The purpose of the completed project was to follow the first National Elder Mistreatment Study, which provided prevalence estimates, with a second study of a subset of the same participants to measure the effects of elder abuse in terms of (1) health and mental health outcomes and (2) criminal justice system participation and satisfaction, as well as to specify additional predictors of these effects.
The original sample of 5,777 adults age 60 and above was collected during 2008 by the AbtSRBI survey research firm and derived using stratified random digit dialing with an area probability sample based on Census-defined 'size of place' parameters with the continental US serving as the sampling location. Interviews were conducted in either English or Spanish. Standardized computer assisted telephone interviewing (CATI) procedures were used to query participants about a variety of mistreatment experiences, potential correlates, and demographics. The NEMS Wave I cooperation rate was 69%, and was calculated according to the American Association for Public Opinion Research (AAPOR) Rate # 2 as the number of completed interviews, including those that screen out as ineligible, divided by the total number of completed interviews, terminated interviews, and refusals to interview. The follow-up NEMS Wave II was also collected by CATI in 2016 under the direction of AbtSRBI. An auto-dialing program tested each phone number used in NEMS Wave I to test whether it was still in operation. This yielded 3,973 operative phone numbers of the original 5,777, including 752 participant numbers who reported being the victim of psychological, physical, or sexual mistreatment since age 60 at Wave I. Note that because these were land lines, an operative number may have been assigned to another household in the intervening 8 years, or respondents may no longer live in the household. Thus, reaching an 'operative number' is not necessarily the same as reaching a 'participant', a relevant factor when calculating cooperation rates (see below). Because the mistreated group was more limited in size, contact attempts were made for all working phone numbers of participants reporting mistreatment at NEMS Wave I. In addition, a randomly selected subsample of operative phone numbers from Wave I comparison participants who did not report Wave I mistreatment were also called (N = 2,149). As with the NEMS Wave I, the cooperation rate was calculated using the AAPOR Rate #2. Note, screen-outs occurred when individuals answered an operative number but disclosed that the original participant was no longer available for participation due to unknown relocation or death; thus, screen outs are not refusals. Among those operative numbers dialed where participants were available, cooperation rates were 66% for the NEMS Wave I since age 60 mistreated group (N = 183) and 57% for the comparison group (N = 591), for a total follow-up NEMS Wave II sample of 774.
Data were collected from 774 older adults 8 years following their participation in Wave I of the NEMS. This represented the results of contacting every locatable participant who reported psychological, physical, or sexual (but not financial) abuse at Wave I (achieved subsample n = 183 of the original 753 Wave I victims) and a comparison sample of 591 randomly selected Wave I non-victims from the remaining 2,149 working phone numbers of the original 5,024 non-victims (at Wave I).