Written by: Lauren Milligan Newmark, Ph.D. | Issue # 129 | 2025
- The MILQ study assessed milk intake and milk nutrient concentration from four global populations in The Gambia, Bangladesh, Brazil, and Denmark with the goal of producing reference values for human milk nutrients.
- Using two different methods to quantify milk intake from 1 – 8.5 months, the study found a mean daily milk intake of 781 grams, or roughly 0.75 liters.
- Milk intake was most similar between study sites during the period of exclusive breastfeeding, and became more variable as mixed feeding became more common.
- Milk nutrient concentration did not vary in response to milk intake, which means Infants with higher daily milk intakes also had a greater intake of milk nutrients.
Baby bottles come clearly marked with milliliter and ounce lines, making it easy for parents using bottles to track exactly how much milk their infant consumes at each feeding. But no such volume markers exist for parents of exclusively breastfed babies, leaving many wondering whether their infant is getting enough milk.
Milk researchers face a similar obstacle. Although the nutrient composition of human milk can be accurately determined with relative ease, milk transfer isn’t observable, and reliably measuring milk intake for breastfeeding infants can be difficult and costly [1, 2]. Making things more challenging, milk intake varies from day to day and throughout the course of lactation, meaning small study populations or short-term research projects may not produce trustworthy data on human milk intake [1, 2].
Given these challenges, it is perhaps not surprising to learn that there are no adequately established reference values (RVs) for human milk nutrients [1, 3]. While Recommended Dietary Allowance (RDA) guidelines help children and adults consume nutritionally adequate diets, there are no such benchmarks to evaluate the dietary needs of human infants. These RVs would not only be crucial for identifying potential nutritional deficiencies in breastfed infants but would also help guide recommendations from infant formula manufacturers.
The Mothers, Infants, and Lactation Quality (MILQ) study was started just over seven years ago by an international team of researchers with the goal of establishing RVs for human milk nutrients across the first 8.5 months of life [1, 3]. To meet this goal, the MILQ study first had to quantify how milk intake naturally differs among various global populations and changes throughout the duration of lactation.
Seven years may seem like a long time from study inception to research publication, but that time frame speaks to the comprehensive approach of the team and the time needed to collect and analyze the required data. Similar to the rigorous process for determining RDAs for children and adults, establishing RVs for human milk nutrients required a large body of scientific evidence on both milk nutrient content and milk intake and had to represent mothers and infants from across the globe, not just one population.
To achieve this, the researchers focused on recruiting healthy and well-nourished women and their infants from four distinct global populations: The Gambia, Bangladesh, Brazil, and Denmark [1]. They expected some natural variation in milk composition and how much infants consumed across (and within) these populations while also assuming that well-nourished mothers would produce adequate nutrition for their infants. Thus, any variations observed would represent the normal, healthy range of nutrient intake for human infants [1].
Determining which mothers were well-nourished required specific, objective criteria that could be applied to all four study populations. This included data on body composition (e.g., pre-pregnancy maternal BMI, upper arm circumference), health (e.g., hemoglobin levels to rule out anemia), and dietary quality (scored based on results from a diet questionnaire) [1]. The MILQ study originally enrolled 1,882 mother-infant pairs in the study. However, due to issues with sampling or participant withdrawal, ultimately ended up with a study population of 558 mother-infants who provided milk samples for nutrient analysis as well as data on milk intake from three time points between one month and 8.5 months postpartum (1 – 3.49 months, 3.5 – 5.99 months, and 6 – 8.5 months) [1]. In the end, the study had 1,458 data points on milk intake from the four populations.
In Brazil, Bangladesh, and The Gambia, milk intake was determined using a stable isotope dilution method, commonly referred to as the “dose-to-mother” method, whereas Denmark used test weighing. Both methods have the benefit of not disrupting the normal feeding patterns of infants, and each can track milk intake over the course of the day—and in the case of dose-to-mother, over the course of 14 days [1].
The dose-to-mother method is rather ingenious, taking advantage of a safe, stable isotope (deuterium oxide) that can be consumed by the mother in water. After drinking, the isotope becomes distributed throughout the mother’s body water, including in her breast milk. As the infant drinks this milk, they ingest some of the isotope. The levels of this isotope in the saliva collected from mothers and infants on days 1, 2, 3, 4, 13 and 14 represent the transfer of the isotope from mother to infant in milk. By applying some impressive mathematical models, the research team can use maternal and infant isotope data to calculate the total volume of milk consumed by the infant during that two week period.
Test-weighing involves a lot fewer steps. Mothers were provided with digital scales and asked to weigh their infants immediately before and immediately after each feeding over the course of 24 hours. This method is less accurate than dose-to-mother and also believed to underestimate milk intake by approximately 5%. To correct for this, the MILQ study corrected all data from Denmark by a factor of 1.05.
Across all four populations and all three time points, the mean milk intake was 781 grams per day (g/d), which is just about 0.75 liters or the size of a standard bottle of wine [1]. This value is slightly higher than the mean value of 690 g/d (or 0.67 L) previously reported from a meta-analysis of 167 global studies [4]. Importantly, test weighing in Denmark produced similar results to the three populations that used dose-to-mother, suggesting that test weighing is a valid method for assessing milk intake when dose-to-mother isn’t feasible [1].
Milk intakes were consistent between the four populations at the first time point, but became more variable at the second and third collection point. This finding suggests similar milk production across exclusively breastfeeding mothers between one and three months. But, as fewer mothers reported exclusive breastfeeding and nonmilk foods were introduced, variation in milk intake increased over time. By the 6 – 8.5 month time point, Brazil and The Gambia had comparatively higher intakes of human milk compared with Denmark and Bangladesh [1]. These findings highlight the importance of long-term studies with repeated sampling, sampling from multiple populations, and consideration of exclusive breast feeding or mixed feeding strategies when interpreting the data.
In addition to establishing expected volumes of milk intake across the first 8.5 months of lactation, the MILQ team was interested in understanding how milk nutrient concentration may vary with respect to milk intake. Does the milk nutrient concentration change in response to patterns of consumption? Do infants who consume 600 g/d of human milk still get the same total amount of protein, fat, vitamins, and trace elements per day as infants that consume 800 g/d? The study authors were surprised to find that milk nutrient concentrations were not correlated with daily milk intake [1]. However, the total daily intake of a nutrient (calculated from volume x nutrient concentration) was positively correlated with volume—more milk, more nutrients [1].
The implications of this finding for infant nutrition are uncertain. In a population of well-nourished and healthy mothers, infants who consume widely varying amounts of milk (both low and high volumes) likely both receive sufficient nutrition [1]. However, in undernourished populations, the absence of compensatory or “harmonizing” [1] mechanisms (where the milk becomes more nutrient-dense to make up for low volume) could result in significant nutritional deficits for infants that consume less milk. These are questions the MILQ team plans to address in a future publication—so stay tuned!
One clear implication is the need for milk researchers to consider both milk nutrient concentration and milk intake when evaluating an infant’s nutrition. As anyone who has tracked their daily calorie intake knows, serving size matters. By successfully establishing two reliable methods for quantifying “serving size”, the MILQ team has created a path for future studies on human lactation to accurately include this essential piece of data.
References
- Moore SE, Devi S, Kurpad A, Peerson JM, Christensen SH, Islam MM, Kac G, Michaelsen KF, Silva GT, Allen LH, Mølgaard C. Breast milk Intake from 1 to 8.5 months of lactation in the multisite Mothers, Infants and Lactation Quality (MILQ) study. Advances in Nutrition. 2025 Aug 26:100456.
- Miller EM, Aiello MO, Fujita M, Hinde K, Milligan L, Quinn EA. Field and laboratory methods in human milk research. American Journal of Human Biology. 2013 Jan;25(1):1-1.
- Allen LH, Hampel D, Shahab-Ferdows S, Andersson M, Barros E, Doel AM, Eriksen KG, Christensen SH, Islam M, Kac G, Keya FK. The Mothers, Infants, and Lactation Quality (MILQ) study: a multi-center collaboration. Current Developments in Nutrition. 2021 Oct 1;5(10):nzab116.
- Rios-Leyvraz M, Yao Q. The volume of breast milk intake in infants and young children: a systematic review and meta-analysis. Breastfeeding Medicine. 2023 Mar 1;18(3):188-97.