Allergies among animal handlers to their subjects is a problem of high prevalence and relatively low morbidity but with often serious career implications. This study sought to determine whether historical factors, such as general allergic symptoms and atopic indicators, could be used to predict which animal handlers would be affected with allergies to laboratory animals (ALA). In a prospective study, 169 laboratory workers with animal exposure were surveyed by questionnaire for these factors in 1985 and again 2 years later. The presence of three general allergic symptoms (relative risk = 4.29) and three historical atopic indicators (relative risk = 2.50) were moderately predictive of the new onset of ALA. This suggests that those animals handlers with allergic histories are at greater risk of developing ALA, although this is a relatively nonspecific marker. In a concurrent prevalence study, the measurement of airborne rat urinary protein was significantly associated with the presence of ALA (prevalence ratio = 1.75). The measurement of airborne allergens could be helpful in monitoring the effectiveness of air handling equipment with the ultimate goal of reducing the incidence of ALA.