What is the question?

• The question is: What is the relationship between shorter versus longer durations of any human milk feeding and diabetes outcomes in offspring?

What is the answer to the question?

• Moderate evidence from observational studies suggests that, among infants fed some amount of human milk, shorter versus longer durations of any human milk feeding are associated with higher risk of type 1 diabetes. 
  
  Limited but consistent evidence suggests that the duration of any human milk feeding is not associated with fasting glucose or insulin resistance in childhood or during the transition from childhood into adolescence. 
  
  There is insufficient evidence to determine whether or not there is a relationship between shorter versus longer durations of any human milk feeding and type 2 diabetes, prediabetes, or HbA1c throughout the lifespan, and fasting glucose and insulin resistance in adulthood. 

Why was this question asked? 

• This important public health question was identified and prioritized as part of the U.S. Department of Agriculture and Department of Health and Human Services Pregnancy and Birth to 24 Months Project.

How was this question answered? 

• A team of Nutrition Evidence Systematic Review staff conducted a systematic review in collaboration with a group of experts called a Technical Expert Collaborative.

What is the population of interest? 

• The population of interest was generally healthy infants and toddlers (ages 0-24 months) who were in studies examining diabetes outcomes throughout the lifespan.

What evidence was found? 

• This review includes 37 articles. 
• These articles compared infants fed human milk for shorter durations with infants fed human milk for longer durations. The infants could be fed any amount of human milk.
• These articles examined available evidence related to diabetes outcomes in offspring, including fasting glucose, HbA1C, glucose tolerance/insulin resistance, and the incidence and prevalence of prediabetes, type 1 diabetes, and type 2 diabetes. 
• Most of the evidence examined type 1 diabetes, fasting glucose, and insulin resistance.
  • Individuals who were fed human milk for shorter durations may have higher type 1 diabetes risk than individuals who were fed human milk for longer durations. 
  • Individuals who were fed human milk for shorter durations and individuals who were fed human milk for longer durations do not seem to have differences in fasting glucose or insulin resistance in childhood or during the transition into adolescence.
• There are limitations in the evidence as follows: 
  • There were not a lot of studies examining type 2 diabetes or intermediate diabetes outcomes such as fasting glucose, HbA1C, and insulin resistance.
  • Some of the articles studied groups of people that may have been too small to detect whether there is a relationship between shorter versus longer durations of any human milk feeding and type 1 diabetes in offspring.
  • Most of the studies asked mothers to remember how long they had fed their offspring human milk when they were infants, and if mothers of individuals with type 1 diabetes remembered or reported differently from mothers of individuals without type 1 diabetes, it could impact the findings.
  • It is possible for factors other than infant feeding to impact the outcomes, and these factors were addressed differently by different studies.
  • There was not a lot of evidence from the U.S. about fasting glucose or insulin resistance, and metabolic risk in the U.S. may be higher than it is in the populations with available evidence. 

How up-to-date is this systematic review? 

• This review includes literature from 01/1980 to 03/2016.

Background

• This systematic review was conducted as part of the U.S. Department of Agriculture and Department of Health and Human Services Pregnancy and Birth to 24 Months Project.
• The goal of this systematic review was to examine the following question: What is the relationship between shorter versus longer durations of any human milk feeding and diabetes outcomes in offspring?
• This systematic review examines comparisons of infants who were fed human milk for shorter durations with infants who were fed human milk for longer durations. Human milk feeding was defined as feeding human milk alone or in combination with infant formula and/or complementary foods or beverages such as cow’s milk.  Human milk was defined as mother’s own milk provided at the breast (i.e., nursing) or expressed and fed fresh or after refrigeration or freezing.  Donor milk (e.g., banked milk) was not examined in this review.  
• This systematic review examines available evidence related to diabetes outcomes in offspring, including fasting glucose, HbA1C, glucose tolerance/insulin resistance, and the incidence and prevalence of prediabetes, type 1 diabetes, and type 2 diabetes.

Conclusion Statement and Grade

• Moderate evidence from observational studies suggests that, among infants fed some amount of human milk, shorter versus longer durations of any human milk feeding are associated with higher risk of type 1 diabetes. 
  
  Limited but consistent evidence suggests that the duration of any human milk feeding is not associated with fasting glucose or insulin resistance in childhood or during the transition from childhood into adolescence. 
  
  There is insufficient evidence to determine whether or not there is a relationship between shorter versus longer durations of any human milk feeding and type 2 diabetes, prediabetes, or HbA1C throughout the lifespan, and fasting glucose and insulin resistance in adulthood.
  
  Grade: Moderate – type 1 diabetes; Limited – fasting glucose and insulin resistance in childhood and the transition into adolescence; Grade Not Assignable – fasting glucose and insulin resistance in adulthood, type 2 diabetes, prediabetes, HbA1C

Methods 

• The systematic review was conducted by a team of staff from the Nutrition Evidence Systematic Review in collaboration with a Technical Expert Collaborative.
• A single literature search was conducted to identify literature for several related systematic reviews that examined infant milk-feeding practices and different outcomes. The search was conducted in CINAHL, Cochrane, Embase, and PubMed, and used a search date range of January 1980 to March 2016. A manual search was done to identify articles that may not have been included in the electronic databases searched. 
• Articles were screened independently by 2 NESR analysts to determine which articles met predetermined criteria for inclusion.
• Data from each included article were extracted, risks of bias were assessed, and both were checked for accuracy. 
• The body of evidence was qualitatively synthesized, a conclusion statement was developed, and the strength of the evidence (grade) was assessed using pre-established criteria including evaluation of the internal validity/risk of bias, adequacy, consistency, impact, and generalizability of available evidence.

Summary of Evidence 

• Thirty-seven articles met the inclusion criteria for this systematic review, including 30 with evidence about type 1 diabetes, 1 with evidence about type 2 diabetes, and 6 with evidence about intermediate diabetes outcomes (fasting glucose and insulin resistance). 
• Evidence about the association between shorter versus longer durations of any human milk feeding and higher risk of type 1 diabetes was moderate. Across 22 independent observational studies (30 articles), 12 reported significant associations. With the exception of 1 study that had limited external validity, the significant associations between the duration of any human milk feeding and type 1 diabetes risk were inverse associations. The ability to draw stronger conclusions was primarily limited by insufficient sample sizes, concerns about internal validity such as the potential for confounding, and the retrospective collection of exposure data, which increases the risk of misclassification of the exposure.
• Evidence about the lack of an association between the duration of any human milk feeding and fasting glucose and insulin resistance during childhood and the transition into adolescence was limited. One cluster randomized controlled trial and 3 prospective cohort studies provided consistent evidence. The ability to draw stronger conclusions was primarily limited by the small number of studies and the limited evidence from the United States (where metabolic risk may be higher).
• Evidence related to type 2 diabetes, prediabetes, and HbA1C, and about fasting glucose and insulin resistance beyond early adolescence was scant.