Important methodological and demographic differences between the studies, such as those involving age, injury severity, and seasonality, were ruled out as causes of these differences. Results found in the literature for other studies are compared and contrasted. These data suggest that recall periods of greater than 2 months are likely to significantly underestimate injury rates.

Possible intervention strategies are suggested, with emphasis on direct communication and active involvement on the part of the growers and harvesters.

Further development of a medical records-based surveillance system should include hospital emergency rooms and focus on identified health center performance factors.

There is a potentially large amount of toxic chemicals, some in decaying containers, that poses a risk to soil and groundwater in the state. The planning of future collections in New York State is haphazard. Carefully designed pesticide collection projects can be effective and should be a priority for local and state health officials.