The objective of this study was to determine how often and how children <5 y-of-age involved in unintentional poisonings gained access to those poisons by opening properly secured child-resistant packages. From June 1994 through February 1995 we surveyed caretakers of children <5 y-of-age presenting with the chief complaint of ingestion, inhalation, or ocular or dermal exposure of potential or perceived toxins. Excluded were exposures to plants, bites or stings, or patients inadvertently overdosed by a caretaker. The survey was administered to caretakers of 168 patients. Mean age was 26 mo (range 9 mo to 56.9 mo). The toxin was in its original container in 71% and transferred to another container or found outside of its container in 29% of cases; 33% involved a child-resistant closure. The child gained access by opening a properly closed child-resistant closure in 20% of the exposures and in 18.5% by opening a properly closed non-child-resistant closure. Child- resistance did not ensure child impenetrability. This suggests the need to reevaluate child-resistant closure testing protocols or retest currently used closures. Packaging misuse, including improper closure use or transferring medications or household products from the original container accounts for a large percentage of pediatric toxin exposures, thus emphasizing the importance of poison prevention education.