1.

Interview of potential victims; all modes

-Victims can be identified with regularity through self-report.

-Nature and severity of the mistreatment acts as well as many antecedents can be characterized.

-Structured operational criteria for case definitions can be most rigorously applied.

-Some elder mistreatment victims will not admit mistreatment events for various reasons.

-Cognitive impairment may preclude accurate reporting.

-Interview setting in which perpetrators are present may deter reporting.

-Access to some elder mistreatment victims may be limited by sampling incompleteness, household refusals, language barriers, or illness.

-Not appropriate for study of fatal events.

2.

Interview of families, other caregivers, and others in trust relationships

-Perpetrators are known to relate some elder mistreatment events.

-Proxy respondents may report some elder mistreatment events more accurately, as well as some respondent characteristics (e.g., belligerent behavior).

-Structured, operational criteria can be uniformly applied.

-Many perpetrators may not report elder mistreatment events.

-Proxy respondents may have imperfect knowledge of events.

-Some proxy respondents may have illness, frailty, or cognitive impairment.

-Some characteristics of respondents can’t be obtained by proxy interview (e.g., emotional status).

3.

Clinical or social service institutional record review

-Level of objective documentation likely to be high.

-Treatments and social interventions documented.

-Historical data more likely to be obtainable.

-Only selected and more severe cases may be present.

-Case definitions may not be consonant with those of investigators.

-Clinical observations are often unstandardized in measures and notation.

-Ancillary and risk factor information may not be collected in a systematic manner.