Women's perceptions of being pregnant and having pregestational diabetes
Introduction
Diabetes mellitus is one of the most common pre-existing medical conditions complicating pregnancy (Confidential Enquiry into Maternal and Child Health, 2007). The St Vincent Declaration's (Workshop Report, 1990) target that the outcomes of pregnancy in women with diabetes should approximate those of women without diabetes has not yet been reached; this is despite improvements in obstetric and neonatal care and diabetes management. Pregnancy in women with pregestational diabetes has been shown to be associated with adverse outcomes for both mother and fetus, with higher rates of labour induction (Landon, 1987; Nielsen and Nielsen, 1993; McAuliffe et al., 1999; Williams and Pickup, 2004), operative and premature childbirth in the mother (Nielsen and Nielsen, 1993; Hanson and Persson, 1993; Hawthorne et al., 1997; Evers et al., 2004) and five times higher rates of stillbirth, neonatal mortality and congenital anomalies in the offspring (Hawthorne et al., 1997; Casson et al., 1997; Hadden et al., 2001; Platt et al., 2002; Macintosh et al., 2006).
Although preconception care reduces congenital malformation in the babies of mothers with diabetes, most women do not discuss preconception care with diabetes health professionals when planning their pregnancies (Hawthorne, 2005; Confidential Enquiry into Maternal and Child Health, 2007). Ideally, all women with diabetes should plan their pregnancies with advice from their diabetes care team, as there is a strong evidence to show that such care can alter behavioural, medical and other health risk factors known to influence pregnancy outcomes (Curtis et al., 2006). In reality, however, a recent survey showed that over half of women with pregestational diabetes mellitus are poorly prepared for pregnancy (Confidential Enquiry into Maternal and Child Health, 2007). In this survey, less than half of women with diabetes were taking folic acid prior to conception, less than half received preconception counselling, and only one-third were recorded to have a test of glycaemic control in the six months prior to pregnancy. Furthermore, two-thirds of women are reported to have evidence of sub-optimal glycaemic control before conception and in the first trimester of pregnancy (Confidential Enquiry into Maternal and Child Health, 2005).
Although there is a plethora of literature reporting outcome data related to diabetes management in pregnant women (Kapoor et al., 2007), few qualitative studies have explored, in-depth, the issues around reproductive health and diabetes from the woman's perspective. Pregnant women considered to be ‘high risk’ are more vulnerable (Berg et al., 2003) and more anxious (Hatmaker and Kemp, 1998; Gupton et al., 2001) than those with normal risk status. One of the few qualitative studies exploring pregnancy and diabetes describes the woman as having a ‘high risk’ body which, for the unborn child's sake, is controlled by blood glucose levels (Berg and Honkasalo, 2000). A further study of 18 women with type 1 diabetes, by the same author, proposes a phenomenon summarised as ‘to master or to be enslaved’, highlighting the internal emotional battle for women of wanting to protect their child whilst remaining in control of their own experience (Berg, 2005). This analogy is an interesting way of illuminating the co-existence of negative and positive feelings felt by women. These Swedish studies offer important insight into women's perspectives; however, global and cultural variations in health-care practice limit their transferability.
Evidence to inform strategies that are likely to improve the maternity services offered to women with diabetes remains limited. Consequently, the diabetes community does not know which interventions are both effective and acceptable to women, and so could result in improvements in outcome. Multifaceted interventions that consider the complexities of women's beliefs, behaviour and social environment may be required to improve outcomes and the pregnancy experience. It may be unacceptable, for example, to developing a drop-in service for women with diabetes if consideration has not been given to the content of the session or the acceptability of its format to women of different socio-cultural groups. Furthermore, it may be that different options are needed to allow flexibility for individual needs. It is therefore imperative that a greater understanding of women's views is sought in order to target such interventions appropriately. It is only when research explores the perceptions of women living with diabetes that clinicians can begin to appreciate the challenges that these women face, and thus practice more effectively and empathetically. This study has therefore provided a key starting point in bridging the gap in current knowledge, having generated a greater understanding of issues relevant to women with diabetes in relation to their reproductive health.
Section snippets
Research design and methods
A hermeneutic phenomenological approach was used to explore the perceptions of women with diabetes from two different cultural backgrounds, namely White British and South East Asian, who had varied reproductive health experiences. Phenomenology aims to develop insights from the perspectives of those involved in a particular ‘lived experience,’ a term that is synonymous with this approach (Mapp, 2008). An interpretive phenomenological approach, i.e. hermeneutics, was used based on the philosophy
Sample and setting
A purposive sample of pregnant and non-pregnant women was obtained from those attending specialist diabetes clinics in two areas in the North West of England. Purposive sampling is appropriate when using a phenomenological approach due to the importance of selecting individuals who have knowledge of the phenomena concerned (Clifford, 1997). The sample was to include pregnant and non-pregnant women of different parity, ethnicity (White British and South East Asian) and diabetes type.
The first
Recruitment
Pregnant women with type 1 and type 2 diabetes were identified from clinic lists and information sheets were provided. Opportunity was given to discuss the study and, at a subsequent clinical visit, written consent was obtained.
The non-pregnant cohort attended clinic less frequently, usually once every six months. Therefore, women were identified from consultants’ clinic lists and information sheets were posted. Women were given time to consider participation and contacted to ascertain their
Data collection
Baseline data were collected on a brief questionnaire, prior to interview. Information sought included age, year diagnosed with diabetes, frequency of self-testing, diabetes management (i.e. diet, tablets, insulin), additional significant health problems, number of hospital admissions in the last five years and whether these were diabetes related or reproductive health related.
A semi-structured interview schedule was used to enable exploration of the broad areas of interest, whilst encouraging
Analysis
Interviews were audio recorded and transcribed verbatim, using pseudonyms to protect identity. Interviews conducted in Punjabi and Urdu were transcribed and translated by the lay researcher. Back translation was also conducted, i.e. the quality of the translations was verified by an independent translator translating back into the original language.
The data were managed manually, i.e. without the aid of a software package. The interpretive process was guided by the principles of the
Findings
Of 49 women who received the information sheet, 22 women consented to take part in the study and were interviewed; 20 individual interviews and two focus groups were carried out (n=3, n=2). Three women attended twice, at their request (two attended both focus groups; one had an individual interview and attended the second focus group).
Participant profiles are shown in Table 1.
Although African women were not part of the original eligibility criteria, an information sheet was sent to one woman in
Main themes
Three main themes contribute to the lived experience of the participants: relinquishing personal control, pregnancy overshadowed by diabetes and haphazard preconception care. Although presented separately, these themes are inter-related. The pregnant and non-pregnant women had similar views on pregnancy and diabetes, with the exception of ‘disruption to everyday lives’; a sub-theme of relinquishing personal control. Unsurprisingly, non-pregnant women had less disruption. Similarities were
Relinquishing personal control
Personal childbirth control has been reported as central to a positive experience (Lavender et al., 1999; Gibbins and Thomson, 2001). However, in this study, most women commented on the lack of personal control that they felt whilst being pregnant. They attributed this to the need to be compliant for the benefit of their own and their babies’ health. This lack of control presented itself in two main ways. Firstly, women expressed frustration at losing control to a system of care which disrupted
Pregnancy overshadowed by diabetes
All pregnant women commented on their disappointment that, during their hospital visits, the over-riding focus was on diabetes management, as opposed to the fact that they were pregnant. Rather than feeling reassured by the centrality of the diabetes to their antenatal visits, this made some women more anxious. Sabina, for example, said:
She explained if blood sugars are not controlled, baby could be aborted. You could have a heart attack. There are chances where it could have a negative effect
Haphazard preconception care
Only 35% of women with diabetes attend for preconception counselling (Confidential Enquiry into Maternal And Child Health, 2007). Women in this study appeared to be more inclined to go for preconception care if they had been directly referred, for example by their general practitioner, and were having difficulty conceiving. Yana, for example, had three years of infertility and attended for preconception care due to ‘pressure from outside’. This pressure came from her extended family and friends:
Discussion
This small exploratory study has provided insight into the lived experience of women of reproductive age with diabetes. As with many qualitative studies, this study generated a wealth of important data on women's experiences; however, given that the study aimed to explore views related to a number of variables (parity, diabetes type, ethnicity and pregnancy state), the sample size was limiting. Groups were not homogenous. The South East Asian women, for example, represented a number of
Conflict of Interest
None of the authors have any known conflict of interest.
Acknowledgements
The authors wish to thank the women who participated in this study, the lay researchers (Parveen Ramzan, Safira, Patel Narmin Rokad and the diabetes specialists (Christine Bond, Pam Inniss, Sue Spellman, Lorna Wood and Gill Morrison). The study was funded by Diabetes UK.
References (33)
- et al.
Women's expectations and experiences of childbirth
Midwifery
(2001) - et al.
Complicated and uncomplicated pregnancies: women's perception of risk
Journal of Obstetrics and Gynecologic Neonatal Nursing
(2001) Preconception care in diabetes
Seminars in Fetal and Neonatal Medicine
(2005)- et al.
Neonatal morbidity in pregnancy complicated by diabetes-mellitus—predictive value of maternal glycemic profiles
American Journal of Obstetrics and Gynaecology
(1987) - et al.
A prospective study of women's views of factors contributing to a positive birth experience
Midwifery
(1999) - et al.
Outcome of 328 pregnancies in 205 women with insulin-dependent diabetes-mellitus in the county of Northern Jutland from 1976 to 1990
European Journal of Obstetrics Gynecology and Reproductive Biology
(1993) Pregnancy and diabetes: how women handle the challenges
Journal of Perinatal Education
(2005)- et al.
Pregnancy and diabetes—a hermeneutic phenomenological study of women's experiences
Journal of Psychosomatic Obstetrics and Gynaecology
(2000) - et al.
Childbirth experience in women at risk: is it improved by a birth plan?
Journal of Perinatal Education
(2003) How do you analyse qualitative data