and culturally diverse populations. Also, much more work needs to be done on how elder mistreatment victims are detected and managed in various health care systems and in communities with varying levels of long-term care and adult protective service availability. Several papers in this volume note the use of qualitative techniques to further define the various elder mistreatment “syndromes” and characteristics; further application of these methods would seem to be of value. Finally, very little is known about elder mistreatment occurrence and related risk factor status among minority populations in the United States, including cultural variation in how mistreatment is defined and perceived. In general, a more diversified approach to research on risk factor and occurrence assessment would achieve several ends: (a) more critically defining populations with higher and lower occurrence rates, (b) determining the generalizability of putative elder mistreatment risk factor findings across such diverse populations, (c) more precisely providing sample size estimates for intervention studies within these populations, and (d) exploiting cross-cultural variation in elder mistreatment occurrence to better understand its causes.

Piggybacking Assessment Modules on Existing Population Surveys

One way to promote research on elder mistreatment occurrence is to add detection items and instruments to existing field surveys, particularly those that cover large geographic areas or are national in scope. This is discussed extensively by Corder (2001).

On one hand, this could provide several potential advantages: it may allow substantial resource savings when compared with conducting surveys de novo; national estimation of elder mistreatment rates could be substantially enhanced; existing surveys may contain important respondent and family health, social, and economic variables that can be explored as both risk and outcome variables; and some surveys may have longitudinal data collection, allowing a time dimension not otherwise available in cross-sectional surveys.

On the other hand, there may also be important limitations to this approach: sensitive assessment of elder mistreatment may not lend itself to certain modes of data collection, such as mail or telephone surveys; elder mistreatment themes may not be compatible with the other survey content; there may be different requirements and challenges in the use of proxy respondents; there may personal respondent resistance to items related to elder mistreatment; certain demographic or cultural groups may not be adequately represented in the parent surveys of interest; and content and sampling techniques may be unsuitable for many elder mistreatment scientific questions of interest. There may also be limitations on identifying or following up respondents, should substantial evidence of elder mistreat-

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