tion of the clinical syndrome and reports of various series of elder mistreatment victims, usually from geriatrically oriented facilities or programs managing chronic illnesses in older persons, and later from social service agencies. Gradually, recommended elder mistreatment definitions and criteria began to appear, and many of these recommendations have been published (see Chapter 2). Often, the basic demographic and clinical characteristics of elder mistreatment patients were defined in part by the nature of the study facilities as well by the patients/victims. Early clinical descriptions paved the way for later studies of elder mistreatment risk factors. Using case records from public health nurses, Phillips (1983) reported one of the first case-control studies of risk factors for elder abuse.
Estimates of the occurrence of abuse and neglect have varied from about 2–10 percent annual incidence, although the bases for these estimates are modest and uncertain (Branch, 2001). The issue of incidence versus prevalence and the recurrent nature of the problem among individual victims and other issues (discussed below) make these estimates very insecure. For example, Thomas (2000) reviewed both formally published and other data on elder mistreatment occurrence. The lack of population-based studies in this review is clear, and much of the information reviewed came from institutional and social service agency sources. Other than representative household samples, some research has explored samples of adult protective service workers, assessing their observations and experiences (Dolan and Blakeley, 1989) and other public and private institutional employees who may have contact with abused elders, such as police authorities, hospital personnel, and bank employees (National Center on Elder Abuse). While important, these approaches identify only those (potential or actual) elder mistreatment victims who have come to public attention and probably underestimate the true elder mistreatment occurrence rates.
One of the first and historically most important population-based studies of elder mistreatment was conducted by Pillemer and Finkelhor (1988). This was a prevalence study using a probability sample of noninstitutionalized persons age 65 and older residing in metropolitan Boston. Interviews were conducted over the telephone and in person using structured questionnaires and standardized criteria for three domains of elder mistreatment: physical abuse, psychological abuse, and neglect. About 72 percent of the eligible respondents were interviewed. Including all three elder mistreatment domains studied, they reported an overall rate of 3.2 percent.
A few other population-based studies have been published. Comijs et al. (1998) studied physical and psychological abuse in a cohort of Dutch elders in Amsterdam, using structured interviewing techniques. Overall, the one-year prevalence of elder mistreatment was 5.6 percent, with verbal aggression being the most common; the prevalence of physical aggression was 1.2 percent. In a telephone interview study of national samples from