able to facilitate an in-depth exploration of perceptions that are essential to understanding barriers arising within a given social context. Qualitative research is therefore essential to improve access to antenatal care:

“if the situated experiences and perceptions of socio-economically disadvantaged women are not accessed to inform health care delivery, these women will probably continue to encounter significant barriers to prenatal care and relationships that reinforce positions of powerlessness” (ibid p1175-6)

A qualitative approach was therefore considered highly appropriate for a study of women’s perceptions and beliefs, one with a Health Services Research focus, which aimed to influence (and improve) future practice through the development of ideas and interventions.
3.1.2 Influences on the methodology

Methodology can be defined as the principles and ideas, the theoretical framework, on which researchers base their procedures and strategies, i.e. their methods (Holloway and Wheeler 2010, Hall, McKenna and Griffiths 2012). It is important to explicate the ontological and epistemological principles which underpin the choice of methodology for any research. As Bondas (2011) argues:

“An understanding of the epistemological roots of the method chosen in relation to the topic of interest may enhance the research process and, ultimately the quality and depth of knowledge that is subsequently developed.” (p1).
In particular, good quality applied qualitative research must be theoretically informed. There are some very different theoretical, epistemological and political starting points in qualitative research, although many of them share a rejection of one or more of the elements of the positivist tradition (Green and Thorogood 2014). They reflect the inevitable tensions between rigour and originality, and the delicate balance of reflexivity with the voices of participants in the research. The desire to produce theories whilst reflecting the diversity of perspectives, and the need for systematic approaches whilst acknowledging the iterative and creative processes involved, are also manifest. All of these tensions are evident in this study.
The explication of the methodology begins with an acknowledgement that there is no single ‘superior’ approach to qualitative research, rather a number of complementary and overlapping approaches with more or less suitability for different research aims and settings. The qualitative methodology most appropriate to answer the research question is one which is:

1. 
   Largely Inductivist in its approach: whereby theory is generated from the research data;
   
2. 
   Constructivist in its ontological position, with postmodern and feminist influences; considering that behaviours and attitudes related to late booking for antenatal care are constructed as a result of interactions between people, rather than existing separately;
   
3. 
   Interpretivist in its epistemological position: aiming to understand the social world of late booking women by examining how they interpret their world; using the concepts of ‘lifeworld’ and verstehen (empathetic understanding); and
   
4. 
   Influenced by some of the principles of grounded theory.
   

These are explored in the following section.
3.1.2.1 Inductivism

With an inductive approach theory is the outcome of the research: drawing generalisable inferences out of observation, allowing what is relevant about the topic to emerge (or rather be constructed) from analysis. This is an iterative process of weaving backwards and forwards between data and theory, in clearly defined stages but not a linear process. This is associated with qualitative research methods and particularly grounded theory, and is in contrast to the hypothetico-deductive approach of most quantitative research, which involves testing or disproving hypotheses based round a pre-determined theory (Bryman 2008, Bowling 2009).

The study aimed to develop theories around late booking from the stories of women. The largely inductive approach, though informed by existing evidence, aimed to ‘keep an open mind’ rather than applying any predetermined theories about why women delay accessing antenatal care. This was reflected particularly in the data collection and analysis methods used in the study. However both qualitative and quantitative research can contain a balance of inductive and deductive elements, used for different purposes at different points in a study, and there is great potential (and necessity) for interplay and flexibility (Murphy et al 1998). As Sword (1999) comments:

“most qualitative studies lie somewhere between a loosely structured, emergent, inductively grounded method and a deductive, confirmatory technique” (p1175)

The generation of theories around late booking has entailed such combined thinking.
3.1.2.2 Constructivism

Constructivism is an ontological position (ontology: the nature of reality or existence) which views all knowledge as a compilation of human-made constructions, as opposed to an impartial discovery of an objective truth. Reality, or the world as we comprehend it, is therefore socially constructed. It is a product of social processes including context, time and culture, which can’t be separated out, rather than something external to us which is imposed. Individuals play an active role in the construction process, which influences perceptions and actions: ‘truth’ is therefore consensual (Murphy et al 1998, Bryman 2008). Lincoln and Guba (1985) suggest that the truth can be defined as the best informed and most sophisticated consensus, with the inquirer and inquired interlocked. Similarly participants, researchers and readers construct research through social interaction (Corbin and Strauss 2008, Holloway and Wheeler 2010). This is in contrast to positivist and objectivist/realist perspectives which suggest that there is one stable pre-existing reality out there waiting to be discovered. It has a strong tradition in qualitative health research where it plays a vital role in challenging assumptions or received wisdom and reading ‘beyond’ statistics (Green and Thorogood 2014).

Constructivism implies that social phenomena are produced through social interaction and are constantly being revised; the researcher’s own accounts are one part of this construction/reconstruction/revision, i.e. there is no single definitive answer. This suggests that knowledge is indeterminate, not fixed or inert; it may act as a point of reference but is always in the process of being formed (Bryman 2008). The categories we use to define and understand our world are social products – constructed in and through interaction, varying in time and place. Different social and cultural groups construct categories of meaning and behaviour in different ways (for example late booking): women’s views and doctors’ views are likely to be different. Particular meanings and connotations, for example negative or blaming, could be part of this construction, for example influenced by representation in the media (Bryman 2008).
Constructivism is allied to postmodernist views which reject the traditional view of absolute knowledge, stressing the plurality of perspectives and the importance of local context, and questioning the aspiration of objectivity and neutrality in research (Holloway and Wheeler 2010). Malterud (2001a) argues that postmodernist researchers are prepared to accept partial understanding of their topic and the identification of new questions as the outcome of their research, as opposed to definite answers. The postmodernist perspective also includes feminist views which suggest that ‘knowledge’ is a masculine construction and that fundamentally men and women have different views of the world based on different experiences, which inevitably will influence research. Qualitative research is seen as more consistent with the values of feminism, avoiding a value-neutral approach and accepting the need for empathy and ‘conscious partiality’ (Mies 1993). Feminist thinking openly acknowledges the subjectivity required to engage with participants as real women in real life settings, to create research of women for women (Harding 1987, Bryman 2008, Green and Thorogood 2014).
The constructivist influence is reflected in several aspects of this study. Firstly, is the acknowledgement of my role as researcher in the construction of the findings with the women, of honesty regarding reflexivity and its influence in the research. The empathy between researcher and researched, and encouragement of all ‘participants’ to play an active part in this construction, also reflects feminist influences. There is a recognition that the findings have not simply emerged from the data but are as a result of social processes. The findings aim to reveal the diversity and detail of perspectives, influences and the social context for the study, rather than trying to find the definitive ‘answer’ about late booking; to create something original and relevant, with validity for Sheffield in the first instance.
3.1.2.3 Interpretivism

The positivist epistemological position (epistemology: the theory of knowledge and its acquisition) is centred around a belief in objective reality, which can be ascertained by the senses (empiricism), tested subject to universal laws and rules and is generalisable to all settings. Positivist research is largely deductive and quantitative in its approach, seeking causal relationships, focusing on prediction and control using ‘facts’ (the result of systematic observation and measurement) and using predetermined theories/hypotheses as the starting point for research. Key to the positivist perspective is the necessary distance between researcher and researched to avoid bias, and the use of ‘objective’ measurement systems. Positivist traditions shape much health and healthcare research, for example clinical trials and standardised surveys or structured interviews (Bowling 2009, Holloway and Wheeler 2010).

The 19^th century views that led to the foundation of the positivist movement and the ‘scientific method’ are now viewed as simplistic, however many authors still identify the clash between positivism and hermeneutics (the study and interpretation of human behaviour). They argue that positivism is unachievable in research into human behaviour, being more concerned with facts, measurable behaviour and cause and effect, than context, feelings and subjective interpretations. Humans are different, they are complex and reflective, so a different approach to research is required (Bowling 2009, Green and Thorogood 2014). This echoes Alfred Schutz’s view from the 1940s of the fundamental difference between the natural sciences and the social sciences and the need for an epistemology that reflects this (Bryman 2008).
Interpretivism broadly encompasses many different non-positivist approaches. These approaches aim to interpret social phenomena, centring on the way that humans make sense of their experiences and attach meaning to them, and the importance of life context on these experiences. They focus on a reflective reconstruction and interpretation of other people’s actions and again mirror some feminist thinking (Pope and Mays 2006, Willis 2007). Max Weber’s concept of Verstehen reflects this, with an emphasis on empathetic understanding, a focus on internal, personal meanings rather than external influences/controls that have little meaning for ordinary people. It also recognises that understanding human experience is different to, but as important as, explaining it (Weber 1979, Bryman 2008).
Interpretivism is rooted in the philosophical approaches of phenomenonology and hermeneutics, whereby everything is subject to the acts of perception and interpretation. Research in such traditions aims to understand the ‘lifeworld’ of particular people: the world of unique but everyday experiences and meanings, putting aside, but not dismissing the influence of, existing prejudices, assumptions and theories (Bondas 2011). This lifeworld is created by those in it, through social interaction, and is not standardised across social and cultural groups, or existing separately to them (the objectivist view). It reflects Mead’s (1934) view of individuals as creative and thinking, exercising choice, not just reacting to external influences.
Interpretivist research focuses on discovering these social meanings, interpretations and subjective understandings, rather than an ‘objective reality’ (Green and Thorogood 2014). It requires multiple layers of interpretation: identifying people’s interpretations, interpreting these and then interpreting them again in terms of overarching concepts and theories. As Thome et al (2004) suggest, research aims to ask questions such as ‘what makes this lived experience what it is’ and ‘what is unique about this?’. Because experiences are context and language bound, interpretive approaches recognise that the values and interests of all the participants (including the researchers) become part of the research process. This presents challenges as it suggests that complete objectivity and neutrality are almost impossible to achieve and that studies cannot be replicated. It is therefore essential to proceed in a well-structured and systematic way, taking into account one’s own position and influence in the research setting - reflexivity (Bowling 2009, Holloway and Wheeler 2010).
An interpretive approach is evident in this study, which attempts to examine the complexity of human behaviours and attitudes in relation to antenatal care initiation, rather than adopting an empirical approach to late booking as an ‘objective reality’. As evidenced in chapters one and two, late booking has many facets and definitions; perspectives around it are subtle and complex, requiring layers of interpretation in the analytical process. Trying to understand the lifeworld of late booking women necessitates acknowledging the unique context that led to delayed initiation of care, and demonstrating an empathetic recognition of the value of these perspectives for our understanding. A positivist ‘cause and effect’ methodology is thus potentially simplistic. Interpretivist and constructivist approaches share common aspirations, such as all participants actively contributing to the research process, which has been an important part of this study. There are also common tensions, reflected in the research: between the need to put aside preconceptions whilst acknowledging that these inevitably have some influence, and a balancing of the unique nature of the study with a systematic approach, to ensure validity (Braun and Clarke 2006).

3.1.2.4 The influence of grounded theory

The chosen methodology acknowledges the fundamental influence of phenomenology, with its emphasis on participation and the interpretation of multiple, socially constructed and context-bound realities. In terms of practical application however, the study has adopted a more structured approach to data collection than is usual in phenomenological research. This is in response to the pragmatic health service focus and the stated aims of the study to examine the attitudes and beliefs of late booking women, in an attempt to improve understanding, access and, ultimately, health outcomes. The study mirrors the pragmatic and interpretivist influences associated with action research but stops short of the explicit focus on practical methods and behaviour change. It also reflects the examination of late booking as a distinct point in women’s lives rather than as ongoing ‘lived’ experiences, typical of phenomenological and ethnographic research and often explored through observation.
Grounded theory is an approach to qualitative research which matches the aspirations of the study and its theoretical framework. It has provided a systematic structure, influencing many of the processes and output of the study. Glaser and Strauss (1967) developed grounded theory as an inductive method of qualitative research which allows social theory to be generated systematically from data (i.e. from the ‘ground’ upwards) rather than being defined in advance, through a systematic process of rigorous and structured analysis. They argued that through such an approach qualitative research could advance from description and interpretation to theory development. Grounded theory has been hugely influential in changing the way that qualitative research is undertaken. It is a useful method for research into human behaviour and attitudes, and the factors that influence these, especially when little is known about the area of interest (Hall, McKenna and Griffiths 2012). It is a particularly relevant approach for examining relationships in healthcare, for example in midwifery, as Hall, McKenna and Griffiths (2012) observe:

“Grounded theory is an ideal research approach for exploring how midwives and women interact, and to examine the factors that mediate their decision making in clinical practice” (p136).

Grounded theory has been called “a compendium of different methods” (Corbin 2009: 41); it can refer to a methodology, a method of inquiry and to the product of inquiry. It has grown and developed in different directions, however in essence it remains “a set of flexible analytic guidelines that enable researchers to focus their data collection and build inductive… theories through successive levels of data analysis and conceptual development” (Charmaz 2008: 204). Grounded theory emphasises the reciprocal relationship between data gathering, analysis and theory. Theories are not causal relationships but are built from multiple observations; they are generalisable ideas or concepts designed to express plausible relationships or testable propositions rather than absolute ‘truths’. At its most effective, grounded theory can provide rich, detailed interpretations; outcomes which reflect the aims of this study (Strauss and Corbin 1998, Pope et al 2006, Lacey and Luff 2007, Bowling 2009).
The process of grounded theory is a creative one, grounded in a scientific approach: part of its appeal has been the structured and detailed procedures it provides for the generation of theory from data (Lacey and Luff 2007). It is not a linear process but involves cumulative, frequent revisiting of the data and revision of the methods in the light of new analytical ideas that emerge as data collection and analysis progresses. Many different methods and interpretations of grounded theory exist. Charmaz (2006) suggests a constructivist view (more recently accepted by Corbin and Strauss (2008)), which has been influential in this study, arguing that theories are co-constructed by the researcher and the participants, rather than ‘emerging’ from the data as Glaser and Strauss first suggested; a mixture of induction and deduction, moving between data and theory (an ‘abductive’ approach of reasonable inference (Charmaz 2012)).
Effective grounded theory is a complex and creative process requiring ‘reasoned reflections’, ‘principled convictions’ and ‘aesthetic merit’ (Charmaz 2008). However, it has been recognised that grounded theory can be defined more broadly, and that there are many elements of the inductive grounded theory approach that are useful for any analysis of qualitative data. Even Strauss and Corbin in their most recent book (2008) define grounded theory “in a more generic sense to denote theoretical constructs derived from qualitative analysis of data” (p1). This and other practical considerations have been influential in its development (Green and Thorogood 2014). In practice, as in this study, researchers frequently use the analysis procedures outlined in grounded theory without taking on board the whole methodological approach to research design. As Charmaz (2012) comments: “Most grounded theorists follow an iterative approach, many make comparisons, few construct theory. But the potential is there” (p2). Some have described this as ‘modified grounded theory’ and have praised this as an evolution of the method, a flexible way of thinking (Charmaz 2006, Pope, Ziebland and Mays 2006, Holloway and Wheeler 2010). The influence of grounded theory is evident throughout the research, particularly the use of constant comparison method: the overlapping processes of data collection, analysis and theory development, each enhancing the other.