In September 2000 the Washington State Department of Health (DOH) was awarded a grant from the National Institute for Occupational Safety and Health (NIOSH) to enhance their pesticide surveillance system. The goal of this effort was to increase the value and effectiveness of Washington State's Pesticide Incident Monitoring Surveillance system (PIMS) by examining and improving the quality of data. One year of case reports for pesticide illnesses among agricultural workers was examined to ascertain the sources of case reports as well as their timeliness and completeness. Only 13% of these cases were reported to PIMS by more than one source. The vast majority (75%) were identified through the analysis of worker compensation claims, however the average time lag between the incident and the report to DOH was significantly greater than for other report sources (>25 days vs. <5 days). Case reports received more than 25 days after the incident were also significantly more likely to be classified as "suspicious" or "insufficient information." Overall nearly 80% of the case reports were received as a result of a visit to a health care provider (e.g., direct report, indirect report through Workers' Compensation claim). A series of six focus groups of current or former farmworkers in Yakima County, WA, were conducted to explore whether they sought health care when they felt they were ill due to a pesticide exposure while on the job. The majority indicated that they would not seek care for mild to moderate symptoms. About three-fourths indicated that they had experienced illnesses that they ascribed to pesticide exposure. Economic factors were the main reasons for not seeking health care. These included loss of wages while at the doctor, loss of job, and the cost of appointments and medications. Many also felt a reluctance to use the local health care system for pesticide illness, even though they often used and were satisfied with these providers for other health care needs. Many expressed a sense that doctors did not believe them when they said that an illness was due to pesticide exposure. Three years of data from the eight major hospitals and clinics in Yakima County were matched to data in PIMS to determine the proportion of visits for pesticide-related illness that were captured by PIMS. Of the 169 encounters which had been assigned a diagnosis code that explicitly referred to a toxic effect of pesticides (i.e. 989.0 - 989.4) or to pesticide exposure as an external cause of injury (i.e., E863.0 - E863.9), 116 appeared to be initial visits. Only 42 of these 116 records were found in PIMS. Medical charts for the remaining 74 cases were requested and abstracted. Of the fifty-five cases involving agricultural workers, 30 had been captured by PIMS (54.5%). In three cases, there was a note in the medical record indicating that the case had been reported to either PIMS or WPC. With this adjustment, PIMS should have captured 60% of the eligible cases among agricultural workers. These results were reviewed with staff from the hospitals and clinics which had contributed data to better understand reporting and coding practices. There was confusion regarding the reporting requirements and definition of pesticide-related illness. Reporting pesticide-related illness was felt to be difficult in clinical and emergency department settings due to the time needed to report and the confusion regarding how and what to report. Further, clinicians and coders reported that they generally do not assign an ICD diagnosis code specific to pesticide poisoning if there is uncertainty that the observed signs and symptoms are specifically a result of pesticide exposure. A series of interviews were conducted with representatives from a variety of stakeholder groups to assess their perceptions of how well the pesticide surveillance program was working and to solicit input on how the value of the information generated could be improved. Most stakeholders were generally satisfied with the pesticide surveillance system. The primary concern of stakeholders and users of PIMS data was the timeliness of the annual reports. While many stakeholders reported using the data, few specific data uses were cited. There were few requests for specific, new analyses. While exposure to pesticides continues to be an important issue to farmworkers, farmworkers in our focus groups were also concerned about a lack of hand washing facilities in the field, the condition of sanitation facilities in the field, the lack of PPE for farmworkers who are not directly mixing, loading or applying pesticides, and the small proportion of employers who they considered to be 'bad actors' and who they thought did not follow health and safety regulations. Based in part on these results, DOH has begun to implement electronic reporting of pesticide case reports to DOH from the Washington Poison Control and the Department of Labor and Industries workers' compensation database. While this will improve the timeliness of the case reports, and is expected to reduce the proportion of cases which receive a classification of "suspicious" or "insufficient information," it will only partially address the problem of under-reporting.
Washington State Department of Health, Office of Environmental Health Assessments, 7171 Cleanwater Lane, Building 2, P.O. Box 47846, Olympia, WA 98504-7846