A concise history of infant formula (twists and turns included)

By Andrew J. Schuman, MD

Finding an acceptable alternative to breast milk has proved to be a complicated quest that continues today with an ever-growing assortment of modified and specialized infant formulas.

If you are a "mature" pediatrician—one older than 40 years or so—there is a good chance that, if you were not breastfed as an infant, you were fed a formula created by mixing 13 oz of evaporated milk with 19 oz of water and two tablespoons of either corn syrup or table sugar. Every day, parents prepared a day's worth of this formula, transferred it to bottles that they had sterilized in a pan of boiling water, and stored it in a refrigerator until used. In addition to formula, infants received supplemental vitamins and iron.¹

Infant nutrition has a fascinating history that began long before pediatricians recommended evaporated milk formula, and eventually commercial formula, as alternatives to breastfeeding. In this first article in an occasional series that puts the practice of pediatrics into historical perspective, we'll take a look at how infant formulas were developed and how they evolved over time.

From wet nursing to dry nursing

Before the era of "modern" medicine, breastfeeding was the preferred method of feeding infants, just as it is today. But if a mother's milk supply was inadequate or she chose not to nurse, the family often employed a "wet nurse" to nourish infants. This practice was common in Europe during the 18th century and in America during the colonial period. Families would hire a wet nurse to reside in the family's home or send the infant to live in the wet nurse's home and retrieve the baby after he or she had been weaned.

Wet nurses were selected with the utmost care, because it was believed the quality of the milk the baby received determined his or her future "disposition." Brunette wet nurses were preferred to blondes or redheads because their breast milk was thought to be more nutritious and their disposition more "balanced."

During the 18th century in Europe, the demand for wet nurses was so great that bureaus were established where wet nurses could register and reside until their services were needed. Governments regulated the bureaus strictly. Laws mandated that wet nurses undergo routine health examinations and prohibited them from nursing more than one infant at a time.

Eventually, wet nursing fell out of favor, and attention turned to finding an adequate substitute for mother's milk.² The practice of feeding human babies milk from animals, called dry nursing, began to flourish in the 19th century. Milk from a variety animals—goats, cows, mares, and donkeys—was used. Cow's milk became the most widely used because of its ready availability (although donkey's milk was thought to be healthier because its appearance most closely resembled that of human milk). Physicians argued about the best way to prepare milk. Some suggested giving it fresh from the animal, while others recommended that it be warmed or boiled first, and still others suggested diluting it with water and adding sugar or honey.³

Before the baby bottle came into use, milk was spoon fed to infants or given via a cow's horn fitted with chamois at the small end as a nipple. When baby bottles were adopted during the Industrial Revolution, many popular designs evolved. Some were submarine-shaped and made from metal, glass, or pottery. They had a circular opening in the top that could be occluded with a cork, and one end tapered to a hole with a rim for securing a nipple. Another popular design was the spouted feeder, which resembled a teapot and was equipped with a handle and a long spout that terminated in a nipple-shaped bulb. The nipple opening of both types of bottles was covered with punctured chamois cloth, parchment, or sponge.² Rubber nipples became widely available and very popular after their invention by Elijah Pratt, an American, in 1845.

After an infant was weaned from breast milk or cow's milk he or she was given an infant food called pap, which consisted of boiled milk or water thickened with baked wheat flour and, sometimes, egg yolk. A more elaborate infant food, called panada, was made from bread, flour, and cereals cooked in a milk- or water-based broth. Detailed recipes for various kinds of infant paps and panadas have been published in cookbooks throughout history.¹

The search for a breast milk substitute

A longtime goal of nutritionists and physicians was to develop an adequate breast milk substitute. In the early 19th century, it was observed that infants fed unaltered cow's milk had a high mortality rate and were prone to "indigestion" and dehydration compared with those who were breastfed. In 1838, a German scientist, Johann Franz Simon, published the first chemical analysis of human and cow's milk, which served as the basis for formula nutrition science for decades to follow. He discovered that cow's milk had a higher protein content and a lower carbohydrate content than human milk. In addition, he (and later investigators) believed that the larger curds of cow's milk (compared to the small curds of human milk), were responsible for the "indigestibility of cow's milk."²

Empirically, physicians began to recommend that water, sugar, and cream be added to cow's milk to render it more digestible and closer to human milk. By 1860, a German chemist, Justus von Leibig, developed the first commercial baby food, a powdered formula made from wheat flour, cow's milk, malt flour, and potassium bicarbonate. The formula, which was added to heated cow's milk, soon became popular in Europe. Leibig's Soluble Infant Food was the first commercial baby food in the US, selling in groceries for $1 a bottle in 1869.

In the 1870s, Nestle's Infant Food, made with malt, cow's milk, sugar, and wheat flour, became available in the US, selling for $.50 a bottle. In contrast to Leibig's Food, Nestle's formula was diluted with water only, requiring no cow's milk to prepare, and thus was the first complete artificial formula available in this country.

Several cow's milk modifier formulas were introduced over the next 20 years, and by 1897 the Sears catalogue was selling no fewer than eight brands of commercial infant foods, including Horlick's Malted food ($.75 per bottle), Mellin's Infant Food ($.75 per bottle), and Ridge's Food for Infants ($.65 per bottle).⁴ Despite their widespread availability, these proprietary formulas realized only modest sales in the late 19th century because they were expensive in comparison to cow's milk. Most mothers continued to breastfeed their infants.

Physicians take charge

In the late 19th century, many physicians thought that infant nutrition should be directed not by formula manufacturers but by physicians themselves. Many believed that commercial formulas were nutritionally inadequate and therefore inappropriate for young infants.

Thomas Morgan Rotch of Harvard Medical School developed what came to be known as the "percentage method" of infant formula feeding, which was popular among medical professionals from 1890 to 1915. He taught that because cow's milk contains more casein than human milk, it must be diluted to lower the percentage of casein. The process of dilution, however, decreases the sugar and fat content to less than that of human milk. To correct these deficiencies, cream and sugar were added in precise amounts.

Cow's milk formulas prescribed by the percentage method were compounded by a milk laboratory or, more often, by a home method that was time and labor intensive. Physicians were taught to monitor growth carefully and to examine the infant's stool and modify the formula based on its appearance.³

By the 1920s, physicians were frustrated by the complexity of formula prescribing and modifications associated with Rotch's percentage method. They eventually began to recommend either commercial formulas or simple homemade formulas made with evaporated milk.

Pasteurization and healthier milk

In the late 19th and early 20th centuries, physicians came to understand that diseases were caused by germs and could be transmitted by consuming contaminated foodstuffs. In particular, raw milk, which spoiled readily (refrigeration was not widely available until about 1910), was found to transmit a variety of diseases, including tuberculosis, typhoid fever, cholera, and diphtheria.

In 1864, Louis Pasteur discovered that keeping wine at a high temperature killed the bacteria that caused the wine to sour. The pasteurization process was employed by the dairy industry in 1890—not to make milk "healthier" but to prevent milk transported in unrefrigerated railroad cars from souring. Several years later, it was discovered that pasteurization also protected against milk-borne diseases.³

Many physicians vigorously opposed pasteurization, however, because they believed that the process significantly diminished the nutritional value of milk. In fact, pasteurized milk was found to be deficient in what were later identified as vitamins C and D, and children consuming pasteurized milk received daily doses of orange juice and cod liver oil (rich in vitamins A and D) to prevent scurvy and rickets. Pasteurization of milk became a universal practice in the US by about 1915.

Big breakthrough: Evaporated milk

Perhaps the greatest advance in milk science occurred before the Civil War. Gail Borden discovered and patented a process of heating milk to high temperatures in sealed kettles, which removed close to half the water content of the milk. By adding sugar as a preservative to the resulting product, Gail Borden invented sweetened "condensed" milk that had a long shelf life and could be transported easily without fear of spoilage. Condensed milk was an invaluable ration for soldiers during the Civil War and was later promoted to mothers as an infant food. Because of its high sugar content, however, physicians discouraged its use as an infant formula.

A method of producing unsweetened evaporated milk was developed by John B. Myenberg in 1883. The process involved evaporating approximately 60% of the water from milk in a sealed metal still, then sterilizing the condensed milk by heating it to above 200°. This process altered the physical properties of milk, homogenizing it and rendering the curd smaller and more digestible than boiled pasteurized milk. Studies published in the 1920s and 1930s demonstrated that large numbers of babies fed evaporated milk formula grew as well as breastfed infants did.⁵ Physicians and parents, reassured by this evidence and encouraged by the low cost and widespread availability of evaporated milk, almost universally endorsed evaporated milk formula to feed infants. In the 1930s, physicians were taught to mix evaporated milk formula by combining 2 oz of cow's milk per pound of body weight per day with ¹Ž₈ oz of sugar per pound of weight per day and enough water to provide an infant with 3 oz of fluid volume per pound per day. During the Great Depression, corn syrup replaced sugar as a source of carbohydrate because of cost and availability. Gradually the formula was simplified to the one described at the beginning of this article.

By the 1940s and through the 1960s, most infants who were not breastfed received evaporated milk formula, as well as vitamins and iron supplements. It is estimated that, in 1960, 80% of bottle-fed infants in the US were being fed with an evaporated milk formula.³

Seeking a "humanized" formula

In the early 20th century, the focus of nutrition scientists shifted from modifying the protein content of infant formula to making its carbohydrate and fat content more closely resemble that of human milk. Some researchers believed that the carbohydrate content of cow's milk should be supplemented with maltose and dextrins; at their request, E. Mead Johnson, the founder of the Mead Johnson company, produced a cow's milk additive called Dextri-Maltose. Dextri-Maltose was introduced at the 1912 meeting of the American Medical Association (AMA) and was sold only by physicians to mothers.

A few years later, in 1919, a new infant formula was introduced that replaced milk fat with a fat blend derived from animal and vegetable fats. This formula, which more closely resembled human milk than cow's milk, was called SMA (for "simulated milk adapted"). SMA was also the first formula to include cod liver oil. Soon after SMA was introduced, Nestle's Infant Food added cod liver oil to its formula, as did most other infant formulas.⁴

In the 1920s, other "humanized" infant formulas were produced and marketed to the American public. Nestle produced a formula with a vegetable-oil-derived fat blend, called Lactogen, that was positioned to compete with SMA.

Another humanized infant formula was developed by Alfred W. Bosworth, a milk chemist working for the biochemistry department of Harvard Medical School, and by Henry Bowditch, a Boston pediatrician who was employed at the Boston Floating Hospital. They experimented with an infant formula derived from cow's milk by adding varying amounts of vegetable oils, calcium, and phosphorus salts and preparing blends with different lactose concentrations. Bosworth and Bowditch tested more than 200 formulas in clinical trials before they considered their infant formula complete.

In 1924, Bosworth agreed to have his formula marketed by the Moores and Ross Milk Company of Columbus, Ohio. The new formula was produced at the Franklin Brewery plant in Columbus and was originally sold by physicians in plain cans upon which they could place their own label. In 1926, the formula was renamed "Similac" because it was "similar to lactation"—a name proposed by Morris Fishbein, MD, editor of the Journal of the American Medical Association.

In the late 1920s, the Mead Johnson company introduced Sobee, the first soy-based formula. Several years later, the company marketed Pablum, the first precooked fortified infant cereal. Pablum was a mixture of wheat, oats, corn, bone meal, wheat germ, alfalfa, and dried brewer's yeast fortified with minerals and vitamins.

The rise of proprietary formulas

Despite the introduction of proprietary infant formulas in the 1920s, most parents continued to use evaporated milk formula because it was easy to prepare and affordable. It was not until the 1950s that commercial formulas began to slowly gain acceptance (Figure 1 in the print edition, Adapted from Fomon SJ: Infant feeding in the 20th century: Formula and beikost. J Nutr 2001;131:409S).

In the decades that followed, a variety of new formulas came on the market. Nutramigen, introduced in 1942, was the first protein hydrolysate infant formula. Ross Laboratories' Similac concentrate became available in 1951, and Mead Johnson's Enfamil (for "infant milk") was introduced in 1959. In that year, Ross first marketed Similac with iron. Iron-fortified formula was poorly accepted initially because of the widespread belief that iron fortification caused gastrointestinal disturbances such as diarrhea and constipation.

During the 1960s, commercial formulas grew in popularity, and by the mid-1970s they had all but replaced evaporated milk formulas as the "standard" for infant nutrition. During this time, the percentage of women who breastfed their newborn reached an all-time low (25%), in part because of the ease of use and low cost of commercial formula and a belief that formulas were "medically approved" to provide optimal nutrition for young infants (Figure 2 in the print edition, Adapted from Fomon SJ: Infant feeding in the 20th century: Formula and beikost. J Nutr 2001;131:409S).

A major factor in the acceptance of commercial formulas was their use in hospitals to feed newborn infants during the 1960s and 1970s. To encourage acceptance, formula companies began to provide inexpensive or free formula to hospitals in ready-to-feed bottles, enabling the phasing out of hospital formula preparation rooms. Mothers who witnessed how well their newborns accepted these easily prepared formulas were often convinced to continue this practice at home. Moreover, although pediatricians did not dissuade mothers from nursing, it was not strongly encouraged, as it is today.

The modern era: Fine-tuning formulas

The American Academy of Pediatrics Committee on Nutrition first made recommendations for vitamins and mineral levels for infant formulas in 1967. These recommendations have been revised periodically.⁶ (See "AAP's Committee on Nutrition: Infant formula and beyond,".) In 1969, the committee endorsed iron fortification of infant formula; in the years that followed, the incidence of iron deficiency anemia dropped strikingly.⁷

In 1978 and 1979, 141 cases of hypochloremic metabolic alkalosis in infants, resulting from consumption of two soy formulas, Neo-Mull-Soy and Cho Free (produced by Syntex, Inc.), were reported to the Centers for Disease Control. This prompted the passage of the Infant Formula Act of 1980, which set maximum and minimum standards for many nutrients in formulas and mandated testing and manufacturing standards as well.

Perhaps the greatest achievement of nutrition research over the past several decades has been the introduction of specialty formulas and human milk modifiers used to feed premature and very low-birthweight infants. For term and near-term infants, formula manufacturers have continued to improve their "standard" formulas to more closely resemble breast milk. In 1997, Ross's Similac was reformulated to change the whey:casein ratio (then, 18%:82%) to 52%:48%, which more closely resembles that of human milk (70%:30%). The ratio of Mead Johnson's Enfamil is 60%:40%. Both Mead Johnson's and Ross's formulas contain added nucleotides in amounts similar to those in breast milk, and this year both companies have introduced formulas that contain long-chain polyunsaturated fatty acids.

Over the past few years, these two companies also have begun marketing "niche" formulas, including lactose-free formulas (both companies), a soy formula with dietary fiber to hasten recovery from gastroenteritis (Ross), and a formula with rice starch for babies with reflux (Mead Johnson).

Today's young infants are the beneficiaries of a long and complicated history of infant formula. While we continue to encourage mothers to breastfeed their infants, babies who are fed formula from birth or are weaned to formula from breast milk receive the best nutrition medical science has to offer.

ACKNOWLEDGMENT

The author thanks Virginia A. Mason for her assistance in preparing the manuscript of this article.

REFERENCES

1. Siberry GK (ed): Harriet Lane Handbook, ed 14. St. Louis, Mosby Year Book, 1996

2. Spaulding M: Nurturing Yesterday's Child: A Portrayal of the Drake Collection of Paediatric History. Philadelphia, BC Decker, 1991

3. Cone TE: History of American Pediatrics. Boston, Little, Brown, and Company, 1979

4. Apple RD: Mothers and Medicine: A Social History of Infant Feeding. Madison, Wis., University of Wisconsin Press, 1987

5. Marriot WM, Schoenthal L: An experimental study of the use of unsweetened evaporated milk for the preparation of infant feeding formulas. Arch Pediatr 1929;46:135

6. American Academy of Pediatrics, Committee on Nutrition: Proposed changes in food and drug administration regulations concerning formula products and vitamin-mineral dietary supplements for infants. Pediatrics 1967;40:916

7. American Academy of Pediatrics, Committee on Nutrition: Iron balance and requirements in infancy. Pediatrics 1969;43:134

DR. SCHUMAN is adjunct assistant professor of pediatrics at Dartmouth Medical School, Lebanon, N.H., and practices pediatrics at Hampshire Pediatrics, Manchester, N.H. He is a contributing editor for Contemporary Pediatrics. He serves on the speakers' bureaus of Ross Laboratories and Mead Johnson.

The pediatrician's infant formula trivia quiz

1. What percentage of infants in the US are breastfed at birth?

a. 85% b. 69.5% c. 25%

Answer: b. According to the most recent data, the 2001 Ross Mother's Survey (Ryan AS et al: Pediatrics 2002;110:1103), 69.5% of newborns in the US are breastfed at birth. That is significantly higher than the 50% recorded a decade ago.

2. What percentage of infants are breastfed at 6 months?

a. 50% b. 39% c. 32.5%

Answer: c. According to the 2001 Ross Mothers' Survey, 32.5% percent of infants are breastfed at 6 months.

3. Which formula manufacturer has the largest market share in the US?

a. Ross b. Mead Johnson c. Carnation

Answer: b. According the latest (2000) published information from the US Department of Agriculture (USDA), Mead Johnson holds 52% of the infant formula market, including 68% of the market from the federal Women, Infants, and Children (WIC) nutrition program. In 1994, Mead Johnson's market share was 27%. Ross has a 35% market share (down from 53% in 1994), and Carnation has 12% of the market. PBM products (such as Parents Choice, manufactured by Wyeth), has a 1% market share. Approximately 27 billion ounces of formula are consumed each year in this country—accounting for about $2.9 billion in revenue for formula manufacturers. (Source: Oliveira V et al: Infant formula prices and availability: Final report to Congress. Economic Research Service, USDA 2001, www.ers.usda.gov/publications/efan02001/efan02001d.pdf )

4. Which type of formula is most popular?

a. Powder b.. Ready-to-feed c. Concentrate

Answer: a. Powder. Sales of powdered formula rose from 42% in 1994 to 62% in 2000; sales of liquid concentrate declined from 42% to 27%. Powder is the most economical formula preparation.

5. Infants with diarrhea are often given Pedialyte (Ross) before they resume regular formula. When was Pedialyte introduced?

a. 1956 b. 1966 c. 1976

Answer: b. 1966

6. Where do parents purchase most infant formula?

a. Supermarkets b. Pharmacies c. Mass merchandisers (Walmart, Costco, etc.)

Answer: a. In 2000, 69% of formula in the US was purchased in supermarkets; 28% was bought from mass merchandisers; and 3%, from pharmacies.

7. What percentage of formula sold in the US is milk-based?

a. 97% b. 87% c. 77%

Answer: c. 77%

8. Gerber introduced its own brand of infant formula in 1989 that vanished from store shelves in 1997.Who manufactured Gerber's formula?

a. Carnation b. Mead Johnson c. Wyeth

Answer: b. Mead Johnson

9. According to 2000 USDA data, which brand of milk-based powder is the most expensive?

a. PBM (Parents' Choice) b. Similac c. Enfamil

Answer: b. Similac (Ross). According to USDA data, the average cost of 26 reconstituted ounces in 2000 was $2.63. The least expensive brand was PBM, manufactured by Wyeth: In 2000, the average cost of 26 reconstituted ounces was $1.56.

10. Which brand of soy-based powder is the most expensive?

a. Prosobee b. PBM (Parents' Choice Soy) c. Isomil

Answer: a. Prosobee (Mead Johnson). According to USDA data, Prosobee cost $2.90 for 26 reconstituted ounces in 2000. The least expensive soy-based powder was PBM (Wyeth), which cost $1.61 for 26 reconstituted ounces in 2000.

AAP's Committee on Nutrition: Infant formula and beyond

The Committee on Nutrition of the American Academy of Pediatrics was established by the AAP's Executive Board on April 1, 1954. Its first chairman was Charles D. May, then chairman of the Department of Pediatrics at the College of Medicine, University of Iowa. Its charge was as follows:

"This Committee shall concern itself with standards for nutritional requirements, optimal practices, and the interpretation of current knowledge as these affect infants, children, and adolescents."

The Committee on Nutrition initially published educational reports; it did not begin publishing policy statements until the mid-1960s. The committee provided invaluable assistance to the Food and Drug Administration (FDA) by defining nutritional requirements for infant formulas.Its 1967 recommendations for levels of nutrientsin infant formula was used by the FDA to createthe 1971 regulation establishing the minimum requirements for fat, protein, linoleic acid, and17 vitamins and minerals in formula. After an outbreak of chloride deficiency in infants fed certain formulas (see article), the Committee on Nutrition revised its recommendations regarding nutrient content.

The Infant Formula Act of 1980 gave the FDA authority to regulate the labeling of infant formula and establish quality control rules and regulations governing formula manufacturing. The act was revised in 1985, based on recommendations from the Committee on Nutrition, to include minimum concentrations of 29 nutrients and maximum concentrations of nine nutrients in infant formula.

The committee continues to play an important role in pediatric nutrition by issuing policy statements as new information becomes available and publishing the Pediatric Nutrition Handbook. Now in its fourth edition, the handbook provides pediatricians with information on a wide variety of nutrition topics. Recent policy statements from the committee have addressed iron fortification of infant formula (1979, 1989, 1999), the use of hypoallergenic infant formulas (2000), breastfeeding and the use of human milk (1997), soy-protein-based formulas (1998, 2001), and the use and misuse of fruit juice (2001).

 

Andrew Schuman. A concise history of infant formula (twists and turns included). Contemporary Pediatrics 2003;2:91.