Hospital-based surveillance for acute pesticide poisoning caused by neurotoxic and other pesticides in Tanzania.

BACKGROUND AND AIM: Acute pesticide poisoning (APP), particularly with neurotoxic agents, is often under-reported in developing countries. This study aimed to estimate the burden of APP in Tanzania due to neurotoxic and other pesticides in order to propose a surveillance system.

METHODS

The study reviewed hospital admission data for APP retrospectively (2000-2005) in 30 facilities in four regions of Tanzania. A prospective follow-up over 12 months in 2006 focused on 10 facilities with the highest reporting of APP.

RESULTS

The majority of known poisoning agents were organophosphates or WHO class I and II pesticides. APP involving suicide was significantly more likely to be fatal in both retrospective (PRR fatal/non-fatal=3.8; 95% CI=1.8-8.0) and in prospective (PRR=8.7; 95% CI=1.1-65) studies. There was a significant association between suicide and gender (PRR female/male=1.5; 95% CI=1.1-2.0) in the prospective study. Occupational circumstances as a cause of APP, which was relatively small in both studies (8.5% in the retrospective and 10.2% in the prospective study) was less common amongst men compared to women (6.1% for males versus 12.0% for females) in the retrospective study but almost equal in prospective study (10.2% for males versus 10.1% for females). Contrasting retrospective to prospective studies, the annual incidence rate almost tripled (from 1.43 to 4.05 per 100,000) and mortality rate doubled (from 0.11 to 0.22 per 100,000). Case fatality declined accordingly from 7.8% to 5.6% in prospective study. The study revealed a substantial improvement in the completeness of data with prospective data collection. Missing data for circumstances and agents declined by 24.1% and 9.9%, respectively. Despite this improvement, routine reporting could only generate 33-50% of the information needed for a notification of banned or severely restricted chemicals under the Prior Informed Consent (PIC) convention.

CONCLUSION

The two to threefold increase in rates with prospective data collection suggests significant under-reporting of APP by neurotoxic and other pesticides. Routine reporting is likely to under-estimate the burden from pesticides, particularly for women in occupational settings. The burden of APP and the specific pesticides causing serious problems in Tanzania would continue to be missed without improved surveillance systems.