Dietary behaviours and weight management: A thematic analysis of pregnant women's perceptions

Abstract Maternal obesity is associated with increased risk of gestational diabetes and other complications. Although antenatal interventions to help prevent these complications are ongoing, an understanding of overweight and obese pregnant women's opinions and attitudes is lacking. Therefore, this study aims to explore these women's experiences and perceptions of dietary behaviours and weight management during pregnancy. Secondary analysis of qualitative data originally collected to examine lifestyle behaviours in pregnant women was conducted. The data were from a purposive sample of overweight and obese pregnant women attending a public antenatal clinic in Cork, Ireland. The data were explored using thematic analysis. Interviews with 30 overweight and obese pregnant women were analysed. Three themes were developed relating to overweight and obese women's dietary behaviours and weight management perceptions including ‘pregnancy's influence on dietary behaviours’, ‘external influences on dietary behaviours’ and ‘perception of and preferences for weight related advice and resources’. Together these themes reveal women's experiences of diet and how pregnancy factors (physiological changes) and external factors (family and friends) can influence dietary behaviours. Furthermore, perceptions of weight management advice and lack thereof were highlighted with women drawing attention to potential resources for future use during pregnancy. This study provides important insights into overweight and obese pregnant women's dietary behaviours and perceptions of weight management. According to these findings, there is a need for clear and unambiguous information about weight management, acceptable weight gain, food safety and how to achieve a balanced diet.


| INTRODUCTION
Pregnancy is a key stage in a women's life when dietary habits and weight management are of major importance (Lindsay, Heneghan, McNulty, Brennan, & McAuliffe, 2015). Despite this, the prevalence of overweight and obesity has increased drastically over the years resulting in more women being obese at the onset of pregnancy (Goldstein et al., 2017). In Europe, estimates of the prevalence of maternal obesity among women aged 20 to 39 years range from 30% to 37% (Devlieger et al., 2016). In Ireland, 20%-25% of pregnant women have obesity (Fattah et al., 2010;Lynch, Sexton, Hession, & Morrison, 2008). Overweight is defined as body mass index (BMI) ≥25 kg/m 2 , and obesity is defined as a BMI ≥30 kg/m 2 which is assessed at the first antenatal consultation (Centre for Public Health Excellence at Nice National Collaborating Centre for Primary, 2006).
Overweight, obesity and excessive gestational weight gain are associated with complications such as gestational diabetes mellitus, preeclampsia, caesarean section and pre-term delivery (Catalano & Ehrenberg, 2006;Lynch et al., 2008). Additionally, maternal obesity is associated with risks to the infant including macrosomia, infant and childhood obesity (Catalano & Ehrenberg, 2006).
Nutrition and weight gain during pregnancy have important implications for subsequent maternal and offspring health (Krasovec & Anderson, 1991). Insufficient nutrient intake during pregnancy, such as increased intake of saturated fat or processed foods, overeating and increased frequency of snacking or decreased frequency of lunch eating during or after the pregnancy, can have a negative impact on health outcomes (Brantsaeter et al., 2009;Maslova, Halldorsson, Astrup, & Olsen, 2015;von Ruesten et al., 2014;Wu, Bazer, Cudd, Meininger, & Spencer, 2004). Furthermore, maternal diet has been shown to be associated with foetal growth and birth size (Bouwland-Both et al., 2013;Knudsen, Orozova-Bekkevold, Mikkelsen, Wolff, & Olsen, 2008).
The increasing rate of maternal obesity has led to national guideline recommendations for the development of interventions to improve pregnancy outcomes (Yaktine & Rasmussen, 2009). This advice stimulated many clinical trials Poston et al., 2015;Sagedal et al., 2017;Szmeja et al., 2014), predominantly of behavioural interventions addressing diet and physical activity. Systematic reviews of these trials suggest potential for the prevention of gestational diabetes in women with obesity (Rogozi nska, Chamillard, Hitman, Khan, & Thangaratinam, 2015). However, most of these trials have been small and underpowered for clinical outcomes such as gestational diabetes and have focused instead on gestational weight gain (Thangaratinam et al., 2012).
Pregnancy has been identified as an ideal time to promote healthy eating and physical activity as women are more likely to experience strong emotional responses to their pregnancy and be motivated to make changes due to the well-being of the foetus (Phelan, 2010).
Additionally, the continuous contact with health care professionals at antenatal visits provides a vital opportunity to interact regarding diet and weight (Beckham, Urrutia, Sahadeo, Corbie-Smith, & Nicholson, 2015). Despite this opportunity, dietary advice and information on BMI and weight are insufficient for the majority of pregnant women (Markovic & Natoli, 2009).
Although it may be important to target all pregnant women, some studies suggest that interventions can impact women in healthy weight range differently to those who are overweight and obese (Hui et al., 2006;Jeffries, Shub, Walker, Hiscock, & Permezel, 2009;Phelan et al., 2011;Polley, Wing, & Sims, 2002) and that women in higher BMI categories should be considered independently and provided with more intensive interventions (Phelan et al., 2011). Therefore, it is essential to gain a greater understanding of this high risk populations perspectives in order to inform effective antenatal dietary and weight management interventions (Atkinson & McNamara, 2017;Cavanagh & Chadwick, 2005;Furber & McGowan, 2011). The aim of this study is to explore overweight and obese women's experience and perception of dietary behaviours and weight management during pregnancy.

| METHODS
This is a qualitative study using naturalistic inquiry to provide an interpretive description of previously collected qualitative data (Sandelowski, 2000). The aim of the conducted interviews was to collect rich qualitative data on diet, physical activity, weight management, technology use and future interventions. The focus of the original study was to identify barriers and enablers to physical activity using two theoretical frameworks, the theoretical domains framework and the COM-B model. The focus and insights provided in the current paper are in relation to pregnant women's perceptions of diet, dietary behaviours and weight management. For this secondary analysis of the previously utilised qualitative data set, two additional researchers joined the study team. The data were used and analysed for a different purpose and provides significantly different insights to the original study.

Key messages
• The prevalence of overweight and obesity has increased drastically over the years resulting in more women being obese at the onset of pregnancy.
• Understanding women perceptions of diet and weight management is fundamental to inform the development of effective antenatal dietary and weight management interventions.
• Pregnancy factors such as physiological changes and external factors such as family and friends can influence dietary behaviours.
• There is a need for clear and unambiguous information about weight management, acceptable weight gain, food safety and how to achieve a balanced diet.

| Study design and population
Secondary data analysis of qualitative interviews collected from a sample of overweight and obese pregnant women at risk of gestational diabetes was conducted. Data were originally collected to examine physical activity behaviour in overweight and obese pregnant women, as previously described (Flannery et al., 2018). In brief, medical chart review by midwife or researcher on site, identified pregnant women with a body mass index (≥25 kg/m 2 ) recruited during pregnancy from a public antenatal clinic at Cork University Maternity Hospital (CUMH) , Ireland. Eligible participants were approached and informed about the study by researcher (CF) on site at their antenatal appointment.

| Data analysis
Secondary analysis was performed using the interview data from the original study (Flannery et al., 2018 ). A thematic analysis was conducted to develop themes relating to weight management and diet for overweight and obese pregnant women (Braun & Clarke, 2006). NVivo software was used to facilitate data analysis. An inductive approach was used, where transcripts were read and re-read numerous times by the researcher (CF). Transcripts were coded line-by-line by researcher (CF) and a subset of transcripts (n = 6) were coded by two co-authors (MNM, KMS). Following open-coding, categories were developed, discussed, and synthesised to develop broader overarching themes (CF, MNM, KMS ). Discrepancies regarding the codes, categories and themes were resolved through team discussions (CF, MNM, KMS X).

Recruitment continued until no new dimensions, nuances or insights
were found (Clarke & Braun, 2020). The purpose and goal of this analysis and pragmatic considerations were discussed with the team.

| Ethical considerations
This study is reported according to the consolidated criteria for reporting qualitative research (COREQ) statement (Supporting Information S1). Ethical approval was obtained from the University College

| RESULTS
Of the 30 women who took part in the interviews, over half were Irish (n = 16), BMI ranged between 20 and 40 kg/m 2 , and were between 10 and 39 weeks pregnant at time of interview. See Table 1 for all participant characteristics including age, nationality, BMI and gestational age. Interviews ranged from 23 to 50 min in duration.
Three major themes were developed that relate to overweight and

| Pregnancy's influence on dietary behaviours
Pregnancy and pregnancy symptoms influenced women's food preferences and dietary behaviours, with women trying to 'balance the good with the bad'. For some, pregnancy and its associated 'physiological changes' influenced women's food preferences, whereas for others, pregnancy was seen as an 'excuse' to indulge. The final subtheme of 'responsibility' highlighted that although diet can be influenced by body changes and cravings, the pregnant woman was ultimately responsible for her dietary choices, her health and health of the baby.

| Physiological changes
Physiological changes during pregnancy were described by women as a factor that impacted their dietary behaviours. Women who experienced nausea or vomiting discussed how they would restrict certain foods, whereas others experienced food aversions. Women who experienced heartburn and/or indigestion would also limit the consumption of certain foods. Now I suffer really badly with heart burn … so that would stop me eating a lot of food…Like cereal for example, if I eat Special K I get heart burn. So I am constantly being watchful. (BMI 27 kg/m 2 , age 35) Furthermore, some women experienced a change in food preferences, an increase in appetite or food cravings. Cravings for specific foods were also common among women in this study with food items such as fast food, crisps and confectionery sweets.

| Responsibility
Some women in this study expressed having a sense of 'responsibility' for providing the best for the baby even while experiencing body changes, and food cravings. Furthermore, women discussed 'making changes' and being more conscious of their diet and overall health behaviours while pregnant. Despite this, for some, pregnancy provided a reason to not make healthy changes '… like sure I'm pregnant. I'm going to be big anyway' (BMI 27 kg/m 2 , aged 35). Women felt that pregnancy could be used as an 'excuse' and was considered a 'free pass' in terms of their food choices, with some women ignoring or overlooking healthy eating during pregnancy.

| External influences on dietary behaviours
Women's experiences of receiving dietary advice and support throughout pregnancy varied. Women described how their social environment including family, friends and their health care professionals influenced or hindered healthy food choices in pregnancy.

| Social influences
Women's friends, family and relatives, in particular the women's partners, played a significant role in providing dietary advice and support, which was often based on their partners own experience or on hear-

| Perception of and preferences for weightrelated advice and resources
Despite some receiving information on diet, the majority of women in this study 'lacked weight related advice' during pregnancy. Women felt that health care professionals provided limited guidance on appropriate weight gain in pregnancy, with some women enduring uncomfortable and upsetting circumstances when weight was addressed.
Furthermore, women highlighted 'weight management resources for pregnancy' and how these could be beneficial in supporting a healthy diet and weight management in the future.

| Lack of weight-related advice
Similar to that of diet-related advice, most women described the weight-related information in pregnancy as "vague" and "insufficient".
Some women reported trying to get advice from their health care professionals but felt their answers did not support weight or their efforts to manage it. Therefore, women did not feel fully informed or supported in their weight management efforts. Furthermore, most women expressed an interest in caloriecontrolled diets or meals plans as a support mechanism that would enable them to make healthy food choices and help to manage weight gain in pregnancy.
If they sit down with you and say here is a meal plan that you can follow and these are the foods you can have. Then you can say to yourself well I can substitute that and this and can add this. It's much easier like that. In this population, pregnancy itself had an influence on dietary behaviours. Pregnancy-related physiological changes such as cravings or food aversions have previously been identified as an influence on dietary behaviour (Quinla & Hill, 2003;Richter, 2003). Although pregnancy-related symptoms are outside the women's control, they are considered typical features of pregnancy (Quinla & Hill, 2003).
Therefore, woman should consider their diets and healthy dietary behaviours preconception and into early pregnancy. Furthermore, as these physiological changes are expected, future interventions need to provide women with the skills to better manage nausea, vomiting, heartburn and indigestion.
Some women revealed 'responsibility' as a main reason for being more conscious of their diet and overall health behaviour's while pregnant. In other qualitative studies, this is a commonly reported reason for adopting a healthy diet in pregnancy, the health of the baby and mother (Bianchi et al., 2016). However, for others, pregnancy provided an excuse to indulge. Similar research found that because weight gain can be inevitable during pregnancy (Johnson et al., 2013), some women felt it was a time to relax the rules relating to healthy eating and lifestyle. Previous findings have highlighted perceptions of pregnancy as a time where women could eat large portions and gain weight confidently (Chuang, Velott, & Weisman, 2010;Kraschnewski & Chuang, 2014). For instance, one study involving relatively health-conscious pregnant women found that pregnancy was viewed as a time when they could take a break from their usual healthy lifestyles (Atkinson, Shaw, & French, 2016). Traditional beliefs such as 'eating for two' were still very much evident with women openly sharing their experiences. Changing the culture of "eating for two" in society is necessary to successfully help pregnant women to understand the importance of diet and weight management in pregnancy (Atkinson et al., 2016;Kraschnewski & Chuang, 2014).
Previous research has highlighted that a number of social groups play a significant role in dietary behaviours and weight management during pregnancy; this includes women's mothers (Prichard, Hodder, Hutchinson, & Wilson, 2012), their spouse (Pachucki, Jacques, & Christakis, 2011), as well as friends (Cruwys et al., 2012). Partners, peers and family members have been identified as positive influences on the initiation of healthy eating behaviours (Schaffer & Lia-Hoagberg, 1997). However, in this study, women's husbands were highlighted in a slightly negative manner in which they criticised or monitored the woman's dietary habits during pregnancy. A systematic review of health behaviour change interventions for couples demonstrated high concordance between partner's health behaviours A lack of reputable dietary information has been reported by other studies, which examined lifestyle behaviours during pregnancy (Gross & Pattison, 2007;Weir et al., 2010). This was echoed in this study as women specified that information was presented on behaviours to avoid, but weight management information was limited.
Women felt the information they received lacked focus on favourable foods in pregnancy and weight management. Previous research found that women sought out information themselves due to the lack of information and conflicting advice they received from their midwife on lifestyle behaviours in pregnancy (Brown & Avery, 2012). It is likely that issues of weight are not raised or addressed due its sensitive nature (Blackburn, Stathi, Keogh, & Eccleston, 2015

| Strengths and limitations
The thematic approach used to analyse interviews was a strength of this study as it revealed the nuances and experiences of overweight research is warranted to further assess racial or cultural differences and whether information needs to be tailored for these women.

| CONCLUSION
This study provides important insights into overweight and obese pregnant women's dietary behaviours and perceptions of weight management. Future research is needed to develop strategies that address these perceptions and target specific dietary behaviour's as well as addressing inadequate and conflicting educational messages in pregnancy. Including women's husband/partners in dyadic and familyoriented interventions could play an important role in promoting and improving diet and weight management for these women. Our results suggest that clear and unambiguous information about acceptable weight gain, food safety and how to achieve a balanced diet is needed. Tailoring existing and effective weight loss programmes for pregnancy might be another potential option to guide woman's dietary habits and help them in their weight management efforts during pregnancy.

ACKNOWLEDGMENTS
The authors would like to acknowledge the midwives and staff at Cork University Maternity Hospital (CUMH) for their support and assistance with recruitment. We would also like to thank the pregnant women who agreed to be interviewed for this study. CF was funded by the Health Research Board SPHeRE/2013/1 during this research.
The Health Research Board (HRB) supports excellent research that improves people's health, patient care and health service delivery. The HRB aim to ensure that new knowledge is created and then used in policy and practice. In doing so, the HRB support health system innovation and create new enterprise opportunities.