Pesticides in Baby Food: Sampling Plan and Testing Methods
Starting at about four months of age, baby food constitutes an important food source for many infants. Nine out of every ten babies eat at least some commercial baby food, and according to Gerber, the average American baby eats 600 jars of baby food (including baby food juices) in his or her first year.
Ninety-six percent of all baby food sold in the United States is made by Gerber, Heinz and Beech-Nut. Respectively, these companies control 69, 14, and 13 percent of the 1.78 billion jar annual market (Nielson)
All of the big three baby food makers sell products in three stages. Stage-one for children up to six months, stage-two for children from six to nine months, and stage-three for children nine months and older. Gerber is less dominant in the first stage market, controlling 56 percent, compared with 32 percent for Beech-Nut and 10 percent for Heinz.
Our sampling protocol was aimed at the stage-one fruits and vegetables in order to determine the exposure to pesticides that occurs at the earliest, and typically most vulnerable stages of infant development. The only exception was plum products which are marketed exclusively at the second level of baby foods, for infants older than six months.
Unlike second and third stage products, stage-one products are usually composed of single fruits and vegetables so that parents can identify any food allergies their children may have. Our sampling strategy was limited by testing costs. Within these constraints we tested fruits and vegetables that comprise a significant percentage of baby food sold and were likely to contain pesticide residues.
We tested eight foods (applesauce, garden vegetables or pea and carrot blend, green beans, peaches, pears, plums, squash and sweet potatoes) sampled from three major metropolitan areas; Philadelphia, Denver, and San Francisco. A total of 24 samples (eight foods in three cities) were obtained from each company for a total of 72 samples in all. All samples were purchased at retail from grocery stores as they would normally be purchased by consumers.
One or more of the nine fruit and vegetables included in this study is an ingredient in over 200 different baby food products manufactured by the big three, 59% of the products sold. Sales of products containing these 9 fruits and vegetables amounted to over 900 million of the 1.3 billion jars (over 69%) of baby food sold from February 1994 to February 1995, excluding juice (Nielson 1995).
Samples were made of a composite of 24, 2.5 ounce jars, or 12, 4.5 ounce jars identified with the same production code. This produced three distinct and uniform samples from three major metropolitan areas, for each food tested.
We sampled four fruit products; applesauce, peaches, pears and plums. The first three of these food constitute 66% of the fruit baby food sold from the beginners foods category (Table 1). Plums, the fourth fruit sampled, is the next most popular fruit sold as a single fruit product (not blended with other fruits) by all three companies. Pure plum products, however, are all marketed exclusively at the second level of baby foods, for infants older than six months.
Strained bananas, the most popular beginners baby food, was excluded from our testing because data from the FDA and other independent sources indicate relatively low pesticide residues.
At least one of the four fruits we sampled (applesauce, peaches, pears, and plums) is an ingredient in over 95 different products manufactured by the three major baby food companies. Apples are the most widely used product in baby foods, followed by bananas and peaches.
Table 1: The most commonly eaten beginner-level* baby foods.
|Fruits||Jars sold (Millions)||Vegetables||Jars sold (Millions)|
|1. Bananas||37.7||1. Sweet Potatoes||35.0|
|2. Applesauce||35.8||2. Carrots||30.2|
|3. Pears||28.7||3. Squash||25.3|
|4. Peaches||23.5||4. Green Beans||20.3|
|5. Plums**||18.1||5. Peas||20.4|
*for children under six months of age
**2nd stage food (about 6 to 9 months)
Units sold are rounded to the nearest 100,000.
Source: Nielson sales numbers for 1994.
Sweet potatoes, squash, green beans, peas and carrots are the only vegetable baby foods sold by the big three in the beginners foods category. These five vegetables are ingredients in 113 products sold in all stages of baby food marketed by these three companies.
We sampled sweet potatoes, squash and green beans from each of the producers' beginners foods category. In order to stay within budget and include peas and carrots in our sampling plan we chose garden vegetables which contains peas, water, carrots and spinach, from one company, and carrots and peas from another. Because the third company does not make a product with both of these ingredients, we combined their beginners and older pea and carrot products in equal portions to create our test sample.
All samples were tested for pesticides using standard procedures employed by the Food and Drug Administration (FDA) and described in the FDA Pesticide Analytical Manual. The Luke extraction method was used for all polar and non-polar pesticides detectable by either Gas chromatography (GC) or High Performance Liquid Chromotography (HPLC). All pesticides detected by GC were confirmed using dual columns and dual detectors. All pesticides detected by HPLC were confirmed with another method.
For pesticides detectable only by single residue methods such as ethylene thio urea (ETU) or benomyl, procedures specified in the Food and Drug Administration's Pesticide Analytical Manual (PAM) were used.
The limits of quantification for pesticide in this study were, in parts per billion:
|Organophosphate and organochlorine insecticides||2 ppb|
|Organonitrogen compounds||5 ppb|
|Carbamate compounds||5 ppb|
For a more detailed description of the testing methods used, see Appendix 1.