Dr. Merrill claims that almost any lung disease fits our definition of flock worker's lung and that we did not carefully describe steps to exclude alternative diagnoses. We disagree. If one too narrowly restricts the case definition of a new disease, patients with less common manifestations of the disease will be excluded inappropriately. Furthermore, our case definition required tissue confirmation of interstitial lung disease or, when no biopsy specimen was available (n = 1), comparably specific findings. Dr. Merrill also complains that we lumped together cases showing eosinophil-predominant bronchoalveolar lavage fluid with cases showing lymphocyte predominance. However, it is widely recognized that many interstitial lung diseases show varying cytologic patterns in lavage fluid and that such patterns are rarely diagnostic. Notably, a panel of blinded pulmonary pathologists, convened by the National Institute for Occupational Safety and Health, has recently substantiated our findings.

Furthermore, we feel that Dr. Merrill's confidence in masks and general steps to improve air quality in the plant, should hypersensitivity be involved, is misplaced. Workers have died of isocyanate-induced asthma while wearing respirators, and previously unexposed lifeguards have continued to develop granulomatous pneumonitis despite extensive efforts to improve working conditions (1). Lastly, we must be careful not to diminish the gravity of the fact that 60,000 Americans die of occupational illness each year (2). The real tragedy of 700 rock drillers dying of acute silicosis at a single West Virginia construction site (3), several thousand Americans dying annually of asbestos-related disease, 27% of Spanish textile dye sprayers developing bronchiolitis obliterans organizing pneumonia (4), and up to 65% of a municipal pool's lifeguards developing granulomatous lung disease (1) lies in the loss of human health and life rather than in the loss of jobs and profit. Disease cluster investigations, despite their limitations, will continue to foster the identification of new and important occupational diseases (5).

We thank Dr. Beckett for his interest and comments. It is worth noting that the National Institute for Occupational Safety and Health HHE team did not identify a single case of disease. Consequently, without our group's involvement, there could not have been an HHE report.

We appreciate Dr. Raymond's recommendation that physicians refer to their public health reporting responsibilities in service contracts with employers. Rather than referring to TSCA, which already carries the force of law and, moreover, is rarely applicable, we recommend referencing the codes of ethics of the American College of Occupational and Environmental Medicine and the International Commission on Occupational Health.