Wildlife as sentinels for human health hazards: a review of study designs

Abstract

A systematic search of the peer-reviewed biomedical literature was performed for original studies linking environmental hazards with health effects in wild animal populations. Some 104 original studies were identified in the BIOSIS and Medline databases since 1966 and classified in terms of study design. A marked increase in published studies has occurred over the past decade, examining a wide range of hazards and outcomes. Most analytic studies were ecologic or cross-sectional in nature. All cross-sectional studies sampled subjects on the basis of exposure, using a reference population for the selection of controls. Studies of wild animal populations may hold unique advantages for toxicant hazard identification, yet the current range of study designs appears restricted. Increased use of study approaches such as sampling based on outcome, intra-population comparisons, as well as cohort and case-control designs may improve hazard identification and priority setting for confirmatory toxicologic and human studies of effects seen in wildlife. Copyright © 1999 John Wiley & Sons, Ltd.