There is an absence of national statistics for maternal obesity in the UK. This study is the first to describe a nationally representative maternal obesity research data set in England.
Retrospective epidemiological study of first trimester obesity.
Data from 34 maternity units were analysed, including 619 323 births between 1989 and 2007. Data analysis included trends in first trimester maternal body bass index status over time, and geographical distribution of maternal obesity. Population demographics including maternal age, parity, ethnic group, deprivation and employment were analysed to identify any maternal obesity-associated health inequalities. All demographics were tested for multicollinearity. Logistic regression analyses were adjusted for all demographics as confounders.
First trimester maternal obesity is significantly increasing over time, having more than doubled from 7.6% to 15.6% over 19 years (P<0.001), and shows geographic variation in incidence. There are also demographic health inequalities associated with maternal obesity, including increased odds of being obese with increasing age, parity, Black ethnic group and deprivation. There is also an association between morbid obesity and increased levels of unemployment.
The increase in maternal obesity has serious implications for the health of mothers, infants and service providers, yielding an additional 47 500 women per year requiring high dependency care in England. The demography of women most at risk of first trimester obesity highlights health inequalities associated with maternal obesity, which urgently needs to be addressed.
Maternal obesity has significant implications for the health of women and their babies. The Centre for Maternal and Child Enquiries (CMACE) summarizes the risks to the mother as being maternal death or severe morbidity, cardiac disease, spontaneous first trimester and recurrent miscarriage, preeclampsia, gestational diabetes, thromboembolism, post-caesarean wound infection, infection from other causes, postpartum haemorrhage and low breastfeeding rates.1 There is also a recognised, although relatively unexplored, psychological impact on obese pregnant women.2, 3 The risks to the infant are described as being stillbirth, neonatal death, congenital anomalies and prematurity.4 In addition, the increased risks during the antenatal, intrapartum and postnatal periods incur additional demand on NHS maternity services.5
The Health Survey for England (HSE) reported an increase in obesity among women of childbearing age from 12.0% in 1993 to 18.5% in 2006.6 CMACE also reported that of all mothers who died during 2000–2002 in the UK, 30% were obese (body mass index (BMI)>30 kg m−2).7 Between 2003 and 2005, more than half of all mothers who died were overweight or obese (BMI>25 kg m−2), with over 15% being morbidly obese (BMI>40 kg m−2) or super morbidly obese (BMI>50 kg m−2).1 Despite the HSE and CMACE data suggesting that obesity in pregnancy is increasing, there is a paucity of national or international statistics on the true incidence.
Three UK studies have shown that the incidence of maternal obesity has increased from 3.2 to 8.9% between 1990 and 1999 in Cardiff, UK,8 from 9.4 to 18.9% between 1990 and 2002/4 in Glasgow, UK,9 and from 9.9 to 16.0% between 1990 and 2004 in Middlesbrough, UK.10 The scale of obesity in the pregnant population on an international level has also been summarised as being between 1.8% and 25.3% according to data from published studies.11 However, there are difficulties with direct comparison of the international data because of the variation in the definition of obesity, the differences in time periods of the published studies, and the majority of studies representing regions of the United States and Australia.
This study is the first to compile a national level data set of maternal BMI, and to identify trends in maternal BMI over time as well as demographic inequalities relating to maternal BMI on a national and regional level in England.
A survey of routine electronic data collection of anthropometric measurements in pregnant women was carried out among all NHS maternity units in England (n=243) in 2006 (89% response). A total of 135 maternity units reported collecting anthropometric data electronically, and 58 of these indicated that they wanted to participate in the study. Forty-nine maternity units (32 NHS Trusts) were sampled as they reported collecting all data items required for the study electronically. Eight NHS Trusts were later excluded from the study: three owing to incorrect reporting of data collection, two owing to inadequate BMI records, one owing to R&D approval not being completed on time, one owing to staff shortages and one owing to staff changes. Thirty-seven maternity units (24 NHS Trusts) were included in the final sample. The demographics of women of childbearing age in the local authorities of the maternity units recruited into the study were compared with the demographics of women of childbearing age in England using the national census12 and Index of Multiple Deprivation13 reference data. The population was found to be nationally representative compared with women of childbearing age in the general population for all of the demographic variables to be incorporated in the analysis. This included ethnic group, deprivation, employment and parity (compared with the census data on a number of dependent children for each local authority).
NHS MREC approval was granted and R&D approval was gained from all NHS Trusts that provided the data for the study. Anonymized retrospective data were provided by the maternity units for all complete years of electronic data collection in their unit, and the data ranged from 1 January 1989 to 31 December 2007. Data were excluded when the booking BMI or gestational age could not be calculated; the BMI was unrealistic (<13.0 kg m−2);14 and when the gestational age at booking was unrealistic (based on a combination of clinical expertise and the NICE induction of labour clinical guidelines15). Previous research identified a lag effect between obesity in the pregnant population when compared with the general population of women of childbearing age.10 This phenomenon was potentially due to the exclusion of late bookers, which theoretically included a large proportion of the target population of obese women in pregnancy. This study adjusted for naturally incurred weight gain of late bookers (women who booked after their first trimester) using published data on BMI change per gestational week16 rather than excluding late bookers and potentially excluding a large proportion of the obese population.
Data analysis: trends in obesity incidence over time
Women were grouped on the basis of their BMI into the WHO categories of underweight, ideal, overweight and obese.17 Obesity subgroups were also analysed using the definitions of moderately obese, severely obese, morbidly obese and super morbidly obese. The χ2 test for trend (χ12) was used to investigate significant changes in proportions of BMI groups over time. The data did not require adjustment for age, as there was no significant change in population age over time (range in mean age over time 27 years, SD 5, and 29 years, SD 6).
Data analysis: geographical distribution of maternal obesity
The data were grouped into geographical regions using the Ordinance Survey Government Office Region (GOR) boundaries. There are nine GORs in England ranging in population size from 2.5 to 8 million.18 These boundaries are used for a range of administrative functions, and apart from one are coterminous with Strategic Health Authorities. The current trends in BMI groups for each region were calculated using the data for 2007 to identify any regional variation in maternal obesity incidence (with the exception of the two NHS Trusts that could not provide 2007 data and therefore 2006 data were used). Statistical significance in the distribution of BMI Groups was analysed using χ2.
Data analysis: demographic inequalities and maternal BMI
Logistic regression was carried out to analyse the relationship between BMI and demographic variables. Age and parity were continuous data, and ethnic group and employment were grouped on the basis of the national census.12 Ethnic groups included Whites (White British, White Irish and other Whites), Asians/Asian British (Bangladeshi, Indian, Pakistani and other Asians); Blacks/Black British (Black Caribbean, Black African and other Blacks); Mixed (White and Black Africans, White and Black Caribbeans, White and Asians, and other Mixed) and Chinese or other Ethnic Groups (Chinese and all other ethnic groups). Deprivation quintiles used postcode and the Index of Multiple Deprivation reference data.13 The rank of deprivation ranges from 1 (most deprived) to 32 482 (least deprived), and quintiles for the study group were defined in equal proportions. χ2 was used to test for an independent association between predictor variables and BMI group, and multicollinearity tests were carried out using linear regression diagnostics and Pearson's r correlation tests. No multicollinearity was present between the predictor variables, and therefore all were included in the final regression model.
Data were provided for a total of 738 307 deliveries. Following exclusions (16.1%), 619 323 deliveries remained. Some individual cases fulfilled multiple exclusion criteria. The leading reason for exclusion was insufficient data provided to calculate the BMI (88.9%). The characteristics of the included population are described in Table 1.
The pregnancy population change in BMI over time between the start and the end year of study is shown in Figure 1. This illustrates a substantial drop in the ideal BMI range, and a population shift to the right with increasing levels of obesity.
There was a significant trend in the proportion of women in each BMI group over time (Table 2). The increase in the proportion of women who are obese has doubled from 8% to 16% over 19 years of study (P<0.001), whereas there has been a 12% decrease in the ideal BMI group from 66% to 54% (P<0.001). Although the χ12 for underweight was significant with a minimum of 3.9% and a maximum of 6.2% of the population (P<0.001), overall it fluctuated around 5%. There was also a significant trend in the incidence of overweight with a gradual increase of 4% (P<0.001). A significant trend over time was also found for the obesity subgroups. The majority of the obese population in this study are moderately obese and there has been a 4.3% increase in the proportion of women in this group, from 5.7% to 10% (P<0.001). The increase in the remaining subgroups is proportionately lower and decreases as the severity of obesity increases. However, when comparing the rate of increase from 1989 to 2007, the relationship is seen to be increasing at a rapid rate within the morbidly obese group, moderately obese 1.75, severely obese 2.71, morbidly obese 4.0 and super morbidly obese 3.6.
Trends in this study were compared with women of childbearing age using HSE data (Figure 2), which shows a lag effect between the two populations. Trend lines were modelled for the data as a time series (with time points from 1 to 19 being the equivalent of 1989–2007), and the obese pregnancy population trend line shows a good fit with an exponential model (R2=0.9695), indicating that the increasing rates over time are accelerating rather than increasing in a linear manner.
The NHS Trusts that provided data included representation of all GORs with the exception of East Midlands, and there was a significant relationship between maternal BMI and GOR (χ2=826.2, P<0.001, 21 df). The incidence of first trimester obesity for the GORs was compared with the obesity prevalence in the general population of women using HSE data. Incidence of obesity in the pregnant population was lower than in the general population of women for all regions, with a difference of 7.4% in the overall proportion for England, and ranging from a minimum difference of 5.8% to a maximum of 10.7% for the individual GORs (Table 3). There are also different regional patterns of obesity in pregnancy when compared with the general population, although the West Midlands and the Northeast regions are in the top three for both populations. The East Midlands is the third most obese region for women in the general population, and HSE data show that it has previously been the region with the highest prevalence of obesity in women.19 On the basis of regional trends in the HSE population data maternal obesity incidence, an estimation of maternal obesity incidence in the East Midlands was calculated to range between 16.3% and 21.2%, with a mean of 18.8%, placing it among the top four obese regions in the pregnancy population. Figure 3 illustrates the GORs with higher than average, lower than average and equal to average incidence of maternal obesity.
The adjusted results of the logistic regression analysis for demographic predictors of maternal BMI groups are shown in Table 4. There is a significant increase in the odds of being overweight or obese with increasing parity and age. Overall, women who were underweight, overweight or obese were more likely to be employed (than unemployed, housewives or careers, or in education). This relationship did not remain significant when looking at the subgroups of obesity, where there was a significantly increased odds of women being housewives or careers if they were morbidly or super morbidly obese, and increased odds of being unemployed in women who were super morbidly obese. There were increased odds of women living in the more deprived quintiles throughout all BMI groups when compared with women of ideal BMI. There were increased odds for the overall obese women to be living in the most deprived quintile compared with the least deprived quintile, and when the subgroups of obesity were explored, the relationship with deprivation was seen to increase as the level of obesity increased. The ethnic group Black/Black British was the only ethnic group to have increased odds of overweight and obesity. However, this relationship decreased with increasing levels of obesity, and the relationship was no longer significant in the super morbidly obese group.
The results of this first nationally representative study have shown that first trimester obesity increases with time, where there are geographical differences in the incidence of maternal obesity, and where there are demographic health inequalities. The increasing rates of maternal obesity supports previous research carried out in the UK at the individual maternity unit level,8, 9, 10 although the actual proportions vary.
The increase in the proportion of women who are obese over time has important implications. Additional numbers of women who are considered to be at high risk result in additional care and support required during pregnancy. NICE guidance and CMACE recommend that women with a BMI>30 kg m−2 should have consultant care rather than midwifery-led care,1, 20 which places a massive burden on maternity unit resources. At a national level, the change in the proportion of women who are obese has doubled from 45 064 to 92 501 women (using the average number of births per year for all 243 NHS maternity units in England, 592 96021). Thus, approximately 47 500 additional women will require high dependency care in England every year as a result of the change in BMI over time. The small proportional increases in the obesity subgroups also have considerable implications for maternity services. The increase in the proportion of moderately obese women by 4.3% over 19 years results in an additional 25 500 women per year in England being in this BMI category, the 2.4% increase in the severely obese group results in an additional 14 000 women each year, the 1.2% increase in the morbidly obese group results in an additional 7000 women each year and the 0.2% increase in the super morbidly obese group results in an additional 1000 women each year.
The increase in first trimester obesity has major implications to clinical practice with the increasing demand for high dependency care, and the management of complications that arise. The regional differences in the incidence of maternal obesity identified suggests that there will be inequalities with some maternity units feeling the strain of the increasing demand on service more than others. This is particularly evident for maternity services located in the West Midlands, Yorkshire and the Humber, the North East, and the East Midland regions of England.
The lag effect between the pregnancy and general population of women identified in this study has also been described in the previous research.10 Previous research hypothesised that this may be related to physiological factors hindering fertility in the obese population, and may explain the existence of a lag effect identified in this study. There is a relationship between obesity and foetal loss,4 and this study used data on completed pregnancies rather than all pregnancies, because of the need to calculate the gestational age at booking from the gestational age at delivery. Thus, the results of this study may be an underestimation of maternal obesity, especially in the light of the latest CMACE report on perinatal mortality where mothers were obese in 22.9% of all late foetal loss and 30.4% of stillbirths.4
The demographic predictors of being obese in pregnancy highlight health inequalities that largely reflect previous research,10 particularly residing in areas of deprivation, which had the strongest relationship with obesity following adjustment for confounding variables. The additional analyses carried out in this study on the obesity subgroups shows a striking positive relationship with deprivation and increasing levels of obesity. Therefore, women who have the highest clinical risk (super morbidly obese) are those facing the highest level of inequality. A certain degree of caution must be noted with the super morbidly obese group because of the limited size of this BMI group in comparison with other BMI groups. However, the sample is large and the population characteristics are representative of women of childbearing age in the overall general population. The relationship with deprivation and inequalities in pregnancy is highlighted in the CMACE reports, where deprivation is significantly related to maternal death.1 The 2007 report identified that women who live in the most deprived areas are five times more likely to die compared with women living in the least deprived areas,1 and this finding in conjunction with the strong links with increasing levels of obesity and deprivation pose major health inequality issues to women residing in the areas of greatest deprivation in England.
Further inequalities exist with obesity, employment and ethnic group. Although analysis of overall obesity (BMI>30 kg m−2) shows that women are significantly less likely to be unemployed than employed, this result masks the relationship with increasing levels of obesity. There is a relationship with women being more likely to be unemployed or housewives/careers as the level of obesity increases. This finding is supported in the HSE data for women in the general population where obesity was found to be related to unemployment in women following adjustment for confounding variables. Obese women were 33% more likely to be unemployed than non-obese women, and this rose to 55% for severely obese women.22 The impact of unemployment in pregnancy is also highlighted in the 2007 CMACE report, which shows that a third of all women who died in pregnancy were either single and unemployed, or were unemployed with an unemployed partner.1 The limitation with the employment analysis is that the type of employment is not differentiated, and therefore the employment category will include a wide range of socioeconomic variation, from professional employment through to low-paid employment.
The results for ethnic group show a positive relationship with obesity and women being Black/Black British, which is representative of the relationship with women in the general population, where Black African and Black Caribbean women have the highest prevalence of obesity.23 In addition, the latest CMACE report also identified that Black African and Black Caribbean women had a higher risk of mortality during pregnancy when compared with white women.1 Interestingly, this study identified a significantly reduced relationship with Asian women and being overweight or obese, and this remained for all obesity subgroups. As there is an increased relationship with obesity and Asian women in the general population,23 this finding was unexpected to some extent. This inverse relationship with Asian women and obesity may be due to the association between obesity and age in women, where obesity is most raised in post-menopausal women.19 This may be more prominent in Asian women in the general population, making obesity most prevalent in post-menopausal women, and therefore not being reflective of women of childbearing age and the pregnancy population. There could also be physiological implications relating specifically to obese Asian women resulting in a high proportion of obese Asian women having fertility problems and thus leading to their exclusion from the pregnancy population. There is a relationship with infertility and central adiposity,24 and the HSE shows that women who are Bangladeshi and Pakistani have the highest risk ratio for having a waist–hip ratio over 0.85 (2.29 and 1.77, respectively, when compared with the general population).25
The relationship with obesity and increasing age and parity is similar to that observed in the previous research.10 These results also reflect the associations found in the general population, where increasing age and parity are linked with increasing levels of obesity,19 and pregnancy is a recognised life event in women in the promotion of obesity.26, 27, 28
This is the first study to address maternal obesity on a national level, and the strengths of the study are in its large sample size and the representativeness of the population when compared with England. The sample size has also allowed the first opportunity to identify the trends in the obesity subgroups of moderately, severely, morbidly and super morbidly obese.8, 9, 10
The relationship between obesity, ethnic group, deprivation and unemployment identified in this study indicate significant health inequalities in the demographics of those women most likely to be obese in pregnancy. In addition, the relationship between all of these factors, access to maternity services, and the risk of maternal death highlights how closely linked the issues surrounding health inequalities are in pregnant women. Further, national level research is required to identify the trends in Scotland, Wales and Northern Ireland to gain a UK perspective on maternal BMI. Also, there is limited evidence on the effectiveness of interventions in tackling maternal obesity, and further research is required to identify ways to halt the yearly accelerating rise in maternal obesity incidence in England.
Conflict of interest
The authors declare no conflict of interest.
Lewis G The Confidential Enquiry into Maternal and Child Health (CEMACH) Saving mothers’ lives: reviewing maternal deaths to make motherhood safer—2003–2005. The Seventh Report on Confidential Enquiries into Maternal Deaths in the United Kingdom. CEMACH: London, 2007.
Nyman VMK, Prebensen AK, Flensner GEM . Obese women's experiences of encounters with midwives and physicians during pregnancy and childbirth. Midwifery 2008; doi:10.1016/j.midw.2008.10.008.
Wiles R . I’m not fat, I’m pregnant. In: Wilkinson S, Kitzinger C (eds). Women and Health: Feminist Perspectives. Taylor & Francis: London, 1994. pp 33–48.
Confidential Enquiry into Maternal and Child Health. Perinatal Mortality 2005. CEMACH: London, 2007.
Heslehurst N, Simpson H, Ells LJ, Rankin J, Wilkinson J, Lang R et al. The impact of maternal BMI status on pregnancy outcomes with immediate short-term obstetric resource implications: a meta-analysis. Obes Rev 2008; 9: 635–683.
Information Centre. Statistics on obesity, physical activity and diet: England, January 2008. Health Survey for England. The Information Centre: London, 2008.
Confidential Enquiry into Maternal and Child Health. Why Mothers Die 2000–2002. RCOG: London, 2004.
Usha Kiran TS, Hemmadi S, Bethel J, Evans J . Outcome of pregnancy in a woman with an increased body mass index. BJOG: Int J Obstet Gynaecol 2005; 112: 768–772.
Kanagalingam MG, Forouhi NG, Greer IA, Sattar N . Changes in booking body mass index over a decade: retrospective analysis from a Glasgow Maternity Hospital. BJOG: Int J Obstet Gynaecol 2005; 112: 1431–1433.
Heslehurst N, Ells LJ, Simpson H, Batterham A, Wilkinson J, Summerbell CD . Trends in maternal obesity incidence rates, demographic predictors, and health inequalities in 36 821 women over a 15-year period. BJOG: Int J Obstet Gynaecol 2007; 114: 187–194.
Guelinckx I, Devlieger R, Beckers K, Vansant G . Maternal obesity: pregnancy complications, gestational weight gain and nutrition. Obes Rev 2008; 9: 140–150.
Census. Standard Tables: ST001 Age by Sex and Marital Status, ST028 Sex and Age by Economic Activity, ST101 Sex and Age by Ethnic Group, ST007 Age of FRP (Family Reference Person) and Number and Age of Dependent Children by Family Type. Office of National Statistics, 2001.
Index of Multiple Deprivation. The English Indices of Deprivation. Department of Communities and Local Government 2007.
Henry CJK . Body mass index and the limits of human survival. European J Clin Nutr 1990; 44: 329–335.
National Institute for Health and Clinical Excellence. Induction of Labour Clinical Guideline 070: An Update of NICE Inherited Clinical Guideline D. Department of Health National Collaborating Centre for Women's and Children's Health: London, 2008.
Ochsenbein-Kolble N, Roos M, Gasser T, Huch R, Huch A, Zimmermann R . Cross sectional study of automated blood pressure measurements throughout pregnancy. BJOG: Int J Obstet Gynaecol 2004; 111: 319.
World Health Organisation. Obesity; Preventing and Managing the Global Epidemic. World Health Organisation: Geneva, 2000.
Office of National Statistics. Population size: by country and government office region, UK, 2004. DirectGov, 2005.
Department of Health. Health Survey for England 2003. HMSO: London, 2004.
National Institute for Health and Clinical Excellence. Antenatal Care: Routine Care for the Healthy Pregnant Woman. RCOG: London, 2008.
Birth Choice UK. http://www.birthchoiceuk.com/ [Accessed16/06/2007].
Morris S . The Impact of Obesity on Employment in England. Tanaka Business School, 2004.
Department of Health. Health Survey for England 2004. HMSO: London, 2005.
Hollmann M, Runnebaum B, Gerhard I . Impact of waist–hip-ratio and body-mass-index on hormonal and metabolic parameters in young, obese women. Int J Obes 1997; 21: 476–483.
Information Centre. Statistics on obesity, physical activity and diet: England, 2006. Health Survey for England. The Information Centre: London, 2006.
Gore SA, Brown DM, Smith West D . The role of postpartum weight retention in obesity among women: a review of the evidence. Ann Behav Med 2003; 26: 149–159.
Siega-Riz AM, Evenson KR, Dole N . Pregnancy-related weight gain—a link to obesity? Nutr Rev 2004; 62: S105–S111.
Gunderson EP, Abrams B . Epidemiology of gestational weight gain and body weight changes after pregnancy. Epidemiol Rev 2000; 22: 261–274.
We like to acknowledge Teesside University for sponsoring this study, the participating maternity units for providing the data, and staff at the North East Public Health Observatory. This work was supported by Teesside University. JR is funded by a Personal Award Scheme Career Scientist Award from the National Institute of Health Research (Department of Health).
About this article
Cite this article
Heslehurst, N., Rankin, J., Wilkinson, J. et al. A nationally representative study of maternal obesity in England, UK: trends in incidence and demographic inequalities in 619 323 births, 1989–2007. Int J Obes 34, 420–428 (2010). https://doi.org/10.1038/ijo.2009.250
- maternal obesity
- body mass index
- health inequalities
Maternal and Cord Blood Metabolite Associations with Gestational Weight Gain and Pregnancy Health Outcomes
Journal of Proteome Research (2021)
Journal für Gynäkologische Endokrinologie/Österreich (2021)
Quantitative systematic review: Methods used for the in vivo measurement of body composition in pregnancy
Journal of Advanced Nursing (2021)
High-fiber diet mitigates maternal obesity-induced cognitive and social dysfunction in the offspring via gut-brain axis
Cell Metabolism (2021)
Understanding maternal mortality in women with obesity and the role of care they receive: a national case-control study
International Journal of Obesity (2021)