expanded its data sources from APS reports to include trained sentinel reports of substantiated or presumed substantiated cases. The NEAIS-targeted people living in their own homes, age 60 and above. This was an incidence study (new cases during a set time frame), not a prevalence study. Importantly, this study did not interview older adults themselves. Sentinels were professionals who served older adults and were randomly selected from more than 200 agencies. Sentinels were trained to complete data entry forms identical to those used by APS workers for elder abuse. The logic of the sentinel approach is based on the supposition that sentinels enhance sensitivity by detecting those older adult victims of abuse who are nonreporters or are not involved with APS but who nonetheless interact in some way with community-based service agencies. NEAIS data were gathered on domestic (i.e., noninstitutionalized) elder abuse and neglect cases from a nationally representative sample of 20 counties in 15 states. Reports from APS agencies were considered only when substantiated and reports from sentinels were presumed to be substantiated.
The methodology of combining agency record reviews with sentinel reports to detect mistreatment has previously demonstrated success in three studies of child abuse. Moreover, the method is cost-effective, and identified cases are very likely true positives. Multiple data sources are consulted, and these typically have a very thorough familiarity with cases. Finally, multiple forms of mistreatment are identifiable.
Weaknesses of this method include the fact that no direct assessment is made of the population in question. Thus, it is very likely that mistreatment rates derived from this study greatly underestimate the true scope of the problem of elder victimization because a great majority of cases go both unreported and undetected by existing formal and informal monitoring agents. Although this approach has been used three times with child abuse, there are several problems with this method when applied to elder abuse. First, and perhaps most relevant, is the fact that child abuse reporting statutes and subsequent education of an extremely wide range of service providers (e.g., schoolteachers, doctors, nurses, counselors, day-care workers, etc.) regarding these statutes is formally established and mature. That is, awareness of the problem of child abuse is far greater among the general and professional public, and thus sentinels in the child arena will be more familiar with the problem and its symptoms. Moreover, child abuse mandatory reporting and provisions for anonymous voluntary reporting have been in place nationwide and have been accompanied by national education campaigns. As such, it is likely that child protective services receive a significantly larger proportion of existing cases than APS. Indeed, mandatory reporting of elder abuse is not consistent across the nation and is still actively debated. National education campaigns for the general public and for health and social service providers on child abuse also increase the