Sweden and Denmark, Tornstam (1989) queried respondents about whether they had observed or knew about specific cases of persons who had been physically battered, threatened, economically abused, robbed, or severely neglected. The overall rate defined this way was 8 percent, but most cases were due to a single incidence of theft. The definitions were not always consistent with those used in other elder mistreatment studies, but most importantly, the individual respondent was not the unit of analysis.

Podnieks (1992) reported the findings from a representative telephone survey of Canadians age 65 and older. Domains included physical abuse, neglect, psychological abuse, and “material abuse.” The overall prevalence rate was about 4 percent, but this was a cumulative experience since age 65, so the annual rates would be difficult to calculate; the most common form was material abuse (2.6 percent). Ogg (1993) attempted to repeat the Pillemer and Finkelhor survey in London but for methodological reasons was unable to obtain credible occurrence information.

Thus, based on the published, peer-reviewed literature and some efforts at obtaining unindexed, non-peer-reviewed studies, there appears to be little population-based information about elder mistreatment occurrence, including the clinical course and outcomes of proven events. It appears that more population-based approaches to elder mistreatment, including nationally representative samples, are needed. Even less information is known about elder mistreatment occurrence in institutional settings. Event detection is extremely challenging and to date only indirect approaches have been employed to make estimates.

The paper by Acierno (this volume) summarizes the primary ways in which elder mistreatment has been detected: (a) direct interview surveys of potential victims by telephone, personal interview, or self-administered questionnaire), (b) interviews of families or caregivers of possible victims or others with a trust relationship, (c) clinical or social service institutional record review, (d) placement of sentinel reporters within these agencies or organizations, and (e) acquisition of criminal justice information. Historically, these techniques have been applied most consistently and will be likely to continue to be important. Corder (2001) reviews many of the issues in population sampling and surveying relevant to household assessment of elder mistreatment occurrences. A summary of overall strengths and weaknesses of these approaches is shown in Table 4-1.

However, other, less explored methods for identifying elder mistreatment cases are available for research evaluation: (a) a two-stage process, beginning with screening potential victims for risk factors or risk indicators, using questionnaires, medical record review, and various biomarkers, with subsequent more intensive evaluation of high-risk persons, (b) screening fiscal records for types of behaviors associated with financial abuse, (c)

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement