Study of baby bottles and bottle liners and Bisphenol A

Health Canada scientists recently published a study that compared levels of Bisphenol A (BPA) present in non-polycarbonate baby bottles, to levels in polycarbonate bottles. The study also looked at bottle liners.

Context

• The article was published in the journal Food Additives and Contaminants, in June 2009.
• This article reports that, in comparison with the migration (i.e., leaching of the chemical BPA from the bottle into liquid contents) from polycarbonate bottles, non-polycarbonate baby bottles and baby bottle liners showed only "trace" levels of BPA. This study indicated that non-polycarbonate baby bottles, therefore, may be good alternatives to polycarbonate bottles.
• Some bottles were found, in laboratory testing, to contain tiny, "trace" amounts of BPA, most of them in the range of parts per trillion. One part per trillion, as an example, is equivalent to one cent in $10 billion.
• Due to advanced technological techniques, scientists are often able to detect minute quantities of substances that are of no apparent biological significance.
• These trace amounts may have been detected due to improved sensitivity of modern laboratory instruments. Such tiny amounts are not harmful.
• The results reported in this study relate only to the items tested and should not be interpreted as indicative for all bottles and liners from any one source. Note that Health Canada did not test all products available on the Canadian market.

Specific Observations and Conclusions of the Study of Non-polycarbonate Bottles

• In the lab, the bottles and liners were tested for the migration of BPA over various time periods using water and 10% ethanol as a food simulant.
• To measure BPA migration from bottles and liners, the sample was filled with boiling water (Table 1) or 10% ethanol at 85 degrees C (Table 2), then cooled to 60 degrees C and held at that temperature for 2, 22, 96, or 238 hours (approximately 2 hours, 1, 4, or 10 days). Even the least severe of these conditions (2 hours) is more severe than expected normal use. The liquid was then collected from the sample bottles and liners and the concentration of BPA was determined.
• Nine non-polycarbonate bottles and two polycarbonate bottles were tested with water as a food simulant (Table 1). Tests concluded that the two polycarbonate bottles contained higher levels of BPA, in comparison with the non-polycarbonate bottles. (Note: Neither of the two bottles that tested higher had claimed to be "BPA-free."). Ten non-polycarbonate bottles and liners were also tested using 10% ethanol as a food simulant to mimic low alcohol foods, as per the guidance to industry from the United States Food and Drug Administration (Table 2). The presence of alcohol changes the sensitivity of the test. In almost every case, there were no measurable amounts of BPA.
• The results for the two polycarbonate bottles were not included in the published study. For comparison, the results for all eleven bottles are shown below.
• The results for the ten bottle liners tested are also shown in the Tables 1 and 2.

The study concluded that at high temperature, migration of trace levels of BPA from some non-polycarbonate baby bottles and bottle liners was observed. However, these testing conditions were designed to represent a "worst case" scenario, and were significantly harsher than would be found in normal use.

Based on Health Canada's screening assessment of BPA, the trace levels detected in this study are much lower than those that could cause health effects. Health Canada wishes to reassure Canadians that at this time, the Department has no concerns with respect to the safety of baby bottles or bottle liners made from non-polycarbonate plastic.