Background
The latest European Perinatal Health Report identified 22.6% of German women as being overweight (body mass index (BMI) 25.0–29.9 kg/m
2) and 13.7% as obese (BMI ≥ 30.0 kg/m
2) at the onset of pregnancy [
1]. Maternal overweight and obesity can affect the course of pregnancy, as well as delivery and the postpartum health of both mothers and their infants [
2,
3]. In addition to a high pre-pregnancy BMI, excessive gestational weight gain (GWG) is an increasing public health concern due to its potential contribution to pregnancy and obstetric complications, maternal postpartum weight retention and childhood obesity. With reference to the recommendations for adequate weight gain during pregnancy provided by the United States’ Institute of Medicine (IOM) [
4], there is a significant trend towards excessive GWG [
5]. In Germany, more than 40% of pregnant women currently exceed the recommended weight gain thresholds [
6].
Pregnant women who gain weight excessively are more likely to develop gestational diabetes mellitus (GDM) [
7,
8] and to retain weight in the postpartum period [
9]. These risks not only affect overweight women or those with obesity, but also those entering pregnancy with a normal BMI [
8,
10,
11]. Further, high maternal weight gain in pregnancy has been found to increase the risk of high foetal birth weight and obstetric complications [
12]. In addition, high GWG has been shown to raise the risk of childhood overweight and obesity [
13,
14], especially in infants born to women with a normal pre-pregnancy BMI [
15], which can persist later in life [
16,
17].
Over the last decade, a variety of lifestyle intervention studies during pregnancy have tried to limit GWG and to improve maternal and infant health. However, these randomised controlled trials have shown rather modest effects in reducing excessive GWG and its associated health outcomes [
18‐
21]. Nevertheless, several studies involving diet and/or physical activity counselling suggest effects on GWG, and a recent meta-analysis showed a mean reduction of GWG by 0.7 kg due to lifestyle interventions [
22]. Yet, many lifestyle intervention trials have focused only on overweight and/or obese women, and trials recruiting women across the entire BMI range do not always explicitly evaluate the effectiveness in normal-weight women [
23]. Further, only a limited number of small studies have tried to integrate lifestyle programmes into routine antenatal care outside of academic settings [
24‐
26]. Thus, there remains a clear need to develop effective and efficient ‘real-world’ strategies limiting GWG to appropriate levels in women with normal weight, as well as in those overweight and with obesity.
We recently conducted the FeLIPO pilot trial (“Feasibility of a lifestyle intervention in pregnancy to optimize maternal weight development”) in order to evaluate the potential to prevent excessive GWG within the setting of routine prenatal care [
27]. The intervention programme, which focused on a balanced diet, physical activity and self-monitoring of GWG, was delivered by an experienced dietician and was effective in reducing the proportion of women with excessive GWG according to the IOM recommendations (38% vs. 60%,
p = 0.032) [
27]. Based on the results of this pilot trial, the GeliS (“Gesund leben in der Schwangerschaft”/Healthy living in pregnancy) trial was designed to embody a true public health approach, as it was performed within the framework of the well-established German pre- and postnatal care system used by almost every pregnant woman in the country. The intervention offered comprehensive counselling on a healthy perinatal lifestyle at four defined visits and aimed to prevent excessive GWG and associated maternal and infant health outcomes.
Discussion
The results of the GeliS study suggest that providing lifestyle advice addressing diet, physical activity and weight monitoring within routine care during pregnancy is not effective in avoiding excessive GWG or in reducing total GWG. Except for a slight decrease in birth weight and length, the antenatal intervention did not notably affect the risk of developing GDM or any other maternal or foetal outcome.
The GeliS trial was performed as a public health approach in the real-life setting of routine care. As researchers have previously suggested [
37,
38], study procedures were adapted to the daily routine work of gynaecological and midwifery practices in order to allow an easy implementation of the intervention into the German maternity healthcare system. To this end, lifestyle counselling was conducted by previously trained medical personnel.
To our knowledge, there is no other trial to date that has investigated the compatibility of an additional lifestyle advisory component within routine healthcare for pregnant women to such a large extent. Overall, 45.1% of women who received the intervention exceeded the IOM recommendations, compared to 45.7% of women who received standard antenatal care. The lack of an intervention effect is consistent with current research. Despite an intensive intervention programme of eight group sessions, the UPBEAT trial showed only small effects on GWG in pregnant women with obesity (– 0.55 kg) [
21]. A large meta-analysis of lifestyle intervention trials including individual participant data of more than 12,000 pregnant women reached similar conclusions [
22]. According to the i-WIP collaborative group [
22], lifestyle interventions are able to reduce GWG, but the effect was quite small (– 0.7 kg), and 37% of women still exceeded the weight gain recommendations. Standard care differs between countries and, thus, may influence the effectiveness of different interventions [
39]. Moreover, the effect of lifestyle interventions in different groups of women based on BMI category, age, ethnicity, parity and risk status in pregnancy is not clear [
37], which complicates the comparability between studies. Furthermore, the assessment of excessive GWG according to the IOM recommendations is disputed as evidence to support these guidelines and has been suggested as insufficient [
40]. Especially in overweight women and those with obesity, weight gain within the IOM recommendations has been associated with both positive and adverse pregnancy outcomes [
40]. Nevertheless, the IOM guidelines represent the current standard and are frequently applied [
41].
As more than 40% of pregnant women in Germany and elsewhere exceed these recommendations, resulting in potentially adverse short- and long-term consequences for mothers and infants, there is an urgent demand for successful interventions [
4,
6,
14,
42]. Additionally, pregnant women themselves are demanding lifestyle counselling, as determined by the high compliance to the GeliS intervention programme in terms of attendance to the scheduled sessions. More than 85% of women in the intervention group attended all four counselling appointments, showing their willingness to adhere to a healthy lifestyle. Together with high rates of excessive GWG, this emphasizes the gap between the current standard of prenatal care and the need for information and support.
Among the secondary outcome parameters, GDM was diagnosed with a 2-hour OGTT in the GeliS trial. At least one of the GDM diagnostic threshold values [
33] was exceeded by 11.0% of the tested study participants, although this was slightly lower than the overall prevalence estimate for Germany (13.2%) [
31]. As an important finding, the GeliS intervention did not lead to a reduction in GDM. However, most women had an adequate metabolic control, as assessed by the measurement of glycated haemoglobin, and only a minority required active treatment. This result is in line with observations from recent meta-analyses and reviews [
22,
43,
44].
Despite most maternal and neonatal outcomes being unaffected by the intervention, a few differences were observed. Elective caesarean sections were more frequently reported in the intervention group. As mode of delivery was not addressed during lifestyle counselling, a specific effect of the intervention seems to be unlikely. Due to the cluster randomisation, one possible explanation could be differences in procedures between hospitals in performing caesarean sections, which is underpinned by a high variance in the caesarean section rate between regions [
45]. Further, labour had to be induced more often in the control group than in the intervention group. Labour is induced frequently, especially after the estimated due date, but as gestational week at birth was comparable between groups, the difference in the rate of labour induction could again be related to clinic-specific procedures. Differences in hypertension cannot be attributed to the intervention programme since blood pressure measurement procedures may differ between the single practices. Another difference between groups was the proportion of women with preterm labour, which was slightly higher in the control group. This parameter was evaluated as a safety control in order to ensure that encouraging women to engage in physical activity during pregnancy would not lead to premature contractions. Thus, the finding that the proportion of preterm labour was even lower in the intervention group supports the safety of the physical activity component.
The intervention resulted in significant trends towards a lower birth weight and lower birth length in the intervention group. However, the estimated differences between groups were small. Overall, there is insufficient evidence for an effect of lifestyle interventions in pregnancy on neonatal birth weight [
23]. The observed difference in birth length in the intervention group may explain the difference in birth weight. Nevertheless, there were no significant differences regarding infants born LGA, in line with the results of the LIMIT and UPBEAT trials, which also showed no effect of lifestyle advice on the number of infants born LGA [
19,
21]. Similarly, the i-WIP consortium did not report any significant effects for neonates including birth weight [
22].
Despite the lack of observation of major effects from the GeliS intervention on GWG or pregnancy complications, the trial has several strengths. A major advantage is that counselling sessions could be scheduled in combination with prenatal visits, resulting in both high participation and low drop-out rates. The drop-out rate of 11% was lower than the expected rate of up to 20%, reflecting the applicability of the programme and, indirectly, the interest of women to participate in the lifestyle programme; similar observations have been reported [
46]. Moreover, the cluster-randomised design counteracted the spill-over effects of lifestyle counselling content from women in the intervention group to those in the control group, which is a further strength of the GeliS trial. Additionally, compared to available national data, the characteristics of women participating in the GeliS study were representative of the target population with respect to age, pre-pregnancy BMI and smoking status.
However, there are a few limitations worth noting. Women participating in the GeliS trial were predominantly white and relatively well educated, with only a small proportion from ethnic minorities. Therefore, the results may not be completely applicable to the general population of Germany. Moreover, the counselling was not extensively based on concepts of behaviour change, with lifestyle counselling including methods such as self-monitoring and feedback on behaviour. An extension with additional methods such as motivational interviewing was not possible. As discussed by others, intervention programmes are often initiated too late [
20,
38]. The importance of implementing an antenatal intervention early on should be stressed since many women gain much weight in early pregnancy and high early GWG is strongly predictive not only of total excessive GWG [
47], but also of GDM [
8]. In accordance with these findings, GeliS counsellors were encouraged to include expectant mothers as early as possible.
It is noteworthy that the large majority of previous lifestyle intervention studies in pregnant women has been performed by trained experts within academic study centres, which poorly reflect real-life settings. Controlled studies involving exercise groups, objective monitoring of physical activity and continuous observation of dietary behaviour may be more likely to prevent excessive GWG. However, there is an urgent need to develop and evaluate effective strategies in real-life settings of routine prenatal care in order to be applicable at the population level. Apart from our pilot trial, of the two identified studies integrating lifestyle interventions into prenatal care [
24,
25], only one was effective in reducing GWG [
25]. Although this study was integrated into routine care, counselling was performed by a qualified expert instead of trained medical personnel as in the present study. This is in line with the results of our pilot trial, supporting the effectiveness of lifestyle counselling within routine care when performed by an expert [
27]. Unfortunately, the promising results observed in our pilot study FeLIPO (excessive GWG IV: 38.2% vs. C: 59.5%) could not be confirmed through the GeliS intervention, despite the comparability of the GeliS trial in terms of trial setting and counselling content. However, in the FeLIPO trial, counselling was provided by a dietician, while trained medical personnel delivered the sessions in the GeliS study, which may have substantially contributed to the discrepancy of results.
Although counselling sessions in the GeliS trial were conducted according to a predefined curriculum, differences in the quality of the delivered intervention are the most likely explanation for the lack of effect on GWG. Due to the specific characteristics of the study designed as a public health approach, it was not possible to extensively monitor whether the sessions were consistently performed by the counsellors as planned, with only a sample of sessions being supervised by a member of the study team. Indeed, the process evaluation of this sample showed inconsistencies in the delivery of counselling sessions. Not every lifestyle counsellor addressed all planned components of the intervention. In particular, individual feedback based on personal dietary and physical activity habits was not consistently given. To date, the prenatal care provided by gynaecologists and midwives does not include lifestyle advice, and mainly focuses on foetal growth parameters and maternal and foetal complications. Even though counsellors received specific training prior to the intervention sessions and reported feeling adequately trained, a 2-day seminar may not be enough to qualify gynaecologists, medical assistants and midwives as specialists with sufficient expertise for high-quality lifestyle coaching.
A further contributor to the missing effect could be the general issue of scale-up of intervention studies following a promising pilot phase [
24,
48,
49]. Implementing successful interventions in health care systems remains a challenge [
48,
49]. Finally, the lack of statistical differences in study outcomes between the intervention and control groups may be attributed to increased awareness of a healthy lifestyle and behaviour change during pregnancy among women in the control group following reading of the study material and questionnaires on dietary behaviour and physical activity.
Implementing a lifestyle programme into daily work and combining counselling sessions with routine care visits remains a challenge. Therefore, collaboration between medical practices and specially trained and experienced dieticians or lifestyle coaches could lead to an improvement in the quality of lifestyle counselling, and may reduce the proportion of women with excessive GWG and associated health consequences. Additionally, ensuring an appropriate environment, such as a separate room designated for counselling patients, rather than conducting sessions in a busy practice office, could contribute to increased quality of the lifestyle intervention. Finally, a more extensive intervention programme comprising more than four counselling sessions may be more successful in promoting significant lifestyle changes.
Acknowledgements
Project managers at the expert centres for nutrition/community catering at the regional offices (AELF) of the Bavarian State Ministry of Food, Agriculture and Forestry (StMELF) have been coordinating the study in the 10 Bavarian regions. We gratefully acknowledge the valuable contribution from the Munich Study Centre at the Technical University of Munich and the network Healthy Start - Young Family Network, Federal Center for Nutrition (BZfE), Federal Office for Agriculture and Food (BLE), belonging to the national IN FORM initiative (Maria Flothkötter, Katharina Krüger), Bonn. The support from Dr Uta Engels, Sports Centre, University of Regensburg; Prof Karl-Heinz Ladwig, Head of Research Group Mental Health at the Institute of Epidemiology, Helmholtz Centre Munich; Prof KTM Schneider, Division of Obstetrics and Perinatal Medicine, Technical University of Munich; Prof Rüdiger von Kries, Institute of Social Paediatrics and Adolescent Medicine, Ludwig-Maximilians-University Munich; Prof Regina Ensenauer, Research Group ‘Molecular Nutrition’, Dr von Hauner Children's Hospital, Ludwig-Maximilians-University Munich and Heinrich Heine University Düsseldorf; Prof Renate Oberhoffer, Chair of Preventative Paediatrics, Technical University of Munich; Prof Rolf Holle, Institute of Health Economics and Health Care Management, Institute of Epidemiology, Helmholtz Centre Munich; Gabi Pfeifer, Educational Center Nuremberg; and Eveline Rieg, Competence Centre for Nutrition, Freising/Kulmbach, is gratefully acknowledged. We are also indebted to the Board of Trustees: Maria Flothkötter, German Federal Ministry of Food and Agriculture (BZfE) in the Federal Office for Agriculture and Food (BLE), Bonn; Dr Beatrix Heilig and Dr Martina Enke, Bavarian State Ministry of Health and Care; Marion Kratzmair and Dr Wolfram Schaecke, Bavarian State Ministry of Food, Agriculture and Forestry; Rainer Prischenk, Competence Centre for Nutrition, Freising/Kulmbach; and Dr Annette Scheder and Katharina Leopold, AOK Bayern. We would also like to thank the Bayerische Landesärztekammer, Bayerischer Hebammen Landesverband e.V. for providing support by announcing the study in the beginning in their journals and newsletters, thereby helping us to get attention and to recruit practices (thus creating some trust for the study team) – though they had no direct involvement in the recruitment – and the Company Beurer GmbH, Ulm, for providing 1500 wearable pedometers for free. Finally, we want to thank all participating practices, gynaecologists, medical assistants, midwives, pregnant women and their families for their involvement.