Included studies

Thirty-nine studies were included for this review:

• 9 prospective cohort studies (Breckenkamp 2019, Cedergren 2006, Chen CN 2020, Gaillard 2013, Haugen 2014, Lautredou 2022, Morken 2013, Premru-Srsen 2019, and Tanigawa 2022)
• 26 retrospective cohort studies (Beaudrot 2016, Beyerlein 2011, Blomberg 2011, Chen L 2020, Chen-Xu 2022, Chuang 2022, Di Benedetto 2012, Enomoto 2016, Flick 2010, Gante 2015, Gavard 2017, Gawade 2011, Graham 2014, Haile 2019, Harper 2011, Hautier 2022, Hung 2016, Kiefer 2022, Kominiarek 2013, Langford 2011, Liang 2021, McCurdy 2022, Nohr 2008, Park 2011, Simko 2019, and Yee 2013)
• 2 systematic reviews of observational studies (Lipworth 2022 and Whitaker 2022) and
• 2 individual patient data (IPD) meta-analyses (Santos 2019 and Voerman 2019).

The included studies are summarised in Table 2.

Two studies were conducted on twin pregnancies (Lipworth 2022 and Whitaker 2022), all other studies were conducted on single pregnancies.

Twenty-nine studies used the Institute of Medicine (IOM) gestational weight gain guidelines (Beaudrot 2016, Beyerlein 2011, Blomberg 2011, Breckenkamp 2019, Chen CN 2020, Chen-Xu 2022, Di Benedetto 2012, Enomoto 2016, Flick 2010, Gaillard 2013, Gante 2015, Gavard 2017, Haile 2019, Harper 2011, Haugen 2014, Hautier 2022, Hung 2016, Kiefer 2022, Langford 2011, Lautredou 2022, Liang 2021, Lipworth 2022, McCurdy 2022, Park 2011, Santos 2019, Simko 2019, Tanigawa 2022, Voerman 2019, Whitaker 2022), 1 study used a combination of guidelines (Chuang 2022), and 9 studies used independent thresholds (Cedergren 2006, Chen L 2020, Gawade 2011, Graham 2014, Kominiarek 2013, Morken 2013, Nohr 2008, Premru-Srsen 2019, Yee 2013).

Thirty studies assessed the association between gestational weight gain and caesarean birth (Beaudrot 2016, Beyerlein, 2011, Blomberg 2011, Breckenkamp 2019, Cedergren 2006, Chen CN 2020, Chen-Xu 2022, Di Benedetto 2012, Enomoto 2016, Flick 2010, Gaillard 2013, Gante 2015, Gavard 2017, Gawade 2011, Graham 2014, Haile 2019, Harper 2011, Haugen 2014, Hung 2016, Kiefer 2022, Kominiarek 2013, Langford 2011, Lautredou 2022, Lipworth 2022, McCurdy 2022, Morken 2013, Nohr 2008, Simko 2019, Whitaker 2022, Yee 2013). 8 studies assessed the association between gestational weight gain and preeclampsia (Chen CN 2020, Chen L 2020, Chen-Xu 2022, Hung 2016, Lipworth 2022, Premru-Srsen 2019, Santos 2019, Whitaker 2022). 6 studies assessed the association between gestational weight gain and gestational hypertension (Chen CN 2020, Chen-Xu 2022, Enomoto 2016, Lautredou 2022, Santos 2019, Whitaker 2022). 9 studies assessed the association between gestational weight gain and gestational diabetes (Chen CN 2020, Chen-Xu 2022, Chuang 2022, Enomoto 2016, Hung 2016, Lautredou 2022, Lipworth 2022, Santos 2019, Whitaker 2022).

Eight studies assessed the association between gestational weight gain and small for gestational age (Chen-Xu 2022, Enomoto 2016, Hautier 2022, Hung 2016, Lipworth 2022, Park 2011, Santos 2019, Whitaker 2022).10 studies assessed the association between gestational weight gain and large for gestational age (Chen CN 2020, Chen-Xu 2022, Di Benedetto 2012, Enomoto 2016, Hung 2016, Lautredou 2022, Liang 2021, Park 2011, Santos 2019, Whitaker 2022). 2 studies assessed the association between gestational weight gain and childhood overweight or obesity (Tanigawa 2022, Voerman 2019).

All studies adjusted for covariates; however, the committee did not agree on key covariates a priori in the protocol. The adjusted covariates varied across the studies.

All studies used the World Health Organization definition of BMI categories when reporting pre-pregnancy BMI.

The studies used different definitions of gestational weight change (for example, IOM categories of weight change, Swedish categories of weight change) and different methods of weight measurement (for example, self-reported, by healthcare professional at antenatal care appointments). Overall, most studies used the IOM categories for gestational weight change. Studies defined gestational weight change as weight loss, inadequate weight gain, adequate weight gain, or excessive weight gain. Only studies using the same gestational weight change categories and definitions were pooled. Where possible, similarly defined outcomes have been pooled. Since all studies reported results by inadequate or excessive gestational weight change, studies have not been analysed by when gestational weight was measured (for example, gestational weight change at different time points such as 1st, 2nd or 3rd trimester). All studies had a referent population (usually adequate weight gain), but there was some variation in studies (for example, specific weight gain categories such as weight gain of 15-24.9 lbs). Only studies using the same referent population were pooled.

Evidence was identified for all outcomes other than health related quality of life.

See the literature search strategy in appendix B and study selection flow chart in appendix C.

The evidence was stratified by BMI and parity (nulliparous and parous) where possible. No evidence was identified to stratify evidence according to ethnicity and bariatric surgery.

Excluded studies

Studies not included in this review are listed, and reasons for their exclusion are provided in appendix K.

Included studies

No economic studies were identified which were applicable to this review question. See the literature search strategy in appendix B and economic study selection flow chart in appendix G.

Excluded studies

Economic studies not included in this review are listed, and reasons for their exclusion are provided in appendix J.

The outcomes that matter most

The committee agreed to prioritise the maternal outcomes of caesarean birth, hypertensive disorders of pregnancy, and gestational diabetes; fetal outcomes of SGA and LGA; and childhood outcomes of overweight/obesity as critical outcomes, and health related quality of life as an important outcome. These outcomes are in line with a core outcome set defined in the evidence base.

The quality of the evidence

The quality of the evidence was assessed with an adapted GRADE approach and the overall quality of the evidence ranged from very low to high.

The evidence was downgraded due to methodological limitations in studies (for example, issues with study participation or with prognostic factors measurement), indirectness (for example, outcome indirectness such as, including macrosomia for LGA), and imprecision around the effect estimate. Some evidence was downgraded as a result of serious or very serious heterogeneity that was unexplained by sub-group analysis.

Data were identified for all outcomes except health-related quality of life.

Benefits and harms

Overall, the committee thought that the evidence was difficult to interpret because of the different categories used to define gestational weight change and because weight change was reported as total weight gain in pregnancy instead of trimester specific weight gain in the studies reviewed. This made it difficult to give practical advice about appropriate or inappropriate weight change during different stages of pregnancy.

The committee discussed that weight change in pregnancy is a sensitive and complicated issue. They were aware that many people have negative experiences when being weighed, or when weight or BMI is being discussed in clinical settings. For example, they may feel judged if they have a higher pre-pregnancy BMI, they may experience that their other health concerns are attributed to their weight inappropriately or unnecessarily. Things can be particularly complicated if they have a history of eating disorders. Sometimes - discussions around weight may be more difficult if their partner is present. The committee agreed it was important to handle these discussions with care to avoid stigmatising the individual. The committee referred to the recommendations in the NICE guideline on overweight and obesity management about how to discuss weight in a sensitive manner. Offering to measure height and weight and calculating BMI in the first antenatal (booking) appointment is current practice and helps healthcare providers to plan care and assess for risk factors, as is recommended in the NICE guideline on antenatal care. The committee discussed that unless there are clinical reasons, sharing the weight and BMI with the person at the booking appointment is not always necessary, but if it is it should be shared in a sensitive manner or discreetly, for example so that it is not said out loud but written down instead.

The committee noted that most of the evidence identified reported results according to the IOM (now called the National Academy of Medicine, NAM and referred to as such in the recommendations) categories on gestational weight change during pregnancy, which estimates appropriate weight change ranges according to pre-pregnancy BMI groups (underweight, healthy weight, overweight, obesity). The committee also noted that the categories report total gestational weight change during pregnancy rather than trimester specific gestational weight change, which is important as weight change does not occur equally in each trimester. Although NAM provides some trimester specific guidance, this has not been rigorously validated and therefore its relevance to, and safety for, pregnancy outcomes is unclear.

The committee were aware that the IOM/NAM categories were developed by balancing the risk between small for gestational age (SGA) and large for gestational age (LGA), and the risk of caesarean section and postpartum weight retention. However, gestational diabetes and preeclampsia were not considered in the development of the categories. The committee discussed that because of the way it was developed, the IOM/NAM categories should be applied and interpreted with caution. Furthermore, the committee were aware that currently nearly half of pregnancies worldwide have a gestational weight change during pregnancy that is more than the IOM/NAM recommended weight change categories, potentially reducing relevance to the population today. The committee also noticed that currently the IOM/NAM recommended categories for people with a pre-pregnancy BMI in the obesity range are grouped together rather than by different obesity categories (for example, class I, II, or III). However, the committee agreed there are no better alternatives available at the moment. Based on the evidence and their expertise, the committee agreed that the estimates for optimal total weight change during pregnancy remain uncertain and not well defined for all pre-pregnancy BMI groups, but the IOM/NAM recommendations may provide helpful estimates for total appropriate weight gain in pregnancy and referred to these in the recommendations. However, what is particularly uncertain is what the week by week weight change in each trimester should be.

The committee agreed that the evidence does not support weighing everyone throughout pregnancy and this should only be offered when there is a clinical need (for example, if the person has gestational diabetes or need for thromboprophylaxis, or if there are nutrition concerns due to hyperemesis gravidarum). This is further discussed in evidence review G. Because of the uncertainties around optimal weight change in pregnancy, the committee agreed that the focus during pregnancy should instead be on healthy eating and physical activity which have positive associations with health benefits for the mother and the baby.

The committee discussed that not only are there uncertainties about what is healthy weight change in pregnancy, but the normal physiological changes during pregnancy contributing to the weight change, such as weight of the growing fetus, weight of the placenta, increase in maternal blood volume, breast tissue expansion, and volume of amniotic fluid can vary between individuals across all these contributing factors but especially in relation to the weight of the baby.

However, they acknowledged that some people may want to monitor their weight themselves throughout pregnancy. Quantitative evidence was unable to determine the optimal weight change during pregnancy; however, there are estimates of healthy total weight change in a singleton pregnancy according to the pre pregnancy BMI that healthcare professionals can refer to (Table 1 in the National Academy of Medicine’s report on the current understanding of gestational weight gain among women with obesity and the need for future research). But these estimates do not account for trimester specific healthy weight change. The committee noted that there are separate estimates for twin pregnancies, although there is even more uncertainty around these estimates. The committee notes that because of the wide ranges in the NAM/IOM estimates and because they estimate total weight change in the whole of pregnancy, they are not particularly helpful in practice.

The committee recognised that some people who want to monitor their weight might not have access to weighing scales but discussed that these should be available in health care settings or community hubs where people can be advised to go. The committee discussed the difficulties and risks of weight management during pregnancy and agreed that pre-conception weight management should be a priority because weight before pregnancy is potentially a modifiable factor. The committee agreed that pre-pregnancy BMI is a bigger risk factor for adverse maternal and fetal outcomes than gestational weight change during pregnancy. However, weight management before pregnancy is outside the remit of this guideline and is covered in the NICE guideline on overweight and obesity management.

Overall, the meta-analysed data and data from the individual participant data meta-analysis suggested that weight change exceeding the IOM/NAM estimated categories was associated with negative maternal and fetal outcomes. Generally, there was moderate to high quality evidence which suggested an association between excessive gestational weight gain and the outcomes caesarean birth, gestational diabetes, gestational hypertension and large for gestational age for pregnant women with a pre-pregnancy BMI within the healthy, overweight, and obesity ranges. The committee noted that although the relative data suggested these associations for risk, there was no absolute data showing that excessive gestational weight gain in a pre-pregnancy BMI of overweight or obesity ranges leads to more negative outcomes than in those with a lower pre-pregnancy BMI. Despite this, the committee agreed that maintaining a healthy pre-pregnancy BMI would lead to the most optimum outcomes for the mother and the baby.

There was high quality evidence from the individual participant data meta-analysis which showed an association for women with a pre-pregnancy BMI in the obesity range and excessive gestational weight gain for the outcomes of caesarean birth, gestational diabetes, pre-eclampsia and large for gestational age. The committee agreed that this overall trend (weight change exceeding the IOM/NAM estimated categories associated with negative maternal and fetal outcomes) resonated with their clinical and research experience, and therefore these should be discussed. Alongside these risks, it is also important to consider individual level factors that might affect development of these outcomes, for example, comorbidities, family history and obstetric history. The committee noted that there was evidence from 1 study (Enomoto 2016) conducted in Japan where those with pre-pregnancy BMI in the obesity range and who gained excessive gestational weight, showed no association for caesarean birth, gestational diabetes, and gestational hypertension. The committee discussed that because these results were of low quality and were based on Japanese women, they were less applicable to the majority of the population in England and Wales.

Moderate to high quality evidence from the meta-analysis from this review and data from the individual participant data meta-analysis included in this review suggested that weight change exceeding the IOM/NAM estimated categories in people with a healthy pre-pregnancy BMI was associated with gestational diabetes, gestational hypertension, preeclampsia, and baby being large for gestational age. Data from the individual participant data meta-analysis suggested there was a lower risk of small for gestational age babies with excessive gestational weight gain for all pre-pregnancy BMI ranges, when compared to normal gestational weight gain in those with a healthy pre-pregnancy BMI. The committee discussed that pregnant people with a healthy pre-pregnancy BMI who rapidly gain weight during pregnancy is a population that is largely missed and therefore adverse outcomes are likely to be more problematic in this population than in people with a higher pre-pregnancy BMI with excessive gestational weight change in pregnancy, who may be under closer observation. The committee were aware from their knowledge that there is not as much guidance on pregnant people with a healthy pre-pregnancy BMI as it is assumed that this population is at a lower risk of negative maternal and fetal outcomes.

Overall, the committee concluded from the evidence that those with pre-pregnancy BMI in the healthy, overweight and obesity ranges who gain excessive weight during pregnancy are at an increased risk of some obstetric complications, including developing gestational hypertension or gestational diabetes, baby being large for gestational age, or needing a caesarean section. The committee noted that large for gestational age is frequently an indicator for caesarean birth, and therefore this outcome is not independent of caesarean birth.

The committee acknowledged that because the evidence does not justify routine weight monitoring in pregnancy, excess weight gain is also not routinely identified. They agreed that if there are concerns about excessive gestational weight gain during pregnancy raised by the pregnant person, or by a healthcare professional as part of weight monitoring for a clinical reason, this should be explored further, and information about healthy eating and physical activity should be discussed. The committee agreed that there may be various underlying reasons for excessive weight gain during pregnancy, and a holistic exploration of the person’s wellbeing is important. Based on the evidence of excessive weight gain being associated with baby being born large for gestational age and with gestational diabetes, the committee recommended that the baby should be monitored for large for gestational age and a test for gestational diabetes should be considered. It should be noted that those with a BMI of 30 or more at the booking appointment should already be offered an oral glucose tolerance test according to the NICE guideline on diabetes in pregnancy, but the committee agreed a referral to the test should be considered regardless of the pre-pregnancy BMI if there are concerns about excessive gestational weight gain.

In terms of too low weight gain during pregnancy, there was high quality evidence from the individual participant data meta-analysis, which showed an increased risk of baby being small for gestational age in those with pre-pregnancy BMI in the underweight and healthy ranges who had low or inadequate weight gain during pregnancy. Moderate quality evidence from other meta-analysed studies also showed an association between inadequate weight gain and baby being small for gestational age among those with pre-pregnancy BMI in the underweight range. Low quality evidence across all pre-pregnancy BMI ranges also showed an association with inadequate weight gain and baby being small for gestational age.

There was low to moderate quality evidence from the individual participant data meta-analysis among those with a pre-pregnancy BMI in the underweight or healthy weight ranges and low gestational weight gain that showed no association of risk for gestational diabetes, however, low to moderate quality evidence from other meta-analysed studies suggested that weight change below the IOM/NAM estimated categories in those with a pre-pregnancy BMI in the underweight or healthy weight ranges was associated with a high risk of gestational diabetes. The committee discussed that this association might be due to participants in this study being diagnosed with gestational diabetes during their second trimester of pregnancy, therefore receiving dietary and exercise advice to control their gestational weight change during the third trimester. It is likely that this would have strongly contributed to the overall gestational weight change as most weight change occurs over this period. The committee noted that the studies reported weight change as the difference between weight prior to birth and pre-pregnancy body weight, and therefore there is uncertainty around how much weight changed by trimester and the causal relationship between low gestational weight change and gestational diabetes.

From their collective experience, the committee agreed that there may be various reasons for low weight gain during pregnancy, for example, nausea and vomiting in pregnancy, mental health issues, or clinical interventions that have been recommended for the person, for example, after a diagnosis of gestational diabetes.

As with excessive gestational weight gain, the committee acknowledged that because the evidence does not justify routine weight monitoring in pregnancy, low weight gain is also not routinely identified. The committee agreed that when there are concerns about low weight gain during pregnancy raised by the pregnant person, or by a healthcare professional as part of weight monitoring for a clinical reason, this should be further explored through individualised discussions about healthy eating and physical activity in pregnancy because advice required for those with low gestational weight gain will likely be different to advice needed for those with excessive gestational weight gain. The committee also agreed that routine monitoring of the growth of the baby (according to the NICE guideline on antenatal care) to check for small for gestational age should be ensured when there are concerns about low weight gain during pregnancy. The committee discussed that for some people, a lifestyle change of healthier eating and increasing physical activity might lead to weight loss. The committee considered it was important to discuss the potential impact of overly restrictive eating and drinking habits.

Overall, the results suggest that gestational weight gain exceeding the IOM/NAM estimated categories is associated with an increased risk of overweight and obesity in childhood. There was high quality evidence from individual participant data meta-analysis for all pre-pregnancy BMI categories and excessive gestational weight gain that showed a high risk of childhood overweight or obesity at 2 to 18 years of age, and with low gestational weight gain showed a low risk of childhood overweight or obesity at 2 to 10 years, and no association at 10 to 18 years. High and moderate quality evidence from 1 study from Japan supported these findings and also found that people within the obesity pre-pregnancy BMI range with low gestational weight gain also had a high risk for childhood overweight or obesity at 3 years. The committee discussed that gestational weight gain is not only associated with maternal outcomes but is also associated with childhood outcomes. The committee agreed that pre-pregnancy BMI is an important modifiable risk factor of childhood weight status, which is currently a significant public health concern. To support people with weight management, the committee referred to the NICE guideline on overweight and obesity management. The committee concluded that supporting healthy eating and drinking, physical activity and weight management in pregnancy can have positive outcomes for the pregnancy and birth but also later on for the child.

Although there was some moderate quality evidence suggesting an association of low risk for weight loss in pre-pregnancy BMI obesity class I and obesity class II for caesarean birth, the committee discussed that they did not want to recommend weight loss during pregnancy, especially in the first trimester, because it was unclear from the evidence whether the weight loss was intentional or unintentional and there was no evidence of benefit or harm identified for other maternal, neonatal, or fetal outcomes. The committee agreed that people may sometimes lose weight during pregnancy unintentionally, for example, if they change their eating or physical activity patterns. However, the committee agreed that intentional weight loss during pregnancy should be advised against because of potential adverse effects for the unborn baby.

The committee noted that most of the evidence included in this review was based on Caucasian populations and there was not enough evidence to stratify according to ethnicity, so the committee agreed there was insufficient evidence to make recommendations for different ethnic groups. Additionally, the evidence stratified participants using the standard BMI classification, which is known to not be appropriate for all populations. NHS and NICE recommend different BMI thresholds for people from South Asian, Chinese, other Asian, Middle Eastern, Black African or African–Caribbean background because their cardiometabolic risk occurs at a lower BMI level.

The committee discussed the limited evidence on twin pregnancies and acknowledged that there are provisional IOM/NAM estimated weight change recommendations for this population. The committee discussed that gestational weight change in twin pregnancies is largely due to the weight of the fetuses. However, the committee were aware from their knowledge that these recommendations have very little evidence base and are seldom used, and therefore, they agreed not to make recommendations on this population.

Cost effectiveness and resource use

Providing personalised advice during routine antenatal appointments to people who are interested in monitoring their weight change during pregnancy, or if there are concerns about low or excessive weight gain during pregnancy may have important resource implications, comprising health professionals’ additional time spent to offer this advice; the committee also expressed the view that time pressures may become a barrier to implementation. Offering to measure the person’s height and weight and calculate BMI at the first face-to-face antenatal appointment may also have some resource implications regarding health professionals’ time, in settings where this is not current practice, although it reflects current advice (existing NICE recommendation). Discussing the option for referral to a specialist obesity service or a specialist practitioner for anyone with a BMI of over 40 at the booking appointment may also have important resource implications. However, the benefits and cost-savings resulting from the above recommendations are expected to offset, at least partially, the costs of implementing them, by reducing the risk of conditions such as gestational diabetes, pregnancy-related hypertension and preeclampsia for the pregnant person, as well as baby being small or large for gestational age, which have negative implications for the health of pregnant people and their babies and, in turn, increase antenatal and postnatal care costs. Considering a test for gestational diabetes following concerns about excessive weight gain during pregnancy, as well as ensuring routine monitoring of the baby to check if they are potentially small or large for their gestational age if there are concerns about low or excess weight gain during pregnancy, respectively, may also have important resource implications, comprising mainly the ultrasound evaluation of the baby’s size. However, these are likely to be offset, at least partially, by prevention or more timely management of negative outcomes for the pregnant person (such as gestational diabetes) and the baby (such as being small or large for gestational age) further down the care pathway. Offering testing for gestational diabetes to anyone with a BMI of over 30 at the booking appointment is in line with current clinical recommendations and therefore should not lead to a change in practice and resource implications.

Other factors the committee took into account

For this review question, the population in the evidence was women and no evidence was identified or reviewed for trans men or non-binary people. The protocol and literature searches were not designed to specifically look for evidence on trans men or non-binary people but they were also not excluded. However, there is a small chance evidence on them may not have been captured, if such evidence exists. In discussing the evidence, the committee considered whether the recommendations could apply to a broader population, and used gender inclusive language to promote equity, respect and effective communication with everyone. Healthcare professionals should use their clinical judgement when implementing the recommendations, taking into account each person’s circumstances, needs and preferences, and ensuring all people are treated with dignity and respect throughout their care.

Excluded clinical evidence studies

Excluded economic studies