The prevalence of obesity in pregnancy is increasing: around 1 in 5 women attending antenatal care in the UK are obese and in the current obesogenic environment, with a rise in the number of obese teenagers reaching child-bearing age, this figure is likely to fluctuate (Kanagalingam et al, 2005; Shah et al, 2006; Heslehurst et al, 2010). Pregnancy is also a significant causative factor in the development of obesity; women with high weight gain during pregnancy tend to retain more weight at 15 year follow-up (Linné et al, 2004). Thus women who have had a high gestational weight gain (GWG) are more likely to commence their next pregnancy with a higher starting body mass index (BMI).
Obesity and excess GWG have both been linked to an increased risk of complications during pregnancy and birth (Cedergren, 2004; Heslehurst et al, 2008). Complications include: gestational diabetes mellitus, pregnancy-induced hypertension, venous thromboembolism, postpartum haemorrhage and caesarean section (Sebire et al, 2001; Usha Kiran et al, 2005; Bhattacharya et al, 2007). Obesity is also known to increase the risks of shoulder dystocia, birth defects, fetal and neonatal death and stillbirth (Robinson et al, 2003; Chu et al, 2007; Rasmussen and Yaktine, 2009; Stothard et al, 2009). Some of the birth risks are directly related to the increase in large for gestational age (LGA) infant. Antenatal care costs may be 5–16 fold higher in overweight and obese women (Heslehurst et al, 2007).
Scott-Pillai et al (2013) in their retrospective study of 30 298 women giving birth in Belfast identified that for women with a BMI at booking of between 35 and 40 kg/m2 there was a 6.0 times greater risk of developing gestational diabetes, a 1.8 increased odds ratio of requiring a caesarean section, a 2.0 times greater risk of the baby suffering shoulder dystocia and a 2.2 times greater risk of the baby being stillborn. A healthy BMI of 20–25 kg/m2 was used as the standard for comparison.
Obesity and excess GWG are the most probable independent risk factors for the associated increase in maternity complications, but there is limited evidence available to determine what a healthy weight gain should be for the already obese woman. In Kiel et al's (2007) study of 120 251 women, outcomes for pre-eclampsia, caesarean section and LGA babies were much improved in women with a baseline BMI ≥40 kg/m2 if the they lost between 2-9lbs or even ≥10 lbs. Only in the case of small for gestational age (SGA) babies was there a slightly impaired outcome with gestational weight losses greater than 10 lbs. Collectively, the data showed minimal risks for all four of their outcomes, for example where the risk of LGA and SGA births lines intersect, equated to a weight gain of 10–25lbs for class 1 (BMI ≥30 kg/m2) obese women, a gain of 0-9lbs for class 2 (BMI ≥35 kg/m2) obese women and a loss of 0–9 lbs for class 3 (BMI ≥40 kg/m2) obese women.
Health professionals are often uncomfortable raising the issue of maternal obesity (Smith et al, 2012), and referral options and evidence-based interventions to tackle obesity during pregnancy are limited (Dodd et al, 2008). UK guidelines on appropriate weight gain during pregnancy are also lacking but the American Institute of Medicine (IOM) guidance is often used with the BMI specific recommended weight gains of 7.0-11.5 kg for women who are overweight and 5.0-9.0 kg for women who are obese at the start of pregnancy (IOM, 2009). There is no separate guidance for women who commence the pregnancy morbidly obese with a BMI ≥40 kg/m2
The National Institute for Health and Care Excellence (NICE) 2010 guidance on weight management before, during and after pregnancy is clear that restrictive diets and weight-loss during pregnancy should be avoided. There is some evidence suggesting that inadequate gestational weight gain may be a risk factor for intrauterine growth restriction, preterm birth, low birth weight and perinatal mortality. However, none of these studies examine the effects in an already obese pregnant population (Carmichael and Abrams, 1997; Rasmussen and Yaktine, 2009). Similarly, none of these studies considered dietary or lifestyle habits which may have influenced the amount of gestational weight gain.
Advice from NICE and the government has suggested free or subsidised attendance at commercial weight management groups for obese (non-pregnant) patients with a growing evidence base to support this treatment option (NICE, 2006; Jolly et al, 2011).
The healthy eating and lifestyle in pregnancy (HELP) intervention was developed by a senior midwife as an intervention for obese pregnant women given the lack of NHS resources available to support women to manage their weight during the antenatal period. The intervention aimed to equip women with the knowledge, skills and confidence to control weight-gain during pregnancy, as well as maintaining a healthy lifestyle postpartum, through healthy eating and increases in physical activity. The intervention was delivered through a weekly group held at the antenatal clinic, jointly facilitated by an experienced midwife and Slimming World (SW), a commercial weight management organisation who were invited to provide the lifestyle component of the intervention by the midwife.
The main aim of the HELP study was to assess the feasibility and acceptability of referring pregnant women with a BMI ≥30 kg/m2 into the weekly group. The secondary objective was to determine the impact of the intervention on weight status across pregnancy and link this to obstetric and neonatal outcomes. Thus secondary outcome measures include maternal gestational weight change, postpartum maternal weight, baby birth weights, mode of birth, birth complications including postpartum haemorrhage, shoulder dystocia, LGA and SGA and method of feeding.
Methods
Design
This study is a non-randomised single arm feasibility trial, following the Medical Research Council (MRC) framework for complex interventions (Craig et al, 2008) to inform the need, design, processes and outcomes for a future definitive randomised controlled trial. The intervention and approaches are to be underpinned by NICE guidance on principles for effective interventions to support behaviour change and motivational support (NICE, 2014), which underpins SW weight management programmes.
Recruitment
Recruitment took place between May 2008 and September 2010. All pregnant women were screened by the midwife at booking (usually between 10 and 12 weeks gestation) and women with a BMI ≥30 kg/m2 were given an invitation to attend the weekly group support and a study information leaflet. The only inclusion criteria were that the women were over 18 years old, had a BMI ≥30 kg/m2 and that they understood English sufficiently to participate in the intervention groups. All midwives who would be raising the issue of weight management at booking were provided with training. At the first visit to the intervention group, the women were screened by the study midwife for any pregnancy complications or medical factors such as pre-existing diabetes or twin pregnancy, which may effect attendance at the group. Information was given about the study and informed consent taken. An initial weight, height and BMI were recorded.
Intervention
The intervention groups were held weekly within the hospital antenatal clinic, during the early evening to promote access for women who were working or who had no childcare during the day. There was no charge for attendance at the group and all resources were free. The weekly sessions were facilitated by a midwife and SW consultant. All women attending the groups were given a new member pack including the SW pregnancy and breastfeeding leaflet, which details the current recommendations for healthy eating and food safety during pregnancy plus advice on extra energy and nutrient needs during breastfeeding, and a Body Magic leaflet (which promotes increases in individuals physical activity levels). Women were encouraged to follow SW's food optimising healthy eating plan with regular healthy meals and snacks. This plan is based on current healthy eating guidelines with guidance by experienced dietitians; it encourages a varied and balanced diet with particular attention to calcium- and fibre-rich foods. It helps limit the intake of sugary and fatty foods and provides a healthy balance of all major food groups without the need for a severe calorie restriction or avoidance of key food groups or nutrients.
The weekly group sessions were led by a midwife whose remit was to discuss all specific pregnancy-related issues and a SW consultant facilitating healthy eating and lifestyle behaviour change. Sessions were structured to allow all components of the intervention to be delivered: social support, motivation, self-monitoring, goal setting, supporting self-efficacy and midwifery support. The format of group sessions included:
If women did not attend the group, the SW consultant contacted them (via telephone message) to encourage continued attendance. This is normal SW practice in standard groups. Women were able to attend the group sessions immediately following acceptance of the invitation and for up to 6 weeks post-delivery. They were then encouraged to join a community-based SW group at a locality of their choice but at this point were required to pay the weekly group fee.
Outcomes and follow up
Women were followed up until 6 weeks postpartum. Data on session attendance and weights during pregnancy were collected on a weekly basis. Weight and height data recorded at booking were taken as baseline values. The weight data was verified using the Slimming World record cards to ensure accuracy and collect any missing data.
The woman's date of birth, parity, antenatal and labour complications, mode of birth, gestation at delivery, birth weights and feeding method were obtained from the maternity computer system. Data were analysed descriptively and mean values are presented with standard deviations in brackets unless otherwise stated.
Study participants were invited to a small focus group following the feasibility study to inform the main study design. The group was facilitated by SS and KJ and was semi-structured and covered the topics; recruitment, motivation to attend, group content, barriers to attending and general practicalities. This group was audio recorded and later transcribed and analysed using a thematic analytic approach.
Ethical approval
Ethical approval was granted by Cardiff and Vale NHS Trust research department.
Results
Recruitment
There were 148 women recruited to the study (Figure 1); the feasibility study did not collect data on numbers who declined.
The mean pregnancy length at initial group attendance was 12 weeks (2.5) but with a range of 7 to 26 weeks. The mean age of group attendants was 32 years (5.3 SD; range 19-48). Parity mean value was 0.8 (1.0) indicating that most women were on their first pregnancy. A BMI at booking was available for 95 women with a mean value of 37.4 kg/m2 (5.5 SD; range 30-58.7).
Attendance
The mean attendance at the group was 10 weeks (7.7 SD; range 2–35). Of a possible 33–35 sessions, 85% of the pregnant women (n=126) attended at least 3 group sessions and 65% (n=ç6) at least 6 group sessions (Figure 2). Of the 126 women who attended at least three group sessions there was a reduced attendance 3 weeks before giving birth but with re-engagement with the group sessions after delivery. Parity did not appear to influence attendance.
Maternal weight change
For the 44 women who attended the group within 3 weeks before delivery, a mean gestational weight gain of 4.3 kg (0.7 SD) was recorded. For the whole study sample, figure 3 shows the overall weight change recorded at the last pre-delivery attendance, showing the full range (-14.1 to +13.2 kg). Of the sample population, 24 women gained weight within the IOM guidelines, 96 gained less than the IOM guidelines or lost weight and 7 gained more than the IOM guidelines during the gestational period. For the 39 participants who lost weight, the mean weight loss was 2.5 kg (2.9 SD).
For those 58 women who returned to the group after having had their baby, the mean weight gain during pregnancy was 3.2 kg with a maximum weight gain of 13.2 kg recorded.
Table 1 outlines the data for the weight change post-delivery for the 58 women who returned to the group after giving birth showing a mean weight-loss since booking in pregnancy of 5.6 kg (0.7 SD) and range of -17 to +7.3 kg.
| Mean weight change(SD) | Maximum weight change | Minimum weight change | |
| Weight change before giving birth | +7.2 lb (1.5) |
+29 lb |
-24.5 lb |
| Total weight change since joining group | -12.4 lb (1.5) |
+ 16 lb |
-37.5 lb |
Mode of birth
The mean length of gestation was 39.7 weeks (2.0 SD). Of 132 births where mode of birth was recorded, 48% were spontaneous vaginal births, 5% Ventouse, 9% forceps, 18% elective caesarean section and 19% emergency caesarean section. Postpartum haemorrhage was recorded for 48 women with a mean blood loss of 958 ml.
Birth weights
Birth weights were available for 132 of the births and the mean birth weight was 3.53 kg (0.49 SD). One baby was born prematurely at 33.7 weeks. There were 16 cases of macrosomia with birth weights ≥4000 g. Figure 4 illustrates the range of birth weights in the total study sample. For those women who lost weight during pregnancy (n=39), the mean birth weight for their babies was 3.59 kg (0.35 SD). No babies in this study sample were born with a weight below 2500 g, but two were born with a below 3000 g. Five of these babies had a birth weight >4000 g. Figure 5 illustrates the birth weights of those babies born to women who lost weight during pregnancy.
Baby feeding method
Of those who attended the HELP group, 89% of women (n=115) chose to breastfeed following birth. At 28 days post-birth 70% were still breastfeeding.
Adverse events
Of the 128 infants for whom Apgar scores were available, the mean score was 8.4(9.5). Only two infants had a score below four at 1 minute and no infants had a score of five or less at 5 minutes.
There were no cases of shoulder dystocia in any of the deliveries. One baby was stillborn. The woman chose to return to the group immediately after giving birth for support and subsequently achieved a healthier BMI before conceiving again.
Focus group analysis
Two study participants attended a small focus group, numbers were limited by the fluidity of the study sample/timing. The main themes discussed are described below.
Recruitment
Participant 1 stated that the initial approach regarding joining the group was ‘uncomfortable’. She was told by an obstetrician:
‘your BMI is off the scale and you will be attending this group’
but was later told not to attend as she should not be dieting in pregnancy. Participant 2 had already lost weight prior to pregnancy and had heard about the study from a midwife, but when she went to the hospital no information was given. Later at home she noticed a letter inviting her had been put inside the notes,
‘it was as if they were embarrassed to mention it.’
When asked about the terminology that was used when they were approached, they both were uncomfortable with the terms ‘obese’ or ‘morbidly obese’. Participant 1 suggested:
‘don't use obese or morbidly obese, use BMI and a visual chart is good because you can see where you are’.
Motivation to attend the group
Participant 1 indicated her concerns about keeping her baby healthy and to be healthy herself for the future:
‘I wanted to keep myself and my baby healthy…also to change things in the future on my next baby, little things like some in the group talking about water births and things that you can't have unless you are below a certain BMI.’
Participant 2 indicated that she wanted things to be different this time compared to previous pregnancies.
‘I didn't want to put on as much as last time and get diabetes again.’
Group content
When asked what they saw as the purpose of the groups, participant 1 stated:
‘it was never about weight loss it was all about how you felt and your weight journey.’
Participant 2 stated:
‘weight loss aside, it was actually being in a room with people in similar situations in terms of weight and pregnancy.’
The women were asked if having both a SW consultant and midwife was important. Participant 2 stated:
‘you need both absolutely…together it married so you felt that it's like somebody was almost looking after you and somebody was looking after your baby.’
The subject of continuous recruitment to groups was also discussed, as women were at different stages. The women were asked whether this had an impact. Participant 1 stated:
‘I never found that an issue, no, because I mean it was interesting to see the people at the end of their pregnancies having their babies and seeing the results and thinking ooh I could be like them.’
Barriers to attending the group
The issue of barriers to attendance was discussed as sessions had been held within the hospital in the evening, participant 1 stated:
‘I think that sometimes people drop out, its not possibly the pregnancy thing it's more about…what's happening in their life…well I think you've got more chance if somebody coming home on their way home from work popping in than not at all’
Participant 2 stated:
‘It's because you're so tired as well that's the difficulty and because you were working. In winter when you know it's dark at 4 o'clock I know for a fact that I hated coming here in the dark.’
Group practicalities
One participant liked the fact it was in the hospital as a venue. She said:
‘I think it also homes it in that it's related to the fact that you are pregnant as well being in the hospital and it brings it all together doesn't it, and also there is that care from the midwife.’
The subject of attending postnatally was discussed, participant 1 stated:
‘Oh yes M was born on the Wednesday if I could've come down to the group on the Thursday after I would've been there.’
The women were also asked what they felt was the optimum size of the group, participant 1 thought that the group should have a minimum of four or five, and participant 2, six.
Discussion
It is widely acknowledged that having a higher BMI at the start of a pregnancy and excessive weight gain during pregnancy will increase the health risks to both the woman and infant (Bhattacharya et al, 2007; Siega-Riz et al, 2009; Centre for Maternal and Child Enquires and the Royal College of Obstetricians and Gynaecologists, 2010). However, it is unclear what constitutes a healthy gestational weight gain. NICE (2010) do not recommend BMI-specific weight changes during pregnancy.
In 2007, Kiel et al convincingly demonstrated the benefits of restricting GWG in obese women. More recently, in their meta-analysis of evidence from 44 randomised controlled trials, including 7278 women, Thangaratinam et al (2012) concluded that dietary and lifestyle interventions can reduce maternal weight gain and improve outcomes for both woman and baby. However, the overall evidence rating was low to very low for important outcomes, such as pre-eclampsia, gestational diabetes, gestational hypertension and preterm delivery.
The LIMIT study (Dodd et al, 2014), where healthy eating and exercise advice was offered to pregnant women who are overweight or obese showed a significant reduction in the number of babies born over 4 kg in weight. While the relatively intensive advice and assistance to adopt a healthy diet and regular exercise during pregnancy led to an 18% reduction in the chance of a baby being born over 4 kg, there was no significant difference in the maternal weight change between the intervention and control groups. Also, the antenatal lifestyle advice used in the LIMIT study did not reduce the risk of having a baby who weighs above the 90th centile for gestational age and sex or improve other maternal pregnancy and birth outcomes measured.
Regardless of the limited guidance and evidence of best practice, given the prevalence of obesity among women of child-bearing age, there is a need for scalable interventions which support obese pregnant women to prevent excessive GWG and encourage healthy lifestyles and postpartum weight loss.
This study examined feasibility, acceptability, recruitment and retention as well as other key outcomes. Recruitment was slow initially with members of the maternal health care team finding it difficult to raise the issue of obesity during pregnancy and to discuss the associated increased health risks. In some cases, it took time for the team members to recognise the benefits of referring to the HELP group. An experienced midwife-led the pathway and passionately encouraged referral into the group and after the first 6 months the referrals gathered momentum. When the group closed there was general widespread disappointment with few other options available locally to provide the same level of support.
It is suggested that the women who were supported by the group felt very positive about the experience, ‘I felt I had no choices when they told me I was too big, but coming to group and keeping control of my weight has given me my confidence back’ and ‘last time I put on 4 stone and had diabetes, this time I am in control’. It is likely that not only the dietary and lifestyle changes contributed to the outcomes observed but also the social support which a group, well facilitated, provides.
The study was not powered to detect an effect on weight or other secondary outcomes. However, in this small feasibility study of 148 women, with an average start BMI at booking of 37.4 kg/m2, 39 women did lose some weight during pregnancy. This did not appear to increase the likelihood of them having a SGA baby; indeed it may have reduced the risk, given no babies were born to this group of women with a birth weight below 2500 g. They also appeared to be less likely to have a LGA or macrosomic baby weighing greater than 4000 g.
For those women attending the group within 3 weeks of giving birth, the mean GWG of 4.3 kg was just slightly less than the 5.0-9.0 kg recommended by the IOM for obese women. For those women who returned to the group after having their baby (mean 1.8 SD 0.31 weeks postpartum), the mean GWG was 3.3 kg with a maximum GWG of 13.2 kg recorded. In these small sub groups there does not seem to have been any adverse effects to either the woman or the baby through what may be described as an ‘inadequate’ weight gain.
For this feasibility study no data were collected about actual dietary changes made as a result of the intervention. However, as per the SW pregnancy policy, any woman losing above a certain amount of weight at any time point during the pregnancy is required to complete a food diary. Thus there is some anecdotal information available with many of the women changing their previous high-fat, energy-dense, nutrient-poor eating habits for less energy-dense but plentiful amounts of nutrient-rich food choices.
The intervention aimed to help the pregnant women believe that they possessed the necessary skills to change their eating and activity behaviours and that their actions would help to improve both their health and that of their unborn child. This was to be achieved through information sharing, supporting self-efficacy, modelling of behaviours and social support. In the non-pregnant population social support is associated with improved weight loss as well as increases in people completing treatment and maintaining weight loss (Wing and Jeffery, 1999; Elfhag and Rossner, 2005). The social support was perceived as particularly valuable by the woman who had the stillborn baby who returned to the group for both the support and also to prepare for her next pregnancy.
It is hypothesised that the social support may have been one explanation for the particularly successful breastfeeding rates, both at birth and 28 days post-delivery, in those women attending the HELP group. At the time when the data were collected, the breastfeeding rates were 52% at birth and 44% at the 5–6 day screening data compared to the respective figures of 89 and 70% for the HELP study group—with the latter figure of 70% being recorded 28 days post-delivery. Besides the social support, the increased self-confidence may also have contributed to these high rates of breastfeeding, as may the improvements in BMI since obese women may find it more difficult to breastfeed (Anstey and Jevitt, 2011). SW's pregnancy literature does encourage breastfeeding and there is additional dietary advice included to support successful breastfeeding and to ensure that the additional nutritional demands are met. Improving breastfeeding rates continues to be high on the public health agenda, to ensure that each child gets the best start in life (Department of Health, 2010) and breastfeeding for longer may also reduce the subsequent risk of childhood obesity (Armstrong and Reilly, 2002; Harder et al, 2005; Lefebvre and John, 2014).
The HELP pilot study group ran every week for over 2 years with two midwives and two SW consultants involved to ensure holiday cover and continuity. Across the UK there are over 4000 trained SW consultants and 11000 community groups running each week, so there is a very sound infrastructure and robust supporting resources were this intervention to be proven effective in a large trial and subsequently rolled out. Hence it would be relatively easy to replicate the HELP group either in a hospital or community setting with midwife support. Women, when they were no longer able to access the hospital group, were able to go to a local community group.
Limitations
The study has a number of limitations, including the small sample size, the limited outcome and process measures, the small focus group sample size as well as the lack of a control group. These will be addressed in a follow-up randomised controlled trial.
Conclusion
The results of this feasibility study appear quite promising. Referring pregnant women with a BMI above 30 kg/m2 at the booking clinic into the HELP pilot group was found to be feasible and acceptable although the referrals were slow at the start. Women were positive about the group attendance. After attendance at the group, women who were both followed up within 3 weeks of giving birth and who had post-delivery recorded mean weight gains below that recommended by the IOM. Thirty-nine women lost weight during the gestational period but the mean baby birth weight was no different than for those women not losing weight. Indeed the extremes of birth weight were reduced in those women losing weight with fewer small and large for gestational age in this small group of women. Further research is required to determine whether either minimal gestational weight gain or weight loss in the clinically obese woman is desirable providing the restricted weight change is achieved through healthy dietary changes and increase in physical activity levels.
A number of women were lighter post-delivery than when joining the group and besides the usual health benefits this will have the added value of allowing them to start their next pregnancy with a healthier BMI providing lifestyle changes are maintained. The other significant finding from this feasibility study is that the antenatal group support appeared to have an impact on breastfeeding rates with more women attending the group going on to breastfeed their infants. If this is due to the intervention, this is an important finding given the many health benefits associated with breastfeeding.
Given that one in five women are now starting their pregnancy with a BMI in the obese category there is a need for effective, scalable interventions. Further research is needed to develop interventions which limit GWG and encourage post-partum weight loss and test these in rigorously designed randomised controlled trials. Some of the findings from this feasibility study were used to inform the development of a large randomised multicentred trial which recruited (ISRCTN25260464) across 20 UK hospital sites.