Mortality and Disability Outcomes of Elder Abuse
Elder abuse as measured using the VASS predicted dis-ability and mortality. The findings add to previous research by using longitudinal data collected from a broadly representative community-based sample over a 12-year period, and for the first time, a link between elder abuse and disability has been demonstrated. Furthermore, this study provides evidence of the effects of four different types of self-reported abuse on long-term health outcomes, unlike previous prospective research based on a single variable of reported or confirmed abuse or more-specific types of self-reported abuse such as physical and verbal abuse. These findings based on self-reported abuse are significant in light of growing calls for better research to inform prevention of elder abuse because effective prevention programs must be based on an accurate and efficient screening tool that is easily administered at the community level.
After controlling for a wide range of demographic, social support, health behavior, and health condition variables, coercion was the only form of abuse found to predict mortality. The most frequently endorsed item in this scale is "Has anyone taken things that belong to you without your OK?" suggesting that the factor may be aligned with financial abuse, although the second most frequently endorsed item, "forced … to do things you didn't want to do," suggests psychological abuse. In contrast, dejection, measured according to feeling sad and lonely often, feeling uncomfortable with someone in the family, and feeling unwanted, best predicted the risk of disability. Although the items suggest possible overlap with mental health, dejection remained significant after controlling for mental health and social isolation. Thus, the factor can be conceptualized as emotional abuse, although further research is warranted.
Although the data do not provide a clear explanatory mechanism for the link between risk of elder abuse and mortality and disability outcomes, it is significant that the VASS contributes additional explanatory power beyond that of demographic characteristics, including marital status and social interactions with friends, represented by the third step in the HR models. At this step, coercion and dejection remain significant predictors of mortality. Vulnerability and dejection also predicted disability at this step. The findings are suggestive of the importance of the quality of interpersonal relationships, in addition to the well-documented effect of social support. The current study points to a specific negative effect of abusive interpersonal relationships on mortality and disability in older women and suggests that subtle indicators such as "feeling uncomfortable with anyone in the family" or "feeling that nobody wants you around" have strong predictive ability for later disability.
This study extends the small existing literature on the association between elder abuse and risk of mortality and disability. First, it used a large cohort of more than 12,000 older Australian women who were broadly representative of the national population of women aged 70 to 75 in 1996. Second, the women were followed over 12 years, from 1996 to 2008, allowing for 12-year risk indices to be calculated. Finally, it is the first known study to examine the utility of a brief self-report screening survey of older people in predicting future mortality and disability risk. This compares with previous risk studies that have relied on reported or substantiated abuse records rather than self-reports of recent abusive experiences.
The study's reliance on self-reported measures of elder abuse is a strength and a weakness. Its strength lies in the ability to identify a greater number of potentially at-risk older people in a less-intrusive way, but self-report can also be criticized as subject to the cognitive functioning abilities of older people and their willingness to reveal sensitive interpersonal information. There was no independent verification of self-report, although it can also be argued that older people may be more willing to reveal such information through a confidential survey than face to face. In addition, some types of elder abuse, such as financial abuse, may be underestimated.
A second possible limitation is attrition in the sample over time, which means that subsequent measures of disability and vulnerability to elder abuse would be underidentified. In comparison, mortality was reliably ascertained through linkage with the National Death Index. Because women who had subsequent experience of elder abuse would be included in the group without abuse, the effects of elder abuse on subsequent disability and mortality would be diluted. Third, the items included on the VASS may overlap with cognitive disability and depression. Nevertheless, the results remained significant after controlling for mental health–related quality of life. Future research is needed to examine the components of the VASS in relation to definitions of elder abuse.
Discussion
Elder abuse as measured using the VASS predicted dis-ability and mortality. The findings add to previous research by using longitudinal data collected from a broadly representative community-based sample over a 12-year period, and for the first time, a link between elder abuse and disability has been demonstrated. Furthermore, this study provides evidence of the effects of four different types of self-reported abuse on long-term health outcomes, unlike previous prospective research based on a single variable of reported or confirmed abuse or more-specific types of self-reported abuse such as physical and verbal abuse. These findings based on self-reported abuse are significant in light of growing calls for better research to inform prevention of elder abuse because effective prevention programs must be based on an accurate and efficient screening tool that is easily administered at the community level.
After controlling for a wide range of demographic, social support, health behavior, and health condition variables, coercion was the only form of abuse found to predict mortality. The most frequently endorsed item in this scale is "Has anyone taken things that belong to you without your OK?" suggesting that the factor may be aligned with financial abuse, although the second most frequently endorsed item, "forced … to do things you didn't want to do," suggests psychological abuse. In contrast, dejection, measured according to feeling sad and lonely often, feeling uncomfortable with someone in the family, and feeling unwanted, best predicted the risk of disability. Although the items suggest possible overlap with mental health, dejection remained significant after controlling for mental health and social isolation. Thus, the factor can be conceptualized as emotional abuse, although further research is warranted.
Although the data do not provide a clear explanatory mechanism for the link between risk of elder abuse and mortality and disability outcomes, it is significant that the VASS contributes additional explanatory power beyond that of demographic characteristics, including marital status and social interactions with friends, represented by the third step in the HR models. At this step, coercion and dejection remain significant predictors of mortality. Vulnerability and dejection also predicted disability at this step. The findings are suggestive of the importance of the quality of interpersonal relationships, in addition to the well-documented effect of social support. The current study points to a specific negative effect of abusive interpersonal relationships on mortality and disability in older women and suggests that subtle indicators such as "feeling uncomfortable with anyone in the family" or "feeling that nobody wants you around" have strong predictive ability for later disability.
Contributions to Existing Literature
This study extends the small existing literature on the association between elder abuse and risk of mortality and disability. First, it used a large cohort of more than 12,000 older Australian women who were broadly representative of the national population of women aged 70 to 75 in 1996. Second, the women were followed over 12 years, from 1996 to 2008, allowing for 12-year risk indices to be calculated. Finally, it is the first known study to examine the utility of a brief self-report screening survey of older people in predicting future mortality and disability risk. This compares with previous risk studies that have relied on reported or substantiated abuse records rather than self-reports of recent abusive experiences.
Study Limitations
The study's reliance on self-reported measures of elder abuse is a strength and a weakness. Its strength lies in the ability to identify a greater number of potentially at-risk older people in a less-intrusive way, but self-report can also be criticized as subject to the cognitive functioning abilities of older people and their willingness to reveal sensitive interpersonal information. There was no independent verification of self-report, although it can also be argued that older people may be more willing to reveal such information through a confidential survey than face to face. In addition, some types of elder abuse, such as financial abuse, may be underestimated.
A second possible limitation is attrition in the sample over time, which means that subsequent measures of disability and vulnerability to elder abuse would be underidentified. In comparison, mortality was reliably ascertained through linkage with the National Death Index. Because women who had subsequent experience of elder abuse would be included in the group without abuse, the effects of elder abuse on subsequent disability and mortality would be diluted. Third, the items included on the VASS may overlap with cognitive disability and depression. Nevertheless, the results remained significant after controlling for mental health–related quality of life. Future research is needed to examine the components of the VASS in relation to definitions of elder abuse.
SHARE
