Caregiver Mental Health and Potentially Harmful Caregiving Behavior: The Central Role of Caregiver Anger

Gordon MacNeil, PhD; Jordan I. Kosberg, PhD; Daniel W. Durkin, MSW; W. Keith Dooley, PhD; Jamie DeCoster, PhD; Gail M. Williamson, PhD


Gerontologist. 2010;50(1):76-86. 

In This Article


As the demands facing caregivers increase, so too does their risk for providing compromised care (Dooley et al., 2007). It is our opinion that increases in caregiver anger and other negative psychological states over the course of caregiving may lead to increased risks for older care recipients. A better understanding of the influences of caregiver mental health characteristics on PHB enhances our ability to target preventive interventions at the precursors to adversity in the caregiving experience.

This study shows that the influences of anxiety, depression, and resentment on quality of care are largely determined by the degree to which caregivers also experience anger. For instance, we found that caregiver anxiety does not predict PHB in the absence of caregiver anger. This suggests that caregiver anxiety may not directly lead to PHB against care recipients. Rather, the risk of PHB among these "anxious caregivers" may be due to the amount of anger produced by the anxiety. One practice implication of this finding is that efforts should be directed toward reducing anger in anxious caregivers' efforts rather than toward reducing anxiety itself as a means of lessening the risk of PHB.

Depression and resentment directly influence PHB. Additionally, anger plays a significant role in the relations between depression and PHB and resentment and PHB. As anger increases, both depression and resentment appear to become more highly associated with PHB. Therefore, it is quite possible that managing anger can have positive effects on lessening the likelihood that depressed or resentful caregivers will engage in PHB. Our moderated mediation analyses suggest that as caregivers become more depressed or more resentful, anger is more likely to be associated with PHB. The practice implications of this are that identifying anger levels among caregivers who report symptoms of depression is warranted. Reducing depression in caregivers who report high levels of anger may result in reductions of PHB. Similarly, screening for resentment is warranted, as the relations between resentment and anger are similar to those between depression and anger.

Our findings have important implications for the identification of individuals who are appropriate or, perhaps more importantly, who are inappropriate for undertaking the caregiver role for a frail or dependent older person. In short, persons who are angry about their caregiving responsibilities may not provide adequate care to their elderly family members. When anger is coupled with moderate or high levels of depression or resentment, the levels of PHB increase significantly, suggesting the need for additional assessment of the caregiving experience. Caregivers demonstrating high levels of anger in addition to depression or resentment resulting from caregiving responsibilities cannot be considered good candidates as care providers as their care recipients are more vulnerable to the possibility of poor quality care.

The stress that these caregivers experience may emanate from many sources other than their caregiving responsibilities. Thus, professionals need to be sensitive not only to emotions related to the older care recipient but also to any existing tension, pressure, or resentment among caregivers. It seems that regardless of the causes of these emotions in the caregiver, the vulnerability of the older person to the possibility of harmful behavior may be high. Clearly, pragmatic concerns preclude a person's exclusion from a caregiving role, even with elevated anger levels, but such circumstances should raise "red flags" for health and social service providers that additional assessment and intervention should be considered.

There is a clear need for supportive resources and interventions for caregivers experiencing anger as well as depression and resentment. In order to ensure the adequacy and quality of care being provided to vulnerable older persons, comprehensive assessments of caregivers for anger, depression, and resentment need to be periodically conducted. Interventions, including respite services, anger management training, cognitive reframing (for resentment), and increasing pleasant events, should be provided to address these conditions.

Limitations and Future Research Efforts

We recognize that in order to confirm the existence of mediation, it is necessary that both the effects of predictors on an outcome are related to the mediator and that evidence exists of a logically expected causal order (Baron & Kenny, 1986). The current study is cross sectional, and we suggest that logical arguments for the causal ordering of these variables have been presented elsewhere (Williamson et al., 2001).

Many studies of elderly abuse use known-group samples of individuals identified as perpetrators of these behaviors or offenses. Much of the literature on caregiving with older populations has focused specific caregiving systems, such as families in which dementia and terminal illnesses prompt caregiving. Our sample, on the other hand, included a wider focus of informal caregivers of older persons with physical, mental, and/or cognitive problems. Although we are confident that our sample is representative of a significant portion of caregivers of older persons in America, the extent to which our findings are generalizable to caregivers of older persons in other locations is seen to be limited. Subsequent studies that include broader samples are needed.

A number of additional research projects are suggested by this study. Although our sample was obtained from multiple sites, larger studies with broader representation of informal caregivers are needed. Studies employing measures that specifically capture caregiving-related depression and anger would be informative. More studies are needed to further our knowledge of how anxiety and resentment relate to caregiving. As our study relied on cross-sectional data, longitudinal data that examines how the relations we found change over time would be enlightening. Finally, the moderated mediation analysis we conducted is new, and further applications may suggest more important nuances and refinements allowing us to better understand and use this statistical test.

There is much more to know about the relations between mental health variables, including anger and PHB. The exploration of this topic is still in its early stages, but the benefits of continuing the search are great: ensuring the quality of care for older and impaired relatives by members of their own families.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: