COMMENT

Abstract

The first of these two papers (Dixon) describes an anlysis and study of nursery outbreaks. Thirty-one percent of the reports reached the CDC through the efforts of drug manufacturers. Some reports were solicited, others unsolicited and still others detected by means other than direct reports. Dr. Dixon and co-workers report statistical significance in the discontinuation of variable hexachlorophene bathing techniques between randomly selected hospitals and hospitals reporting staphylococcal nursery outbreaks. In the second paper, Kaslow and co-workers (Kaslow) analyze results from a survey of 266 of 296 randomly selected hospitals. In this study, infants who developed the disease after discharge from the nursery were excluded. Relative to the Academy-FDA warning in December 1971, staphylococcal disease rates rose sixfold in nurseries in which hexachlorophene bathing was discontinued while disease rates were unchanged in hospitals continuing hexachlorophene bathing. However, disease rates rose in hospitals when hexachlorophene was not used for bathing infants either before or after the warning. In neither of these papers, do the authors present data on the inverse relationship between hexachlorophene bathing in the nursery and disease caused by gram negative organisms. Authors of both papers address themselves to some of the uncontrolled variables inherent in these types of retrospective studies and caution against too firm conclusions based on their data alone. For example: (1) In the first paper, the CDC authors note that at least 31% of the reports of epidemics received by them were facilitated in some fashion by the manufacturers. They suggest that the conclusions of the first study are also limited by the fact that hospitals continuing or discontinuing hexachlorophene without recognized problems were not studied.