While the logic of this approach is appealing, the ability to assess accurately child maltreatment potential is seriously limited by problems in methodological adequacy, definitional ambiguity in the criteria for child abuse and neglect, the mathematics of predicting low base-rate phenomena, and the financial and social costs. Risk assessments to identify populations in need of preventive programming have yielded very small increases in programming efficiency. Even this small benefit is likely to be eroded if the base rate for abuse and neglect declines, as evidence suggests it has. Further, as prevention efforts become more successful, the base rate will further decrease, additionally decreasing the utility of risk assessment. The alternative strategy is to deliver preventive services to the general population without regard to risk status. This strategy eliminates the stigma associated with being singled out for special services and ensures that no person requiring intervention is denied it. Interventions that improve the quality of parent-child interactions, support single mothers, or teach children how to avoid abusive interactions may be helpful whether or not participants are at risk. 95 references. (Author abstract modified)