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

Abstract and Introduction


Purpose: Caregivers feeling stress and experiencing mental health problems can be at risk for engaging in abusive acts against elderly care recipients. Potentially harmful behavior (PHB) was used as a measure of caregivers' engagement in, or fear of engagement in, behavior that places dependent care recipients at risk of physical and/or psychological maltreatment and may be seen as an antecedent of, or a proxy for, identifiably abusive behavior. The study examined the ability of anger to mediate and moderate the relations of depression, resentment, and anxiety with PBH.
Design and Methods: Data are from the first wave of the second Family Relationships in Late Life study of caregivers of community-dwelling elderly care recipients with whom they coreside. Caregivers (N = 417) completed face-to-face interviews.
Results: Anger was found to mediate the relation between anxiety and PHB. Anger both mediates and moderates the relations of both depression and resentment with PHB in a dynamic way such that the mediating effect of anger increases substantially with increased scores on both depression and resentment.
Implications: 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. Screening for resentment is warranted, as the relation between resentment and anger is similar to that between depression and anger.


Family members represent the "front line" of care provision for the growing number of ill, disabled, and demented older adults residing in the community. Unfortunately, many of these informal caregivers experience considerable stress and are at higher risk for poorer mental and physical well-being than noncaregivers (Bookwala, Yee, & Schulz, 2000). This article reports on an investigation of the mental health challenges to caregivers that can result in compromised care provision to others.

Several studies have linked caregiver depressive symptomatology with the quality of elderly care (Beach, Schulz, Williamson, Miller, & Weiner, 2005; Williamson, Shaffer, & The Family Relationships in Later Life Project [FRILL], 2001). Depressed caregivers are more likely than nondepressed caregivers to be subjects of investigation by social service agencies for possible elderly mistreatment (Bonnie & Wallace, 2003; Wolf, 1996). More depressive symptoms and feelings of resentment among informal caregivers are associated with caregiver self-reports of potentially harmful behavior (PHB), a precursor to elderly mistreatment (Beach et al.; Williamson et al., 2001, 2005). The association between depression and quality of care holds even after controlling for background characteristics and previous mental health status (Williamson et al., 2001). Conversely, care described as high in quality by caregivers and/or their care recipients is more common among caregivers who express fewer depressive symptoms (Dooley, Shaffer, Lance, & Williamson, 2007). Although the effects of depression are well documented in the caregiving literature, not all informal caregivers experience serious or debilitating symptoms of depression, and other negative affective reactions are possible.

Another frequent reaction to the burdens of care provision is the harboring of resentment. For example, caregivers may resent care recipient dependency or the responsibility associated with ensuring the daily welfare of another adult (Williamson, Shaffer, & Schulz, 1998; Williamson et al., 2001). Caregivers also may resent care recipient behavior that makes helping more difficult. In turn, more resentment is associated with lower quality of informal care (Williamson, Shaffer, & FRILL, 2000). Resentful caregivers may be more likely to harbor ill feelings toward their care recipients, leading to tension and conflict in the relationship (Schofield, Murphy, Herrman, Bloch, & Singh, 1997). These circumstances may lead to intense feelings of anger.

The emotional problem of anxiety has not been as extensively examined in caregiving relations as has depression, despite the fact that anxiety disorders may be more prevalent in caregivers than depression among caregivers (Mahoney, Regan, Katona, & Livingston, 2005). In examining the prevalence and covariates of anxiety in dementia caregivers, Cooper, Balamurali, and Livingston (2007) note that about a quarter of these caregivers demonstrate clinically significant symptoms of anxiety. In reviewing 33 caregiver studies, the authors found that along with physical health and caregiver burden, coping strategies of confrontation or avoidance were indicators of increased anxiety. They noted that although burden and physical health covariates were similar to those of caregiver depression, coping styles were more likely to be associated with anxiety than with depression. Others have found that although most depressed caregivers are also anxious, the inverse is not the case (Mahoney et al.). Neither care recipient cognitive status, degree of activities of daily living impairment, caregiver age, nor duration of caring were associated with caregiver anxiety (Cooper et al.). Given that predictors of depression may not be the same as those of anxiety, there is a need to consider anxiety as a separate outcome measure.

Pinquart and Sorensen's (2003) systematic review of interventions for caregivers failed to include attention to anxiety, whereas Schulz and colleagues (2002) included a few studies with anxiety as a characteristic of interest. Mehta and colleagues (2003) noted that interventions that reduce depression may have little effect on reducing anxiety and concluded that it is useful to consider anxiety as a separate outcome.

Although efforts have been made to examine the relations among depression, resentment, and care provision, the consequences of anger on the delivery of informal elderly care have received relatively little research consideration. Caregivers may be less likely to acknowledge their anger in research or social service interviews because of shame at having these feelings or fear of social or legal scrutiny (Gallagher-Thompson, Lovett, & Rose, 1991). Yet, some caregivers do report becoming angry or frustrated in certain caregiving circumstances (Williamson et al., 1998; Zarit, Stephens, Townsend, & Green, 1998), such as when a care recipient exhibits behavioral problems (e.g., Kramer, Gibson, & Teri, 1992).

There is considerable interindividual variability in the frequency of anger and the manner in which anger is managed and expressed. For example, individuals prone to experiencing shame are more likely to feel resentment and/or anger, blame others for negative events, or act aggressively toward others in the face of interpersonal conflict. Individuals prone to experiencing guilt are more likely to adopt more constructive and prosocial conflict resolution strategies (Tangney, Wagner, Fletcher, & Gramzow, 1992; Tangney, Wagner, HillBarlow, Marschall, & Gramzow, 1996). There also is evidence of congruity of anger in close relationships such that when one partner expresses negative emotions and the other partner also experiences negative emotions in the same or different domains (Lane & Hobfoll, 1992). Spouses of depressed individuals commonly report depression, dissatisfaction, or anger (Bookwala & Schulz, 1996; Druley, Stephens, Martire, Ennis, & Wojno, 2003). Averill (1982) has identified several behavioral reactions to anger, including direct aggression, indirect aggression, displaced aggression, and nonaggressive responses.

Some caregiving literature gives attention to adverse consequences of caregiving stress on the quality of care given to older persons as well as to the coping strategies of caregivers. Coping strategies refer to specific behavioral and psychological efforts to master, tolerate, or minimize stressful events (Lazarus & Folkman, 1984). The use of one type rather than another is determined by one's personal style as well as the nature of the stressful event (Folkman & Lazarus, 1988). When caregiving is long term, the predominant methods for dealing with stress appear to be emotion-focused coping strategies (Pruchno & Resch, 1989). Kramer (1993) found emotion-focused coping to be a significant predictor of depression for wives caring for husbands with dementia. Carver and Scheier (1994) identified coping mechanisms that they believe to be dysfunctional: denial, mental disengagement, behavioral disengagement, and the use of alcohol. Thus, coping styles may affect caregiver mental health outcomes.

In a study identifying predictors of elderly abuse and neglect, caregivers who reported physical abuse displayed higher depression scores, whereas those who reported neglect reported higher anxiety scores (Reay & Browne, 2001). A study by Gallagher, Rose, Rivera, Lovett, and Thompson (1989) reported that approximately two thirds of the elderly caregivers reported feeling angry or engaging in angry behavior toward their care recipient; many reported sufficient concern that their anger led them to seek formal services.

A National Research Council Report (Bonnie & Wallace, 2003) on elderly mistreatment concluded that depression is highly characteristic of perpetrators of elderly abuse. Williamson, Shaffer, & FRILL (2001) found that higher depression scores were predictive of more frequent occurrences of PHB, regardless of demographics, interpersonal relationships, and care recipient illness severity. Further analyses of the relationship between depression, resentment, and proactively aggressive caregiving strategies indicates that resentment mediated the impact of depression, suggesting that care recipients are at increased risk for PHB particularly when their depressed caregivers are resentful about their caregiving responsibilities (Shaffer, Dooley, & Williamson, 2007)

Thus, there is evidence that the quality of care given to an older person is affected by the existence of caregiver depression, anxiety, and resentment. What is not clearly understood, however, is the specific relationship between caregiver anger and anxiety, depression, and resentment. Given the important effect resentment has on depression in caregivers, examining the relationship between anger and depression, resentment, and anxiety is warranted.


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: