Increased maternal BMI is associated with an increased risk of minor complications during pregnancy with consequent cost implications
Dr FC Denison, Centre for Reproductive Biology, Queen’s Medical Research Institute, 47 Little France Crescent, Edinburgh EH16 4TJ, UK. Email firstname.lastname@example.org
Objective To investigate the effect of maternal body mass index (BMI) on minor complications, associated additional medication use during pregnancy and the consequent cost implications.
Design Retrospective analysis of case notes.
Setting Labour wards, tertiary referral hospital, Royal Infirmary Edinburgh, UK.
Population Six hundred and fifty-one women with a singleton pregnancy over four separate time periods in 2007 and 2008.
Methods Descriptive statistics, univariate and multivariate logistic regression analysis and cost analysis using standard techniques and inflation indices.
Main outcome measures Minor complications, use of medications during pregnancy and consequent incremental costs from the perspective of the National Health Service (NHS).
Results 42.4% of women were overweight or obese (BMI ≥ 25 kg/m2). Higher BMI during the first trimester (BMI ≥ 30 kg/m2 compared with BMI < 25 kg/m2) was associated with an increased risk of minor complications including symphysis pubis dysfunction (OR 3.97; 95% CI 2.19–7.18), heartburn (OR 2.65; 95% CI 1.42–4.94) and chest infection (OR 8.71; 95% 2.20–34.44) and with drugs used to treat these complications including Gaviscon (OR 3.52; 95% CI 1.78–6.96). The mean incremental (additional) NHS costs per woman for treating minor complications increased with maternal BMI were £15.45/woman, £17.64/woman and £48.66/woman for BMI < 25 kg/m2, BMI ≥ 25 to <30 kg/m2 and BMI ≥ 30 kg/m2 respectively.
Conclusions Increased maternal BMI is associated with increased risk of developing minor complications during pregnancy; use of medications associated with treating these conditions and has significant NHS costs.
The prevalence of obesity in women of childbearing age in the UK is increasing1–3 with recent studies estimating that more than one in five pregnant women are obese.4 Pre-gravid obesity has significant implications for maternal and fetal health. Maternal risks include increased risk of developing major complications such as gestational diabetes mellitus, pregnancy-induced hypertension, pre-eclampsia during pregnancy and having a complicated delivery including caesarean section.5–10 For the offspring, the short- and long-term risks include an increased risk of congenital anomaly and stillbirth, neonatal morbidity and mortality and early onset childhood obesity.6,11
Most studies reporting complications in pregnancy associated with maternal obesity have focused on major morbidity5–10 and have not commented on minor complications. Although symptoms such as heartburn are common and are generally perceived as being benign, they can be associated with significant morbidity, increased healthcare usage, and therefore increased costs.12,13 The objectives of this study were therefore to investigate the effect of body mass index (BMI) at booking on the risk of developing minor complications during pregnancy and to evaluate the consequent cost implications for the NHS.
A retrospective analysis of antenatal notes and labour ward records was undertaken over four separate periods during 2007–2008 (11th–18th July and 13th–20th August 2007 and 10th–20th March and 21st April–3rd May 2008) at the Simpson Centre for Reproductive Health, Royal Infirmary of Edinburgh, Edinburgh, UK. The study periods coincided with the time periods in which undergraduate medical students were doing their 4th year research projects. The case-notes of every delivery occurring on the labour ward during these time periods were interrogated with data including demographics, antepartum and peripartum information, reporting of minor complications during pregnancy and medication use being recorded. Maternal BMI (kg/m2) was calculated from maternal weight and height measurements taken at the first community antenatal booking visit (10–12 weeks gestation). Cases were grouped into the following categories: normal (<25 kg/m2), overweight (≥25 to <30 kg/m2) and obese (≥30 kg/m2).
All pregnancies in which maternal BMI could be calculated from height and weight recorded at the booking visit and which ended with a liveborn singleton delivery over 24 completed weeks gestation were included in the analysis.
The main outcome measure was development of a minor complication during pregnancy as recorded in the notes by a healthcare professional (midwife or doctor). The following minor complications were recorded: headache, carpal tunnel syndrome, constipation, sciatica, hyperemesis, urinary tract infection, chest infections and persistent generalised itch. In addition, major complications (gestational diabetes, pregnancy-induced hypertension and pre-eclampsia) and intrapartum interventions (artificial rupture of membranes, as a proxy for abnormal progress in labour, induction of labour and caesarean section) were also recorded. Postcode was used to calculate the Deprivation Category score (DEPCAT). This postcode sector-based scoring system is based on four indicators (overcrowding, male unemployment, social class 4 or 5 and no car)14 and generates a scaled measure of deprivation where one is the most affluent area and seven is the most deprived.14
Ethical approval was obtained from the local Lothian Research Ethics Committee (Reference 07/S1103/25).
A cost effectiveness analysis was conducted from the perspective of the NHS with the outcome of interest being the risk of minor complications. The NHS staff costs, facilities costs and consumables costs were identified for each category of minor complication, which after multivariate analysis, was associated with maternal BMI. After consultation with three experienced obstetricians, six experienced community- and hospital-based midwives and two physiotherapists, an estimate was made of the expected percentage of women with each complication and the anticipated duration of each problem. This then permitted unit costs to be referred to from standard sources including Scottish Health Service Costs and the British National Formulary.15–17 These information were then used to calculate the total and mean costs of complications per patient. Finally, the incremental (additional) NHS costs per patient for minor complications were calculated for subjects in each BMI category.
As the outcome data was collected over four separate periods during 2007–2008 it was decided that this should be regarded as being collected during financial year 2007/08. Discounting was not deemed to be necessary as none of the costs were incurred in a time period that lasted for more than 12 months.
Descriptive statistics were tabulated for all outcomes and covariates and their association with maternal BMI. Differences in the explanatory covariates across the four time periods were tested using chi-squared tests or ANOVA as appropriate. Logistic regression analysis was used to explore the effect of known or suspected prognostically important covariates (BMI, parity, age, smoking status and DEPCAT) on minor pregnancy complications outcomes. These covariates were fitted first individually and then jointly, all together in a multivariable model. The number of women with a pre-existing medical problem, for example pre-gestational diabetes (n = 5), asthma and thyroid disorders was very small. When these co-morbidities were entered into the multivariate analysis they did not contribute to the model and did not show a statistically significant relationship with minor complications when other factors were taken into account. These variables were therefore removed from the final analysis. Statistical significance was set at P < 0.05 throughout. No adjustments were made for any multiple comparisons. SAS 9.1.3 for Windows XP (SAS Institute, Cary, NC, USA) was used for all statistical analyses.
In 28 of the 671 case-notes interrogated, height, weight or BMI were missing, leaving a final study group of 651 for analysis. Maternal age, anthropometric and demographic characteristics of the study group are summarised Table 1. The proportion of women of normal weight (BMI < 25 kg/m2), overweight (BMI ≥ 25 to <30 kg/m2) and obese (BMI ≥ 30 kg/m2) at the first trimester booking appointment was 57.6%, 26.1% and 16.3% respectively. There was no difference in the characteristics of subjects attending in the four different study periods, specifically no difference in smoking (P = 0.45), DEPCAT scoring (P = 0.41), age (P = 0.71), BMI (P = 0.06), birthweight (P = 0.90) or gestation at delivery (P = 0.79). Maternal BMI was not associated with maternal age, smoking, DEPCAT score or gestation at delivery (all P > 0.05).
Table 1. Descriptive data for maternal age, anthropometrics and demographics for the study group
|Age (years)||29.7 (6.03)|
|BMI at booking (kg/m2)||25.2 (5.28)|
|Smoking habits in early pregnancy|
|>1 per day||108 (16)*|
|Birthweight (kg)||3.42 (0.57)|
|Gestation at delivery (days)||278.2 (17.1)|
|DEPCAT scores (1–3, most affluent)||283 (43)*|
|DEPCAT scores (4–7, most deprived)||388 (57)*|
The higher the maternal BMI, the greater the likelihood of reporting the minor complications of heartburn, symphysis-pubis dysfunction and chest infection with trends for increased reporting of carpal tunnel syndrome (Table 2). There was no statistically significant association between increased maternal BMI and risk of headache, constipation, sciatica, hyperemesis, urinary tract infection or generalised itch (Table 2).
Table 2. Absolute risk (%) and adjusted odds ratios (95% CIs) tabulated for minor complications according to first trimester BMI grouping (<25 kg/m2, ≥25 to <30 kg/m2 and ≥30 kg/m2)
|Symphysis pubis dysfunction||7.7||8.2||26.4||1.00||1.16 (0.60–2.25)||3.97 (2.19–7.18)||<0.0001|
|Chest infection||0.8||1.8||7.6||1.00||2.28 (0.45–11.55)||8.71 (2.20–34.44)||0.01|
|Heartburn||8.0||8.2||18.9||1.00||1.02 (0.52–1.99)||2.65 (1.42–4.94)||0.005|
|Carpal tunnel syndrome||2.1||4.7||5.7||1.00||2.42 (0.88–6.66)||3.02 (1.01–9.04)||0.09|
|Headache||3.2||5.3||7.6||1.00||1.64 (0.67–4.01)||2.27 (0.89–5.79)||0.21|
|Constipation||2.9||2.4||0.9||1.00||0.84 (0.26–2.73)||0.33 (0.04–2.62)||0.57|
|Sciatica||2.7||0.6||1.9||1.00||0.22 (0.03–1.72)||0.65 (0.14–3.08)||0.33|
|Hyperemesis||2.9||2.9||2.8||1.00||1.11 (0.37–3.30)||0.97 (0.26–3.59)||0.98|
|Urinary tract infection||9.6||7.7||9.4||1.00||0.81 (0.41–1.61)||0.88 (0.41–1.88)||0.82|
|Generalised itch||2.9||2.9||4.7||1.00||1.02 (0.35–3.04)||1.48 (0.49–4.46)||0.77|
Increased maternal BMI was also associated with a greater likelihood of developing major complications of pregnancy of induced hypertension and intrapartum interventions of caesarean section and induction of labour (Table 3). Because of the small numbers of women with gestational diabetes (n = 5), this complication could not be fitted to the model. However, the mean BMI was significantly higher in women with gestational diabetes [mean, SEM; 34.2 (0.2)] compared to those without [25.2 (2.8); P < 0.0001]. Finally, increased maternal BMI was associated with an increased likelihood of taking Gavison during pregnancy with trends for taking insulin and antihypertensives (Table 3).
Table 3. Absolute risk (%) and adjusted odds ratios (OR; 95% CIs) tabulated for medication, major complication or intrapartum interventions, according to first trimester BMI grouping (<25 kg/m2, ≥25 to <30 kg/m2 and ≥30 kg/m2)
|Gaviscon||5.6||6.5||17.0||1.00||1.17 (0.55–2.51)||3.52 (1.78–6.96)||0.001|
|Insulin||0.5||1.8||4.7||1.00||3.02 (0.49–18.63)||6.82 (1.21–38.51)||0.09|
|Antihypertensives||2.4||4.7||6.6||1.00||2.05 (0.77–5.49)||3.08 (1.10–8.58)||0.09|
|Antibiotics||12.5||11.8||17.9||1.00||0.96 (0.54–1.69)||1.38 (0.76–2.53)||0.51|
|Inhalers||3.2||3.5||5.7||1.00||1.10 (0.40–3.02)||1.38 (0.47–4.10)||0.84|
|Thyroxine||1.6||2.4||1.9||1.00||1.08 (0.26–4.43)||1.10 (0.21–5.67)||0.99|
|Pregnancy-induced hypertension||2.9||5.9||10.4||1.00||1.83 (0.73–4.59)||4.34 (1.79–10.55)||0.01|
|Pre-eclampsia||0.5||2.4||2.8||1.00||4.91 (0.86–24.04)||6.26 (1.00–39.19)||0.11|
|Caesarean section||18.1||27.9||34.0||1.00||1.65 (1.05–2.60)||2.47 (1.46–4.19)||0.002|
|Induction||31.8||41.6||46.7||1.00||1.50 (1.01–2.23)||1.79 (1.11–2.91)||0.02|
|Artificial rupture of membranes||31.0||39.8||43.3||1.00||1.46 (0.98–2.18)||1.64 (1.01–2.66)||0.06|
The incremental (additional) costs per patient from the perspective of the NHS for treating women with symphysis pubis dysfunction, chest infection, heartburn and carpal tunnel syndrome were £75.30, £73.13, £108.64 and £16.20 respectively. The mean additional NHS costs per person per BMI category associated with treating these minor complications was £15.45, £17.64 and £48.66 for women in BMI categories <25 kg/m2, ≥25 to <30 kg/m2 and ≥30 kg/m2 respectively. Women with a BMI of ≥30 kg/m2 accounted for 39%, 57%, 31% and 27% of the additional NHS costs for symphysis pubis dysfunction, chest infection, heartburn and carpal tunnel syndrome respectively.
Although the association of obesity with increased major morbidity during pregnancy is well described,5,6,8–10,18 there are no studies which have specifically investigated the influence of obesity on minor complications of pregnancy and its associated economic implications for the NHS.19 Here, we demonstrate that increased maternal BMI is also associated with increased reporting of minor complications and medication use during pregnancy. With the increasing prevalence of obesity in the antenatal population,1 this has significant cost implications for the NHS.20
In large studies relating pre-pregnancy weight or BMI to risk of adverse pregnancy outcomes, minor complications have not been the main focus of the study. The majority of these studies are based on interrogation of large databases in which the associations between maternal obesity and pregnancy outcomes are derived from analysing the variables contained within pre-existing databases.5,6 Although major morbidities such as pre-eclampsia and maternal obesity are usually entered, minor complications such as heartburn, symphysis-pubis dysfunction and chest infections, are seldom recorded.6 Despite these complications being common, there is a paucity of data about risk factors for their development during pregnancy. Moreover, recent studies suggest that heartburn21,22 and symphysis-pubis dysfunction23 may have long-term implications and not be as ‘benign’ as is generally accepted.
Studies in non-pregnant women demonstrate a strong positive association with BMI and symptoms of gastro-oesophageal reflux disease,24 However, whether obesity constitutes a risk factor for development of gastro-oesophageal disease during pregnancy is less clear.24,25 Our study clearly demonstrates that increased maternal BMI is associated with reporting of heartburn. Moreover, the strong association between maternal BMI and use of Gaviscon further supports this link.
Symphysis-pubis dysfunction is another common and frequently disabling condition during pregnancy. In a large epidemiological study, Larsen et al.,23 did not demonstrate an association between pelvic girdle relaxation and maternal body weight. However, a number of factors could account for this, in particular methodological differences. Larsen et al.23 undertook their study 10 years ago when the prevalence of obesity in the pregnant population was much lower. Moreoever, their study used a broad definition for pelvic girdle relaxation, which included pain located to the sacro-iliac joints in addition to symphysis pubis dysfunction. It is therefore possible that women with sciatica may have been included under this broad definition. In our study, there was no association between maternal BMI and reporting of sciatica.
Our data is likely to be representative of all deliveries in Edinburgh as we collected data over 2 months in total with information gathered at least a month apart. Almost half of the women studied were over-weight or obese, highlighting the increasing epidemic of obesity in the UK among women of childbearing age.1,2 A limitation of our study is that the information was gathered retrospectively and the complications were obtained from the women’s antenatal notes following their delivery. However, this is only likely to lead to under-reporting of these minor complications during pregnancy.
In our study, it is clear that more women in the higher BMI categories experienced minor complications of pregnancy and this was associated with significant additional costs for the NHS. Our findings are strongly supported by a recent study by Chu et al. (2008),26 which demonstrated that in the absence of major complications of pregnancy such as diabetes and hypertension, maternal obesity was still associated with increased contact with health professionals and use of medications. It is likely that many of these contacts and drug prescriptions are related to the increased risk of minor complications during pregnancy associated with maternal obesity.
Although, using robust methodology, we have described the additional costs associated with minor complications, it must be kept in mind that many women may incur a number of the costs enumerated and not simply one of the complications in isolation. Thus, we may have underestimated the additional costs associated with the minor complications of pregnancy. In addition, we have not included other costs, for example costs of induction of labour, and the numerous and large costs associated with the major complications of pregnancy such as gestational diabetes and pre-eclampsia.
In conclusion, increased maternal BMI is associated with an increased incidence of minor complications during pregnancy and increased use of the medications associated with treating these conditions. This may impact upon maternal wellbeing during pregnancy. Moreover, the additional NHS costs associated with treating minor complication in obese women may have a wider economical impact as a result of increased health service usage and drug costs. Encouraging weight reduction and management strategies before, during and after pregnancy, may therefore beneficial, not only for the individual but also for the wider society.
Disclosure of interests
The authors have no conflicts of interest.
Contribution to authorship
FD designed the study, wrote the manuscript and contributed to health economic evaluation and data analysis. GN analysed the data and approved the final manuscript. BG carried out the health economic evaluation and approved the final manuscript. JL and NH collected the data and approved the final manuscript. RR designed the study, contributed to health economic evaluation and data analysis and wrote the manuscript.
Details of ethics approval
Ethical approval was obtained from the local Lothian Research Ethics Committee (Reference 07/S1103/25).
The study was unfunded.
We thank the labour ward midwives and Tommy’s, the Baby Charity.