Table 3 Summary of included studies

(Baerug et al., 2018) [43]

Recent gestational diabetes was associated with mothers stopping predominant breastfeeding earlier in a multi-ethnic population

Norway

To assess the association between GDM and breastfeeding in a multi-ethnic population

Retrospective Cohort study

823 women enrolled in the study. After removal of participants not meeting study inclusion criteria, 616 women included in the study who gave birth between May 2008 and May 2010

• Women with GDM = 190

• Women without GDM = 426

Women examined at 15- and 28-weeks’ gestation, and at 14 weeks’ postpartum

Maternal data

• Collected at the time of inclusion

• Questionnaire completed by specially trained and certified midwives

Birth data

• Collected from hospital records

Breastfeeding data

Retrospective questionnaire covering the period since birth

á incidence of mothers with a history GDM ceasing breastfeeding earlier than mothers without a history GDM

• aHR 1.33, 95% CI, 1.01—1.77, p < 0.05

Breastfeeding initiated by 99% of mother’s both with and without a history of GDM. Breastfeeding rates during the postpartum period:

• End of week 1 postpartum: 86% of mothers with a history of GDM, versus, 91% of mothers without a history of GDM (p = 0.07)

• End of week 2 postpartum: 78% of mothers with a history of GDM, versus, 88% of mothers without a history of GDM (p < 0.01)

End of week 12 postpartum: 56% of mothers with a history of GDM, versus, 67% of mothers without a history of GDM (p = 0.02)

• Heterogeneity within broad ethnic groups

• The use of plain water not included in the questionnaire, preventing the ability to assess exclusive breastfeeding according to the WHO definition

• No data on if the administration of formula milk given in the infant first week of life occurred before or after hospital discharge

• Lack of the one-hour glucose value

31.5/40 (78.75%)

(Chamberlain et al., 2017) [44]

Low rates of predominant breastfeeding in hospital after gestational diabetes, particularly among Indigenous women in Australia

Australia

To investigate rates of ‘any’ and ‘predominant’ breastfeeding in hospital among Indigenous and non-Indigenous women with and without GDM

Retrospective Cohort study

Secondary analysis of data obtained from the Queensland Perinatal Data Collection (PDC)

670 participants identified who were born from 01/01/2007 – 31/10/2010 to a mother diagnosed with GDM. After removal of participants not meeting study inclusion criteria, 618 participants included in this analysis

• Indigenous infants = 214

• Non-Indigenous infants = 404

Subgroup of 365 infants with medical record review

• Indigenous infants = 209

Non-Indigenous infants = 156

Maternal data

• The Cairns Hospital Clinical Coding System

Pregnancy, birth, and breastfeeding data

Queensland Perinatal Data Collection

â incidence of women with a history GDM predominantly breastfeeding

• OR 0.32, 95% CI 0.27–0.38, p < 0.0001

â rates of breastfeeding amongst Indigenous Australian women with a history of GDM compared with non-Indigenous women with a history of GDM

• OR 0.78, 95% CI 0.70–0.88), ≤ 0.0001)

• Reporting and coding of infant data was variable, and significant changes occurred during the study period, precluding the use of data prior to 2007

• Data only illustrates the 24 h prior to discharge

• Secondary analysis, with limited data on characteristics of women who didn’t breastfeed

34.5/40 (86.25%)

(Chertok & Sherby, 2016) [45]

Breastfeeding self-efficacy of women with and without gestational diabetes

Israel

To identify factors that may contribute to maternal self-confidence in breastfeeding based on diabetes status

Subset of larger prospective case control study

Potential participants were identified by medical staff on a postnatal ward. Recruitment was attended by the research team. Participants included 67 women who gave birth in an Israeli hospital between January and June 2014

• Women with GDM = 32

• Women without GDM = 35

Maternal and breastfeeding data

• Surveys administered using the Breastfeeding Self-Efficacy Scale-Short Form (BSES-SF)

Infant, birth, and glucose data

Collected as per hospital protocol

Infant, birth and glucose data was collected as per hospital protocol

á proportion of women without a history of GDM breastfed early in the first half hour following birth, compared to women with a history of GDM (p = 0.016)

• Mothers with a history of GDM: 29%

• Mothers without a history of GDM: 58.8%

á proportion of women with a history of GDM reported perceived delayed lactogenesis II compared with women without a history of GDM

Fischer’s exact test, p = 0.029

• Lack of consistency among healthcare providers in following hospital protocol about providing formula milk supplementation to babies born to women with GDM

• Small sample size

• No operational definitions for breastfeeding

27.5/40 (68.75%)

(Cordero et al., 2013) [46]

Breastfeeding initiation in women with gestational diabetes mellitus

United States

To examine feeding practices and factors associated with BFI in women with GDM and their infants

Retrospective cohort study

Participants included 303 women who gave birth between 2008 and 2011 who had GDM during their pregnancy

• 176 women with GDM treated with diet

• 127 women with GDM treated with diet and insulin or Glyburide®

Maternal and clinical data

• Electronic medical records

Women’s feeding preference

• Recorded by Nurse on arrival to labour and delivery ward

Infant feeding data

Hard copy and electronic neonatal medical records

0% of women with a history of GDM, who intended to formula milk feed, initiated breastfeeding

27% of women with a history of GDM, with undecided feeding preferences, initiated breastfeeding

54% of women, regardless of feeding preference, initiated breastfeeding

Breastfeeding initiation rates were similar amongst women with a history GDM regardless of their GDM treatment method

• OR 9.87 CI 4.96–19.64

á breastfeeding initiation for mother with a history of GDM associated with:

• Intention to breastfeed: OR 9.87 CI 4.96–19.64

• Higher education: OR 4.20 CI 2.33–7.56

â breastfeeding initiation for mothers with a history of GDM associated with:

• African American race: OR 0.40 CI 0.22–0.73

• Smoking: OR 0.26 CI 0.14–0.49

• Obesity: OR 0.39 CI 0.22–0.68

• Admission to the neonatal intensive care unit:

81% of infants born to mother with a history of GDM, receiving well baby care post birth whose mothers intended to breastfeed, initiated breastfeeding

61% of infants born to mothers with a history of GDM, who were admitted to the neonatal intensive care, unit whose mothers intended to breastfeed, initiated breastfeeding

• Limitations inherent to retrospective studies

• Minimal follow up regarding breastfeeding practices

• Unable to ascertain how far in advance mothers’ intention to or not to breastfeed was made

• No operational definitions for breastfeeding

28/40 (70%)

(Doughty et al., 2018) [47]

Barriers to exclusive breastfeeding among women with gestational diabetes mellitus in the United States

United States

Identify differences in breastfeeding related knowledge, attitudes, beliefs, and experiences between women with GDM and women without GDM

Cross-sectional and prospective cohort study

Subset of the Infant Feeding Practices Study II (IFPS II) conducted between 2005 and 2007

4902 women enrolled. 3033 completed the first postnatal questionnaire. After removal of participants not meeting study inclusion criteria, 1733 mother infant dyads included in this cohort

• Women with GDM = 195

• Women without GDM = 2815

Mailed questionnaires

• 1 antenatal questionnaire

• 10 postnatal questionnaires, administered at monthly intervals following birth

Phone interview

Around the time of infant’s birth

â likelihood of mothers with a history GDM saying breastfeeding is the best way to feed an infant compared to mothers without a history GDM

• aOR = 0.62, 95% CI (0.46, 0.85)

á likelihood of mother with a history GDM stating the fathers of their infants prefer formula milk feeding or mixed feeding compared to mothers without a history of GDM

• aOR = 1.74, 95% CI (1.02, 2.97)

á likelihood of mothers with a history of GDM to state their physician prefers formula milk compared to mothers without a history of GDM

• aOR = 2.82, 95% CI (1.17, 6.79)

â likelihood of mother with a history of GDM to report feeling comfortable breastfeeding Infront of female friends compared to mothers without a history of GDM

• aOR = 0.70, 95% CI (0.50, 0.98)

â likelihood of newborns born to mothers with a history of GDM to stay in their mothers’ hospital room compared to infants born to mothers without a history of GDM

aOR = 0.55, 95% CI (0.36, 0.85)

• IFPS II data collected 10 to 12 years before study published

• Possibility of misclassification of the exposure to a diagnosis of GDM. Diagnosis of GDM self-reported by mothers

• Data on disease severity and treatment not available for this cohort

• No operational definitions for breastfeeding

33.5/40 (83.75%)

(Griffin et al., 2021) [48]

Lactation consultation by an International Board-Certified Lactation Consultant improves breastfeeding rates for mother with gestational diabetes mellitus

United States

To determine if a postpartum IBCLC consultation during delivery hospitalisation improved any or exclusive breastfeeding rates at hospital discharge and 3 months postpartum in women with GDM

and

To determine if obstetrical providers’ acknowledgement of maternal feeding preference affect the rates of IBCLC consultation for patients

Retrospective, comparative, secondary analysis of a prospective cohort

Research staff approached women to discuss participation between January 2016 and December 2016

600 women enrolled in primary study. After removal of participants not meeting study inclusion criteria, 517 women with GDM were included in this secondary analysis

Maternal and in hospital breastfeeding data

• Participant completed baseline surveys in either English or Spanish

• Electronic medical records

• Chart review to collect data regarding in-patient IBCLC consultation

3 months postpartum data

• Scripted telephone survey

á likelihood of mothers with a history of GDM reporting any breastfeeding at postpartum discharge amongst mothers who received IBCLC consultation compare with mothers who did not

• aOR 4.87, 95% CI (2.67, 8.86)

á likelihood of mothers with a history of GDM reporting any breastfeeding at 3 months postpartum amongst mothers who received IBCLC consultation compare with mothers who did not

• aOR 5.39, 95% CI (2.61, 11.16)

There was no difference in exclusive breastfeeding rates between mothers with a history of GDM who did and did not receive IBCLC consultation

• Data collected retrospectively as part of secondary analysis

• Data lacking regarding prior breastfeeding experience, prenatal breastfeeding education, and reasons for breastfeeding cessation and supplementation

• The reason for a mother not receiving a lactation consultation not routinely documented in medical records

• Mothers of infants with lower Apgar scored and neonatal intensive care unit admissions were less likely to receive a IBCLC consultation

• Differences in provider group practices in addressing prenatal breastfeeding education

• Differences in provider group documentation challenged the assessment of specific practices regarding prenatal breastfeeding education

• No operational definitions for breastfeeding

32/40 (80%)

(Haile et al., 2016) [49]

Association between history of gestational diabetes and exclusive breastfeeding at hospital discharge

United States

To examine the association between GDM and exclusive breastfeeding at hospital discharge

Retrospective cohort study

Subset of the Infant Feeding Practices Study II (IFPS II) conducted between 2005 and 2007

4900 participants in the IFPS II. After removal of participants not meeting study inclusion criteria, 2038 women were included in this study

• Women with GDM = 119

• Women without GDM = 1919

Mailed questionnaires

• 1 antenatal questionnaire

• 10 postnatal questionnaires, administered at monthly intervals following birth

Phone interview

Around the time of infant’s birth

GDM prevalence was 5.8%

â likelihood of women with a history of GDM exclusively breastfeeding a hospital discharge compared to women without a history of GDM (P < .01)

• Women with a history of GDM:62.2%

• Women without a history of GDM: 75.4%

â odds of exclusive breastfeeding among women with a history of GDM after adjustment for sociodemographic, behavioural, and anthropometric factors

OR = 0.59, 95% CI, 0.39–0.92

• Cross-sectional study design, precluding conclusions regarding the casual role of history of GDM in the status of exclusive breastfeeding

• Women agreed to participate in the IFPSII survey, making the results vulnerable to volunteer bias, selection bias, and recall bias

• No operational definitions for breastfeeding

28/40 (70%)

(Jagiello & Azulay Chertok, 2015) [50]

Women’s experiences with early breastfeeding after gestational diabetes

United States

To explore the lived experience of early breastfeeding for postpartum women who had GDM in pregnancy

A qualitative phenomenological research design

Potential participants identified by medical staff during hospital postpartum stay or at lactation clinic visits. Interested participants were put in contact with the research team

A purposive sample of 27 women who had been diagnosed with GDM and who initiated breastfeeding between October 2013 and January 2014

Breastfeeding experience data

• Focus groups: Audio recorded, transcribed, and noted taken by a member of the research team

• Interviews: Audio recorded, transcribed, and noted taken by a member of the research team

• Questions used as prompts to initiate conversation and to provide structure

• Session’s audio recorded

Maternal data

• Participant completed surveys

• Breastfeeding challenges and breastfeeding support

- 44% (n = 12) of women used formula to manage breastfeeding challenges

• Milk supply challenges

- Delayed lactogenesis II was reported by 41% (n = 11) of women

- Perceived decreased supply reported by 44% (n = 12) of women

• Concern of infant health

- 33% (n = 9) infants experienced complications including (15% (n = 4) with hypoglycaemia and 7% (n = 2) with jaundice)

• Small study sample size, limiting the transferability of the study findings

• No operational definitions for breastfeeding

30.5/40 (76.25%)

(Kachoria & Oza-Frank, 2014) [51]

Factors associated with breastfeeding at discharge differ by maternal diabetes type

United States

To investigate the factors associated with breastfeeding initiation in mothers with gestational and prepregnancy diabetes vs those without diabetes

Retrospective cohort study

875,988 births in Ohio between 2006 and 2011. After removal of births not meeting the study inclusion criteria, 792,730 were used in this analysis

Maternal, infant, and breastfeeding data

Birth certificates

The association of maternal and infant characteristics including maternal prepregnancy weight, maternal age, maternal race, prepregnancy care, county type, and infants gestational age on breastfeeding initiation, varied by maternal diabetes status

Overweight mothers with a history of GDM were equally likely to breastfeed compared with mothers of normal weight, with a history of GDM

• OR 0.95; 95% CI 0.87, 1.03

â likelihood of mothers with a history of GDM, from Appalachian countries, to breastfeed, compared to mothers with a history of GDM from suburban countries

• OR 0.7; 95% CI 0.7, 0.8

â likelihood of mothers with a history of GDM who receive inadequate care to breastfeed, compared to mothers without a history of GDM

• OR 0.8; 95% CI 0.7, 1.0

• Accuracy of birth certificate data not well researched in Ohio

• Method of GDM diagnosis unknown

• Limited information available regarding maternal complications of pregnancy and delivery

• Lacks specific data about breastfeeding practices

• No operational definitions for breastfeeding

34/40 (85%)

(Laine et al., 2021) [52]

Impact of gestational diabetes mellitus on the duration of breastfeeding in primiparous women: an observational cohort study

Finland

To evaluate in primiparous women whether GDM had an influence on the duration of breastfeeding, and further, to evaluate the factors that influenced on the duration of breastfeeding

Observational cohort study

Participants included 1089 women who gave birth between 2009 and 2015

• Women with GDM = 155

• Women without GDM = 934

Maternal data

• Finnish Medical Birth Register

• Statistics Finland

• Finnish Tax Administration

• Social Insurance Institution

Breastfeeding data

• Health care records

• Based on regular follow-up visits at public child welfare clinics

Infant data

Finnish Medical Birth Register

No differences observed in the duration of breastfeeding between women with a history of GDM and women without a history of GDM (p = 0.17)

• Mothers with a history of GDM: 7.5 (SD 3.7) months

• Mothers without a history of GDM: 7.9 (SD 3.5) months

á duration of breastfeeding for male infants born to mothers with a history of GDM compared to female infants born to mothers with a history of GDM (maternal age, pre-pregnancy body mass index, marital status, educational attainment, duration of pregnancy, and smoking habits adjusted p = 0.042)

• Data only available on any breastfeeding, not separately on exclusively, predominant, and partial breastfeeding

• Missing maternal data on women’s dietary and physical activity habits, as well as gestational weight gain

• All women were Finnish, therefore the generalisability of the study observations is limited

• No operational definitions for breastfeeding

31.5/40 (78.75%)

(Loewenberg Weisband et al., 2017) [53]

Hospital supplementation differentially impacts the association between breastfeeding intention and duration among women with and without gestational diabetes mellitus history

United States

To assess the associations between GDM and exclusive breastfeeding intentions, hospital supplementation, and breastfeeding duration, including whether hospital supplementation mediates the association between exclusive breastfeeding intention and breastfeeding duration

Retrospective cohort study

Subset of the Infant Feeding Practices Study II (IFPS II) conducted between 2005 and 2007

4900 participants in the IFPS II. After removal of participants not meeting study inclusion criteria, 2299 women were included in this study

• Women with GDM = 16

• Women without GDM = 2139

Mailed questionnaires

• 1 antenatal questionnaire

• 10 postnatal questionnaires, administered at monthly intervals following birth

Phone interview

Around the time of infant’s birth

â odds of intention to exclusively breastfeed associated with women with a history GDM

• AOR 0.71; 95% CI, 0.51–0.99

á odds of hospital supplementation associated with both mothers with a history of GDM, and mothers without a history of GDM, who did not intend to exclusively breastfeed

Mothers with GDM: AOR 3.52; 95% CI 1.44–8.57

Mothers without GDM: AOR 3.66; 95% CI 2.93–4.56

Mothers both with and without a history of GDM, who had exclusive breastfeeding intentions, breastfed for similar durations

• Mothers with GDM: 22.3 weeks [95% CI (16.6–28.0); p < 0.001]

Mothers without GDM: 20.7 weeks [95% CI (19.1–22.3); p < 0.001]

• IFPS II is not a nationally representative survey, limiting the generalisability of the findings

• IFPS II data 10 + years old at time of study publication. Small sample size. 160 women with GDM, of whom, 127 women had information regarding supplementation

• Intention to breastfeed assessed after GDM diagnosis, possibly influencing the woman’s intention to breastfeed

• No operational definitions for breastfeeding

35/40 (87.5%)

(Longmore et al., 2020) [54]

Associations of gestational diabetes and type 2 diabetes during pregnancy with breastfeeding at hospital discharge and up to 6 months: the PANDORA study

Australia

To evaluate the association of hyperglycaemia, including type 2 diabetes, with breastfeeding outcomes

Longitudinal cohort study

Cohort derived from 1170 participants in the Pregnancy and Neonatal Diabetes Outcomes in Remote Australia (PANDORA) cohort who gave birth between November 2011 – February 2015. After removal of participants not meeting study inclusion criteria, 1050 mother-infant dyads included in this study

• Indigenous participants = 495

• Non-Indigenous participants = 555

• Women with GDM = 684

• Women without GDM = 222

Maternal data

• PANDORA study

• Medical records

• Self-reported data

• Australian Bureau of Statistics

Infant data

• Medical records

Breastfeeding data

• Direct questions attended by research team

• Telephone or email survey attended by a member of the research team

• Electronic medical records

Among women with a history of GDM, the proportions of those exclusively breastfeeding at hospital discharge was similar among both Indigenous and non-Indigenous Australian women

• Indigenous women: 77%

• Non-Indigenous women: 75%

á likelihood of Indigenous Australian women with a history of GDM to be predominantly breastfeeding at 6 weeks postpartum, compared to non-Indigenous Australian women with a history of GDM. (p < 0.001)

• Indigenous women: 81%

• Non-Indigenous women: 57%

á likelihood of Indigenous Australian women with a history of GDM to be predominantly breastfeeding at 6 months postpartum, compared to non-Indigenous women with a history of GDM. (p < 0.001)

• Indigenous women: 68%

• Non-Indigenous women: 46%

Indigenous and non-Indigenous Australian women with a history of GDM were as likely to achieve predominant breastfeeding or exclusive breastfeeding at 6 weeks or 6 months as women without a history of GDM

• Loss of follow up at 6 weeks and 6 months may have affected the representativeness of the cohort

• Sample size was not calculated for the purpose of addressing differences in breastfeeding outcomes

• Potential for reporting bias

• Women with diabetes may be more likely to report breastfeeding especially if they received information on potential benefits

• Observational study, unmeasured confounding may influence findings

33.5/40 (83.75%)

(Morrison et al., 2015) [55]

Factors associated with early cessation of breastfeeding in women with gestational diabetes mellitus

Australia

To determine factors associated with early cessation of breastfeeding (< 3 months) in women with recent GDM

Cross-sectional online survey

Cohort derived from 15,817 women registered with the National Diabetes Service Scheme in 2010

After removal of women not meeting study inclusion criteria, invitations were sent to 5057 women. Of the women invited, 738 women consented to participate. After further removal of women who did not meet the studies inclusion criteria, 729 eligible responses were included in this study

Maternal, infant, and breastfeeding data

• Self-administered online questionnaire

97% of women with a history of GDM reported ever breastfeeding

19% of women with a history of GDM reported breastfeeding for 3 months

Cessation of breastfeeding ≤ 3 months among women with a history of GDM associated with:

• Breastfeeding problems at home: aOR 8.01, 95% CI 4.57, 14.05

• Return to work prior to three months: OR 3.39, 95% CI 1.53, 7.55

• Inadequate breastfeeding support: OR 1.88, 95% CI 1.10, 3.22

• Caesarean delivery: OR 1.70, 95% CI 1.04, 2.76

• Low socioeconomic status: (SEIFA 1 unit increase) OR 0.89, 95% CI 0.81, 0.97

• BMI (2 unit increase) OR 1.08, 95% CI 1.01, 1.57

Being married or de facto, was a protective factor against early cessation of breastfeeding for women with a history of GDM

• OR 0.14, 95% CI 0.03, 0.62

• 15% response rate. 738 women completed the survey

• Indigenous Australian women underrepresented

• Potential for bias towards women interested in breastfeeding being more likely to respond to the survey

• Women in the study were self-selected, highly educated, and differed somewhat from other Australian women with GDM, therefore, this study was unable to accurately estimate breastfeeding rates or make comparisons with GDM breastfeeding data reported elsewhere

29.5/40 (73.75%)

(Oza-Frank et al., 2016) [56]

Early lactation and infant feeding practices differ by maternal gestational diabetes history

United States

To examine breastfeeding practices through 12 months postpartum by GDM history

Retrospective Survey Design

Population derived from the Mom2Moms (M2M) study. 1244 eligible women were identified who gave birth in 2011. After removal of women not meeting study inclusion criteria, surveys were mailed to the 813 eligible women. 501 completed surveys were returned. After further removal of women who did not meet the studies inclusion criteria, the final cohort included 432 women

• Women with GDM = 34

• Women without GDM = 398

Maternal and infant data

• Maternal medical records

• Self-administered questionnaires

Maternal lactation and infant feeding data

• Self-administered questionnaire

Women with a history of GDM initiated any breastfeeding as often as women without a history of GDM

á likelihood of women with a history of GDM reporting the introduction of formula milk within the first 2 days of life, compared to women without a history of GDM (aOR 3.48; 95% CI, 1.47–8.26)

• Women with a history of GDM: 79.4%

• Women without a history of GDM: 53.8%,

Women with a history of GDM initiated pumping 4 days earlier than women without a history of GDM (P < 0.05)

No difference in the proportion of women reporting breastfeeding difficulty when comparing women with and without a history of GDM

OR 2.08; 95% CI, 0.78–5.52

• The subset of women with GDM in this sample is small and results may not be generalisable to all women with GDM history

• The survey excluded women who intended to exclusively bottle feed, indicating the sample is more representative of women who intended to engage un some breastfeeding

• Compared with non-responders, the sample was more educated, younger, had private health insurance, were non-Hispanic white, and reported fewer children

• Did not have information on level of glycaemic control among women with GDM

• No operational definitions for breastfeeding

29.5/40 (73.75%)

(Oza-Frank & Gunderson, 2017) [57]

In-hospital breastfeeding experiences among women with gestational diabetes

United States

To determine changes in the prevalence of hospital breastfeeding experiences among women with GDM and women without diabetes. And, to determine whether GDM is associated with higher occurrence of experiencing baby-friendly hospital practices because of their known higher rates of breastfeeding difficulties

Retrospective Observational Cohort Study

Cohort derived from the Pregnancy Risk Assessment Monitoring System (PRAMS) between 2004 – 2008 and 2009—2011. 163,627 possible participants were identified in the two cohorts. After removal of participants not meeting study inclusion criteria, 157,187 participants were included amongst the two cohorts

• Women with GDM = 14,409

• Women without GDM = 142,778

Maternal, infant, and breastfeeding data:

• Self-administered questionnaire

2004 – 2008 Cohort

â likelihood of mothers with a history of GDM reporting their baby stayed in the same room, compared to mothers without a history of GDM (p < 0.0001)

• Mothers with GDM: 73.6%

• Mothers without GDM: 78.3%

â likelihood of mothers with a history of GDM reporting breastfeeding in hospital, compared to mothers without a history of GDM (not including pumping) (p = 0.01)

• Mothers with GDM: 73%

• Mothers without GDM: 75.7%

â likelihood of mothers with a history of GDM reporting breastfeeding in the first hour following birth, compared to women without a history of GDM (p < 0.0001)

• Mothers with GDM: 55.7%

• Mothers without GDM: 65.5%

â likelihood of mothers with a history of GDM reporting hospital staff helping them with breastfeeding, compared to mothers without a history of GDM (p = 0.03)

• Mothers with GDM: 80%

• Mothers without GDM: 82.5%

â likelihood of mothers with a history of GDM reporting feeding only breastmilk in the hospital, compared to mothers without a history of GDM (p < 0.0001)

• Mothers with GDM: 41%

• Mothers without GDM: 52.4%

â likelihood of mothers with a history of GDM reporting feeding on demand, compared to mothers without a history of GDM (p = 0.04)

• Mothers with GDM: 81.2%

• Mothers without GDM: 83.6%

2009 -2011 Cohort

â likelihood of mothers with a history of GDM reporting their baby stayed in the same room, compared to mothers without a history of GDM (p < 0.003)

• Mothers with GDM: 77.8%

• Mothers without GDM: 81.8%

â likelihood of mothers with a history of GDM reporting breastfeeding in hospital, compared to mothers without a history of GDM (not including pumping) (p < 0.01)

• Mothers with GDM: 88.2%

• Mothers without GDM: 91.9%

â likelihood of mothers with a history of GDM reporting breastfeeding in the first hour following birth, compared to women without a history of GDM (p < 0.0001)

• Mothers with GDM: 60.2%

• Mothers without GDM: 68.6%

â likelihood of mothers with a history of GDM reporting feeding only breastmilk in the hospital, compared to mothers without a history of GDM (p < 0.0001)

• Mothers with GDM: 39.4%

• Mothers without GDM: 52.7%

á likelihood of mothers with a history of GDM reporting receiving a pump, compared to mothers without a history of GDM (p < 0.001)

• Mothers with GDM: 39.5%

• Mothers without GDM: 31.8%

á likelihood of mothers with a history of GDM reporting receiving a formula gift pack, compared to mothers without a history of GDM (p = 0.002)

• Mothers with GDM: 71%

Mothers without GDM: 68.9%

• The data included in this study was from 16 states and New York City and therefor may not be generalisable to the whole of the United States

• Variation in breastfeeding practices between facilities

• Survey did not include data on all 10 BFHI steps (steps 1, 2 and 10 not asked)

• The observational study design could allow for the possibility of reverse causality

• PRAMS has limited data on maternity leave and employment

• Potential for reporting bias

• Criteria for GDM diagnosis unknown

• ~ 10% of respondents with missing data excluded

• No operational definitions for breastfeeding

30.5/40 (76.25%)

(Stuebe et al., 2016) [58]

A Cluster Randomized Trial of Tailored Breastfeeding Support for Women with Gestational Diabetes

United States

That a breastfeeding support intervention integrated within a lifestyle intervention would increase duration of any and exclusive breastfeeding among women with GDM compared to usual care

Randomised controlled trial

215 women screened for the study. After removal of women not meeting study inclusion criteria, 100 women were eligible and enrolled in the trial. Recruitment from 29/06/2012 – 11/09/2014

• Intervention group: 50 women

• Control group: 50 women

Maternal, infant, and breastfeeding data

• Women completed study assessments at

- 22 – 36 weeks’ gestation

- 6 weeks postpartum

- At the end of the intensive intervention

- 4 months postpartum

- 7 months postpartum

• 10 months postpartum

á Likelihood of women in the experimental group to be breastfeeding, and to be breastfeeding exclusively throughout the follow up period, when compared to women in the control group

á rates of any and exclusive breastfeeding at 4 weeks postpartum amongst women in the experimental group when compared to women in the control group

• Any breastfeeding: 87%. SE 5.4% in the experimental group, versus 64%. SE 7.7% in the control group

• Exclusive breastfeeding: 45%. SE 7.9%. in the experimental group, versus 23%. SE 6.4% in the control group

â likelihood of formula milk introduction amongst women in the experimental group when compared to women in the control group

• aHR 0.50, 95% CI 0.34–0.72

• Cluster randomisation may not balance the groups as efficiently as individual randomisation

• Higher loss of follow up than anticipated

• Low follow up rates diminished the study’s ability to measure the extent to which the intervention increased achievement of Health People 2020 targets, including 3 month exclusive breastfeeding rates and 6 moth any breastfeeding rates

• Low follow up rate might bias results

• Breastfeeding results were self reported, providing for the possibility of social desirability bias to affect results

• No operational definitions for breastfeeding

34.5/40 (86.25%)