appears to be the most sensitive method of detecting elder mistreatment in instances where older adults live with family members and suffer from significant cognitive impairment. Indeed, when older adults cannot serve as reliable historians or reporters of mistreatment, family caretaker assessment maybe the only available alternative. However, when cognitively impaired older adults reside in care facilities, the usefulness of caretaker assessment is less well established. This is because there are multiple care-taking staff for any single individual, the turnover rate of these staff is extremely high, precluding accurate long-term (i.e., multiyear retrospective) assessment, and the consequence of disclosure of abusive behavior is more immediately apparent (e.g., immediate suspension or termination).

Assessment of Older Adults

Epidemiological investigations with young adults and adolescents support direct interviewing of potential victim populations to determine the extent and character of mistreatment. It is logical to conclude that, for cognitively unimpaired older adults, direct assessment will also be useful. The following measures have been used with older adults. An additional interview methodology is proposed later.

The Hwalek-Sengstock Elder Abuse Screening Test (HSEAST) is a paper-and-pencil index of elder mistreatment with some psychometric evaluation. Neale et al. (1991) validated the 15-item screen and found that 9 of these items identified abused or exploited individuals. Items are scored yes or no, and a score of 3.5 or higher is indicative of abuse. Three domains of elder abuse are assessed: overt symptoms, victim risk characteristics, and victim symptom characteristics (the authors categorize these as violation of personal rights or direct abuse, characteristics of vulnerability, and potentially abusive situation). The test has some psychometric support of its construct, concurrent, and discriminant validity. The authors compared responses from 170 older adults who were agency referred with founded abuse cases with agency-referred nonfounded cases (n = 47) and a non-APS agency comparison group of elderly women (n = 47). Significantly higher scores were noted for the abused group, and item-level analysis indicated that nine items provided the basis for this difference. Discriminant function analyses of the nine relevant items revealed correct classification 74 percent of the time, with false negatives (35.7 percent) more likely than false positives (9.3 percent).

Advantages of the HSEAST include its preliminary psychometric validation, along with the fact that it is based on factor analysis of a large item pool. The test is able to assess risk factors along victim and situation lines and can facilitate direction or allocation of additional resources or assessment measures when risk is present. Although it assesses aspects of physi-



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