Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jan 1.
Published in final edited form as: J Fam Violence. 2017 Oct 17;33(1):17–26. doi: 10.1007/s10896-017-9941-4

Functional Decline and Emotional Elder Abuse: A Population-Based Study of Older Korean Adults

Jooyoung Kong 1, Haesang Jeon 2,✉
PMCID: PMC6028192  NIHMSID: NIHMS936643  PMID: 29977104

Abstract

Objectives

Elder abuse is an increasingly prevalent issue in South Korea. The current study examines the association between functional impairment and emotional abuse victimization in Korean adults 65 and older. We also examines the mediating roles of diverse aspects of family resources (i.e., older adult's self-esteem, family cohesion, family assistance, contact with friends/neighbors, and participation in social activities) in the aforementioned association.

Methods

We analyzed 9,691 community-dwelling older Koreans from a population-based survey of the 2009 Survey of Elderly Care and Welfare Need. We conducted a path analysis to analyze the mediational hypothesis.

Results

About 11% of the study sample reported experiencing emotional abuse by a family member in the past year (n = 1,082). The results of the mediational analysis showed that greater ADL/IADL limitations were associated with a) reduced self-esteem and b) receiving more assistance from family, which were ultimately associated with increased risk of emotional abuse victimization.

Discussion

Elder abuse is a family crisis that may occur as a result of maladaptation to the heightened long-term care needs of older family members. Practitioners should take into account older adults' vulnerability in terms of the loss of adaptive resources in the face of functional decline. This study further supports the importance of relieving the burden of family caregivers to avoid the incidence of emotional elder abuse.

Keywords: Elder victimization, family stress theory, self-esteem, family cohesion

Emotional elder abuse has become a prevalent social issue in South Korea (hereafter Korea). Recent national estimates indicate that 9.9% of Korean adults aged 65 and older (i.e., approximately 0.7 million individuals) experienced elder abuse, and more than 70% of the abuse cases involved emotional abuse that is defined as an act that causes emotional pain or distress (Korean Institute for Health and Social Affairs (KIHSA), 2014). In addition, emotional abuse against older Korean adults has increased sharply over the past decade as reflected in the number of reported cases to the elder protective services from 1,499 in 2005 to 2,330 in 2015 (Korea Elder Protection Agency(KEPA), 2016). More importantly, we noted that the majority of the reported elder abuse cases happened at home (i.e., 86% in 2015; KIHSA, 2014), which alerts us that the traditional value of filial piety may have been challenged (Doe, Han, & McCaslin, 2009). In the meantime, public policies to address the extended burdens of care giving due to prolonged longevity have not been kept up with these changes in family norms and rules (Kwon, 2004), which could potentially accelerate violence against older family members. Despite the importance of the issue, the following research gaps have been identified: (a) a paucity of research exists concerning specific mechanisms of how older Korean adults are victimized within families; and (b) much of research on elder abuse in Korea relied on small, conveniently collected samples, which limits the robustness of study findings and their interpretations for implications. To address this research gap, we an alyzeda population-based survey of 9,691 older adults residing in Korea. Using the family stress framework, we examined how older adults' functional impairment could indirectly inflict emotional elder abuse through diverse aspects of family resources. We also discussed the study's contribution to enhance understanding of family dynamics in response to caring for functionally impaired older family members, as well as to provide specific implications to address the significant global concern.

Theoretical Framework: Family Stress Theory

The current study examined emotional abuse against older adults in the context of the family stress framework that focuses on the process by which the family as a unit adapts or fails to adapt to stressful events or situations. The earlier conceptual foundations of the family stress theory lie in the Hill ADCX family crisis model (1949, 1958) and McCubbin and Patterson's Double ABCX Model (1981, 1983). According to their formulations, family stressors are occurrences that can create some changes in family dynamics and relationships (McKenry & Price, 2005). Family stressors result in outcomes that can be arranged on a continuum from maladaptation/crisis on the negative end to bonadaptation on the positive end (McCubbin & Patterson; 1983). In the stress adaptation process, family adaptive resources play a key role preventing or buffering the impact of family stressors on specific stress outcomes (McCubbin & McCubbin, 1989; McKenry & Price, 2005). Family adaptive resources are part of the family's capabilities to deal with stressful situations that include factors at individual-, familial-, and community-levels (Lavee, McCubbin, & Patterson, 1985; McKenry & Price, 2005). Individual-level factors refer to traits, characteristics, or abilities of individual family members that can include psychological resources, such as self-esteem. Familial-level factors refer to attributes of family as a unit that facilitates successful adaptation to stress events. Two aspects of familial-level factors are often considered: family cohesion (i.e., interconnectedness through affection and support) and adaptability (i.e., capability to change). Community-level factors refer to people or institutions available from outside the family unit, such as friends/neighbors, religious institutions, or government agencies.

Functional Limitations as a Risk Factor for Elder Abuse

Consistent with the extensive existing literature, the current study postulates that an older adult's functional impairment may be a potential family stressor leading to a family crisis which in our case is emotional elder abuse. The early work of Steinmetz (1978; 1988) asserts that elder abuse can occur as a result of stress related to caring for older persons with functional limitations. Recent empirical research also supports this argument: Burnes and colleagues (2015) conducted a population-based study of 4,156 older adults residing in New York State and found that older adults with activities of daily living (ADLs) and instrumental activities of daily living (IADLs) impairments were at a greater risk of emotional and physical abuse. Similarly, Johannesen and LoGiudice (2013) conducted a systematic review of the risk factors associated with abuse among community-dwelling older adults and found that functional dependency or frailty that required assistance with ADLs was a key risk factor. A significant association between functional impairment and elder abuse victimization has also been found in the studies based on older Korean populations. For example, in a study examining a sample of 1,000 family caregivers, 2-10 percent of the respondents reported physically or emotionally abusing their older family members (Lee, 2008). Key predictors of the incidence of elder abuse were the care recipient's number of ADL impairments and the burden associated with caregiving roles.

Family Adaptive Resources as Potential Mediators between Functional Limitations and Elder Abuse Victimization

We also posit the changes in family adaptive resources as potential mechanisms linking functional impairments and emotional elder abuse victimization. Consistent with the family stress conceptualization, we assess three dimensions of family resources that include characteristics/capacities of 1) individual family member, 2) family as a unit, and 3) community or people/institutions outside of family. First, we considered care recipient's self-esteem as a key individual-level factor because empirical evidence shows that impairments in daily functioning can cause significant threats to self-concept. For example, Brown (2016) examined a representative sample of older adults in the U.S. and found that functional decline was significantly associated with low levels of mastery and self-esteem. In a similar vein, Kim and Choi (2011) also found that functional impairment was the key risk factor resulting in low level of self-esteem among older Korean women. Second, we considered that older adults' functional impairment can undermine the degree of cohesion among family members. According to Lim and colleagues (2016) who analyzed a convenient sample of family caregivers in Korea, a care recipient's ADL limitations were associated with negative changes in family relationships that involved filial obligation, affectional solidarity, and self-sacrifice. Similarly, Knussen and colleagues (2005) examined a sample of older adults with hearing disability and their caregivers and found that the care recipients' hearing disability was significantly associated with deterioration in caregivers' perception in family functioning. Third, functional limitations can impair access to community resources. A number of studies have shown that older adults' functional limitations can weaken their ability to build up or maintain social relationships from outside family (Monserud & Peek, 2014).

In turn, these negative changes in multi-dimensional family resources can put older adults at a greater risk of emotional abuse. First, although only scant research exists, individual older adult's psychological resources may predict elder abuse victimization. In few studies that examined convenient samples of older Korean adults, self-esteem was a salient risk factor for incidences and severity of elder abuse (Jang & Park, 2012; Kwon, 2004). Second, negative family relationships and functioning have been identified as a risk factor for elder abuse. For example, a low level of positive support from family was associated with self-reported elder abuse (Luo & Waite, 2011). Similarly, studies based on samples of older Koreans have consistently showed that conflictual relationships between older adults and their adult children increased a risk of elder abuse (Han & Kim, 2000; Lee, 2003). On the other hand, a positive parent-adult child relationship quality can serve as a protective factor preventing elder abuse (Von Heydrich, Schiamberg, & Chee, 2012). Third, lack of social support or social isolation is a significant predictor for elder abuse victimization (Kim, 2015). On the other hand, positive social support plays a protective role deterring elder abuse victimization; Dong and Simon (2007) showed that a greater level of social support, such as having someone to listen to and talk to, or having someone to show love and affection to, was associated with a lower risk of elder abuse. According to De Donder and colleagues (2016) who examined factors associated with severity of elder abuse, involvement in social activities was associated with less severe abuse.

Hypotheses of the Current Study

The theoretical consideration and the review of literature suggest that functional limitations could potentially predict emotional elder abuse victimization due to changes in family adaptive resources at all levels (See Figure 1). Specifically, we evaluated the following mediating hypothesis: an increase in ADL/IADL limitations will be associated with reduced self-esteem, less family cohesion, greater receipt of family assistance, less frequent contact with friends/neighbors, and less participation in social activities, which, in turn, leads to emotional abuse victimization among older Korean adults.

Figure 1. Conceptual Model.

Figure 1

Open in a new tab

Methods

Data Set

We used data from the 2009 Survey of Elderly Care and Welfare Need (SECW) conducted by the Korea Institute of Health and Social Welfare. The SECW is nationally representative data that has been widely recognized for providing robust statistics about elder abuse among older Korean adults. The sampling process first began by dividing nationwide into general districts where the majority of housing included detached houses, and apartment districts. Next, the districts were divided into 6 stratums to represent 7 metropolitan cities, and suburbs. Based on the 2008 National Consensus, 300 districts were again selected and were stratified by age, gender, and geographic location. A total of 8,607 households were contacted and 6,745 households, 9,995 community-dwelling adults aged 65, completed the survey, which yielded the survey response rate of 78%. In carrying out the survey, trained interviewers visited their assigned respondents to conduct face-to-face individual interviews from November 2009 to April 2010. The final study sample consisted of 9,691 older adults after excluding 304 incomplete cases. Specifically, 17 cases were excluded because respondents' age was non-identifiable, 203 cases were excluded because the interviewers noted that the survey was not possible (reasons unspecified), and another 84 cases were excluded because the interviewers noted that the respondents were suspected as having serious cognitive problems, such as dementia. The survey was approved by the Statistics Korea and followed the protocols to protect the research participants.

Measures

Emotional abuse victimization

Emotional abuse victimization was measured by self-reports of whether or not an older adult experienced the following events in the past 12 months by their family member. The questions included 1) ignored me or avoided making conversations with me, 2) yelled, screamed, or cursed at me, 3) expressed verbally that they no longer want to live with me, 4) isolated me from meeting friends or attending activities, 5) threatened to kill me, 6) did not do what I requested or forced me to do things that I declined to do, 7) terrorized or menaced me. A sum score was created by counting the number of yes to each of the seven items, which ranged from 0 to 7. In the multivariate analyses, the categories of 3 and above were combined because of low frequencies. The reliability of the items was moderate with Cronbach's alpha value of .66.

Functional limitations

An older adult's functional status was measured by summing the number of ADL and IADL limitations. First, respondents were asked to report the number of ADL tasks that they needed assistance with in terms of dressing, washing their face, brushing their teeth, bathing, eating, moving around, sitting up, transferring from a bed/chair, getting out of a room, using the toilet, and maintaining continence of bowel/bladder. Second, respondents were asked to report the number of IADL tasks that they needed assistance with in terms of grooming, doing house chores, preparing for meals, doing laundry, travelling to close locations, using public transportation, shopping, managing finances, using telephones, and taking medication.

Family adaptive resources

Three sources of family resources were considered: individual, familial, and community resources. First, individual-level family resource was assessed by an older adult's level of self-esteem. Self-esteem was measured by ten items based on Rosenberg's Self-esteem scale (Rosenberg, 1965). The items include: 1) “I feel that I'm a person of worth, at least on an equal plane with others,” 2) “I feel that I have a number of good qualities,” 3) “all in all, I am inclined to feel that I am a failure,” 4) “I am able to do things as well as most other people,” 5) “I feel I do not have much to be proud of,” 6) “I take a positive attitude toward myself,” 7) “on the whole, I am satisfied with myself,” 8) “I wish I could have more respect for myself,” 9) “I certainly feel useless at times,” 10) “at times, I think I am no good at all.” Respondents rated each item on a four-point Likert scale ranging from strongly disagree to strongly agree. The total score was calculated by averaging the ten items. The negatively worded items (3, 5, 9, and 10) were reverse-coded so that higher values indicate greater levels of self-esteem. The reliability of the scale was high with Cronbach's alpha value of .73.

Second, familial resource was assessed by two dimensions: family cohesiveness and receipt of family assistance. First, family cohesiveness was measured by the following ten items about their family relationships: 1) my family helps each other in difficult times, 2) my family know seach family member's friends and treats them nicely, 3) my family tries to do things together, 4) my family members are closer to each other than any other relationships, 5) my family spends time together in free time, 6) my family is closely bounded, 7) my family actively participates in family events and activities, 8) my family plans for events or activities to spend time together, 9) my family members consult each other before making any decisions, 10) my family emphasizes being harmonious and cooperative with each other. Respondents rated each item on a five-point Likert scale ranging from strongly disagree to strongly agree. The total score was calculated by averaging the ten items. The reliability of the scale was high with Cronbach's alpha value of .93. Second, receipt of family assistance was measured by the number of assistance they received from family in the past 12 months based on the following areas: 1) discussion or consultations for individual or familial issues, 2) housework, such as cleaning, preparing meals, and laundry, 3) personal care or long-term care assistance, 4) financial assistance. A total score was created by counting the number of yes to receiving assistance in these four areas, which ranged from 0 to 4.

Third, community resource was assessed by two dimensions: frequency of contact with friends/neighbors and frequency of participating in social activities. First, frequency of contact with friends/neighbors was measured by asking respondents about how often they contacted their friends or neighbors who they can share their problems or concerns with in the past 12 months. The response choices were coded as a) never contacted or not having friends/neighbors, b) less than once a year, c) once or twice a month, d) once a week, and e) more than once a week. Second, frequency of participating in social activities was measured by asking respondents about how often they participated in social gatherings or organizations (e.g., informal groups, religious groups). The response choices were coded as a) never participated, b) less than once a year, c) once or twice a month, d) once a week, and e) more than once a week.

Covariates

Based on the prior studies on risk factors for elder abuse (e.g., Johannesen & LoGiudice, 2013), we added covariates in the analysis model, including the respondents' gender (male), marital status, age (years), educational attainment, household income, and self-reported health status. Marital status was indicated by six categories (i.e., having a partner or spouse, widowed, divorced, separated, never married, and others), that was dichotomously recoded as married and non-married. Educational attainment was coded as five categories based on the variable distribution: 1) illiterate, 2) no formal education but literate, 3) graduated from elementary school, 4) graduated from middle school or high school, and 5) graduated from college or above. The unit of household income was 10,000 won (approximately 10 U.S. dollars) and was top-coded at 500 to deal with skewness. Self-reported health status was assessed by four-point Likert scale ranging from very poor/poor (1) to very good (4).

Analytic Procedures

To test the mediational hypothesis, we employed a path an alysis using Mplus version 6. As depicted in Figure 1, the associations between the predictor (i.e., ADL/IADL limitations), five mediators (i.e., self-esteem, family cohesion, receipt of family assistance, contact with friends/neighbors, and participation in social activities), and the outcome variable (the experience of emotional elder abuse) were estimated simultaneously. In addition, the list of covariates were regressed on both mediators and the outcome variable.

Because one-third of respondents were nested within households, we calculated the intraclass correlation coefficient (ICC) to determine whether to pursue a multilevel model. The ratio of the between-cluster variance (i.e., .112) to the total variance (i.e., 112 + .582) was .03. The small ICC value (approaching zero) indicated little variability between households, and thus we did not control for the nested data structure. A weighted least square parameter estimate (WLSMV) was used to specify an ordered categorical dependent variable (Muthén & Muthén, 2007). The significance of indirect effects was assessed using the Sobel test. The following criteria of the goodness-of-fit indices were considered to evaluate the model fit (Hu & Bentler, 1995): (a) comparative fit index (CFI) ≥ .95, (b) nonnormed fit index (NNFI) ≥ .95, (c) root mean square error of approximation (RMSEA) < .05, and (d) weighted root mean square residual(WRMR) ≤ .90. Cases with missing data were less than 1% of the respondents, so list wise deletion was used to handle missingness.

Results

Table 1 presents descriptive statistics of key variables. About half of the sample was male (43.6%, n = 4,228), and 70.6% of the sample (n = 6,844) were married or had a partner. The average age was 72.3 years old (SD = 5.66). On average, the respondents completed elementary school education (M = 3.1, SD = 1.1). On average, respondents experienced less than one episode of emotional elder abuse (M = .19, SD = .60). About 11% of the study sample reported emotional abuse victimization by a family member in the past year (n = 1,082, 11.06%). Of the different types of emotional abuse, being ignored was the most common experience (n = 668, 6.9%), followed by being yelled, screamed, or cursed at (n = 565, 5.8%), and being controlled (n = 299, 3.1%). On average, the respondents reported 0.8ADL/IADL limitations (SD = 1.8). Overall, respondents received three different types of help from their family (M = 2.6, SD = 1.0); more than 90% of the respondents received financial help (n = 8,886), 80% received help in terms of discussing/consulting for individual or familial issues, and approximately 40% received help in terms of housework and personal or long-term care. The study respondents contacted their friends or neighbors less than once a year (M = 2.4, SD = 2.4), and participated in social activities once or twice a month (M = 2.8, SD = 2.2).

Table 1. Descriptive Statistics of Study Sample(N =9,691).

Variables N/Mean (SD) %/Range
Emotional abuse victimization .19 (.60) 0/3
 Never experienced 8,609 88.84
 Experienced at least once 1,082 11.17
Number of ADL and IADL limitations .75 (1.76) 0/10
Self-esteem 2.69 (.38) 1.2/4
Family cohesiveness 6.41 (.83) 1/8.7
Receipt of family assistance 2.56 (1.00) 0/4
Frequency of contact with friends/neighbors 2.36 (2.42) 0/6
Frequency of participating in social activities 2.81 (2.20) 0/6
Male 4,228 43.63
Married 6,844 70.62
Education 3.10 (1.08) 1/5
Age 72.33 (5.66) 65/99
Household income 183.15 (193.50) 0/2,500
Self-reported health 2.85 (.93) 1/4
Open in a new tab

Notes. Descriptive statistics are reported prior to correction for skewness. Frequency of contact with friends/neighbors were coded as 1) never contacted (no family/friends), 2) less than once a year, 3) once or twice a month, 4) once a week, 5) more than once a week. Frequency of participating in social activities were coded as 1) never participated, 2) less than once a year, 3) once or twice a month, 4) once a week, 5) more than once a week. Education was coded as 1) illiterate, 2) no formal education but literate, 3) graduated from elementary school, 4) graduated from middle school or high school, 5) graduated from college or above.

Mediational Analysis

Table 2 and Figure 2 illustrate the results of the mediational analysis. Overall, the final model fitted the data well (CFI = 1.00; NNFI = .95; RMSEA = .03). The initial model was respecified by permitting mediators to covary as guided by modification indices. First, greater ADL and IADL limitations were not directly associated with emotional abuse victimization (b = -.01, p = ns). The number of ADL/IADL limitations was significantly associated with family resources across all levels. That is, greater ADL/IADL limitations were associated with reduced self-esteem (b = -.05, p< .001), greater assistance received from family (b = .16, p< .001), less frequent contacts with friends/neighbors (b = -.22, p< .001), and less frequent participation in social activities (b = -.30, p< .001). In turn, self-esteem and family cohesiveness were negatively associated with emotional abuse victimization (b = -.67; b = -.22, ps< .001, respectively) whereas the receipt of family assistance was positively associated with emotional abuse victimization (b = .04, p< .001). The test of indirect effects showed that self-esteem and the receipt of family assistance were significant mediators between ADL/IADL limitations and emotional abuse victimization. That is, greater ADL/IADL limitations were associated with reduced self-esteem, which was ultimately associated with a greater risk for emotional abuse victimization (b = .03, p< .001). In addition, greater ADL/IADL limitations were associated with greater assistance received from family, which was ultimately associated with a greater risk of emotional abuse victimization (b = .02, p< .001).

Table 2. Path Analysis of ADL/IADL Limitationson Emotional Abuse Victimization through Multi-level Resources (N =9,691).

Parameter Estimates Unstandardized (SE) Standar-dized
ADL/IADL limitations → Self-esteem -.05 (.00)*** -.15
ADL/IADL limitations → Family cohesiveness -.01 (.01) -.01
ADL/IADL limitations → Receipt of family assistance .16 (.01)*** .18
ADL/IADL limitations → Contact with friends/neighbors -.22 (.03)*** -.10
ADL/IADL limitations → Participation in social activities -.30 (.02)*** -.15
ADL/IADL limitations → Emotional abuse victimization -.01 (.02) -.01
Self-esteem → Emotional abuse victimization -.67 (.05)*** -.25
Family cohesiveness → Emotional abuse victimization -.22 (.01)*** -.17
Receipt of family assistance → Emotional abuse victimization .10 (.02)*** .09
Contact with friends/neighbors → Emotional abuse victimization .01 (.01) .03
Participation in social activities → Emotional abuse victimization .00 (.01) .00
Significant Indirect Effects Unstandardized (SE) Standar-dized
ADL/IADL limitations → Self-esteem → Emotional abuse victimization .03 (.003)*** .04
ADL/IADL limitations → Receipt of family assistance → Emotional abuse victimization .02 (.003)*** .02
Goodness-of-fit Indices
χ2(df) 8.84 (1)**
Root-mean-square error of approximation [90% CI] .03 [.01, .05]
Comparative fit index 1.00
Nonnormed fit index .95
Weighted root mean square residual .36
Open in a new tab

Notes. Significance levels are denoted as

*

p < .05,

**

p <.01,

***

p <.001.

Each analysis model controlled for gender, marital status, education, age, household income, and self-reported health status.

Figure 2.

Figure 2

Open in a new tab

Results of Path Analysis. Standardized path coefficients are shown. Significance levels are denoted as * p < .05, ** p <.01, ***p <.001. Bold arrows denote paths involving significant indirect effects: ADL/IADL limitations → self-esteem → emotional abuse victimization: b = .04, p < .001; ADL/IADL limitations → receipt of family assistance → emotional abuse victimization: b = .02, p < .001. This mediational model includes covariates of gender, marital status, educational attainment, age, household income, and self-reported health status.

Discussion

The current study considered emotional elder abuse as a family-level issue that may occur by not responding adaptively to functional impairments of older family members. We also focused on the mediating roles of diverse aspects of family resources in the association between older adults' functional impairments and emotional elder abuse victimization.

Effect of Functional Decline on Family Adaptive Resources

The key findings of this study showed that an older adult's functional decline had direct effect on several aspects of family adaptive resources. That is, an increase in ADL/IADL limitations was associated with older adults' reduced self-esteem, greater assistance received from family, less frequent contact with friends/neighbors, and less participation in social activities. Elder abuse literature based on the family stress framework has focused on the role of family adaptive resources as buffering the effect of family stressors or preventing elder abuse (e.g., Florian & Dangoor, 1994). However, the current study contributes to the existing literature by showing that functional decline may have a direct impact on undermining the capabilities of the whole family, as well as an individual older adult. That is, functional decline involves the deterioration of family adaptive resources that are known to be the major protective mechanisms against elder abuse. Therefore, the intervention strategies relying on an older adult's existing resources may not be efficacious in terms of preventing elder abuse victimization. Instead, intervention programs should focus more on maintaining or strengthening older adults' resilient resources in the face of functional decline.

Mechanisms Linking Functional Decline and Elder Abuse Victimization

Furthermore, we found significant mediating paths between functional declines, changes in family resources, and elder abuse among older Korean adults. Specifically, an increase in ADL/IADL limitations was associated with lower self-esteem, which was ultimately associated with a greater risk of emotional abuse victimization. Considering that high self-esteem plays a role as a positive, effective coping resource that helps deal with stressful situations (Taylor & Stanton, 2007), an older adult's feelings of reduced self-worth due to functional limitations may weaken his or her ability to manage and resolve conflicting situations with family members. Alternatively, as social exchange theory postulates, this might be an issue of power imbalance that elder victims become unable to defend them selvesdue to a lack of self-worth against abusers who have greater power by doing the caregiving roles (Momtaz, Hamid, & Ibrahim, 2009; Wolf, 1997).

Additionally, a greater receipt of family assistance was another significant mechanism; namely, older Koreans who receive greater assistance from family members were at a higher risk of emotional elder abuse. Considering a significant correlation between caregiver's subjective burden and the extent of assistance provided (Campbell et al., 2008), older adults' self-reports of dependence on family (i.e., extent of receiving assistance from family) can serve as a proxy to assess the level of burden on family caregivers. In that regard, this study supports the findings of prior studies that the subjective caregiver burden is a significant predictor for committing elder abuse against older family members. For example, Lee (2008), who examined a sample of family caregivers residing in Korea, found that the burden of care giving for a family member with physical impairments was significantly associated with caregivers' impulsive feelings to commit elder abuse. This result urges the importance of alleviating the burden and distress of family caregivers through expanded policy support. In Korea, currently, publicly-supported long-term care services intervene in only the small fraction of functionally impaired older adults (Lee & Moon, 2016). For example, in 2008, the Korean government launched the universal long-term care insurance program to provide financial and non-financial supports to family caregivers. Despite passage of decade, only 6.4 percent of the total older population receives benefits from the program of which the coverage rate is much less than the average 12.7 percent of the OECD member countries(Lee & Moon, 2016). Thus, expanding the number of beneficiaries will be one way to relieve the burden of family caregivers and prevent elder abuse inflicted by family members.

We also note that, in the current study, social support from outside family (i.e., frequency of contacting with friends/neighbors and participating in social activities) was not a significant mediator between an older adult's functional decline and emotional abuse victimization. This finding can be attributed to the specific characteristics of the study sample. This sample of older Korean adults reported a low level of involvement in social activities, which may well represent the isolated status of older adults in Korea. For this particular cohort of older Korean adults, family is the major source of social support; many of them do not rely on non-familial sources for support and they are not actively involved in social relationships outside family (Oh et al., 2006). However, in other cultural contexts where older adults' involvement in social relations and activities are more vital, as evidenced by prior studies, social support from non-family members could play an important protective role against elder abuse victimization (e.g., Dong & Simon, 2007). This anticipated difference also emphasizes the importance of examining elder abuse issues from different cultures and contexts to better understand the etiology of this family violence as well as its correlates and consequences.

Limitations

This study has some limitations to be noted. First, the cross-sectional nature of the data limits the argument related to mediational relationships among key variables. For example, this study cannot exclude the possibility that emotional abuse victimization may have led the sample of older adults to have low levels of self-esteem. To obtain robust evidence about the specific pathways to elder abuse victimization, more sophisticated data analysis techniques based on longitudinal data are required. Second, psychometric properties of the emotional abuse items were somehow limited in that a) the items were not derived from a validated scale, and b) the reliability of the scale was moderate (i.e., α = .66). In addition, face-to-face interviews at respondents' house may have involved measurement errors because older adults may not disclose their victimization if the abuser is present. (Schofield, Powers, MMedStats, & Loxton, 2013). In terms of collecting data on elder abuse, telephone interviews that are more anonymous and less intimidating than in-person contact may reduce older adults' reluctance to share their sensitive interpersonal information (Acierno, 2003). Interviewers should also consider ways to better communicate with older respondents with sensory impairments such as vision and/or hearing loss. Another limitation is that we have not considered other types of elder abuse that include physical, sexual abuse and financial exploitation. Because multiple forms of violence often co-occur on an older adult (Jackson & Hafemeister, 2013), focusing solely on the emotional aspect of abuse may limit comprehensive understanding of the causes and consequences of elder abuse. Lastly, we did not include covariates such as older adults' medical conditions or cognitive functions that may confound with elder abuse victimization (Johannesen & LoGiudice, 2013).

Implications and Future Directions

Despite these limitations, the current study contributes to the enhanced understanding of older Korean adults' experience of emotional abuse within family. Emotional elder abuse is subtle by nature and is hard to be detected, unlike physical abuse, and thus the issue requires more systematic attention from advocacy groups, researchers, policy makers, and legislators. One encouraging aspect is that the Elderly Welfare Act was revised in 2016 to sanction any act that causes emotional distress or anguish to an older person, which clearly shows that the awareness about emotional elder abuse has grown in Korea. However, we also note that elder abuse issues have a low rate of reporting to official sources (i.e., 0.5% in 2015; KEPA, 2016). Furthermore, less than 40% of older Korean adults stated that they would be willing to report their abuse cases to elder protective services or law enforcement agencies possibly because of stigma associated with being victimized in family or not wanting to prosecute their family members (KEPA, 2016). Through senior or community centers where older adults and family caregivers are relatively easy to access, intervention campaigns or educational programs to increase the awareness of elder abuse issues should be organized. For practitioners working with vulnerable older adults, it is important, first, to be aware that functional decline can diminish older adults' capabilities of maintaining independence in several aspects and then to create intervention plans to minimize the loss of adaptive resources. Particularly, they can focus on protecting or enhancing the self-esteem of older adults even in situations of increasing dependence on others. For example, in Korea, many local community welfare centers or senior centers provide specific programs to enhance older adult's self-esteem. We also recommend the practitioners to provide telephone or on-line consultation programs to help enhance self-esteem for older adults with severe functional impairments. Additionally, practitioners can help an older adult diversify support resources through social work organizations or religious institutions. More importantly, further research is need to support these practice efforts. We particularly emphasize the value of cross-cultural research to better address this global concern as many countries are currently dealing with similar sets of issues – that is, the increase in the number and life expectancy of older people, changes in family norms and structure, and limited roles of the public arena in terms of caregiving – that serve as contextual factors for elder abuse.

Acknowledgments

This study was partially supported by National Institute on Aging Grants T32 AG049676 (D. Almeida) and K02 AG039412 (L. Martire) to The Pennsylvania State University.

Contributor Information

Jooyoung Kong, Center for Healthy Aging, College of Health and Human Development, Pennsylvania State University, 422 Biobehavioral Health Building, University Park, PA 16802.

Haesang Jeon, Research Department for Women's Human Rights & Safety, Korean Women's Development Institute, 225, Jinheung-ro, Eunpyeong-gu, Seoul 03367, Korea.

References

  1. Acierno R. Elder mistreatment: Epidemiological assessment methodology. In: Bonnie RJ, Wallace RB, editors. Elder mistreatment: Abuse, neglect, and exploitation in an aging America. Washington, D. C.: The National Academies Press; 2003. pp. 261–302. [PubMed] [Google Scholar]
  2. Brown RL. Functional limitation and depressive symptomatology: Considering perceived stigma and discrimination within a stress and coping framework. Stigma and Health. 2016 Jul 7; doi: 10.1037/sah0000042. Online First: http://dx.doi.org/10.1037/sah0000042. [DOI] [PMC free article] [PubMed]
  3. Burnes D, Pillemer K, Caccamise PL, Mason A, Henderson CR, Berman J, et al. Lachs MS. Prevalence of and risk factors for elder abuse and neglect in the community: A population-based study. Journal of the American Geriatrics Society. 2015;63(9):1906–1912. doi: 10.1111/jgs.13601. [DOI] [PubMed] [Google Scholar]
  4. Campbell P, Wright J, Oyebode J, Job D, Crome P, Bentham P, et al. Lendon C. Determinants of burden in those who care for someone with dementia. International Journal of Geriatric Psychiatry. 2008;23:1078–1085. doi: 10.1002/gps.2071. [DOI] [PubMed] [Google Scholar]
  5. De Donder L, Lang G, Ferreira-Alves J, Penhale B, Tamutiene I, Luoma M. Risk factors of severity of abuse against older women in the home setting: A multinational European Study. Journal of Women and Aging. 2016;28(6):540–554. doi: 10.1080/08952841.2016.1223933. [DOI] [PubMed] [Google Scholar]
  6. Dong X. Elder abuse: Systematic review and implications for practice. Journal of the American Geriatrics Society. 2015;63(3):1214–1238. doi: 10.1111/jgs.13454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dong X, Simon MA. Is greater social support a protective factor against elder mistreatment? Gerontology. 2007;54:381–388. doi: 10.1159/000143228. [DOI] [PubMed] [Google Scholar]
  8. Florian V, Dangoor N. Personal and familial adaptation of women with severe physical disabilities: A further validation of the double ABCX model. Journal of Marriage and the Family. 1994;56(3):735–746. [Google Scholar]
  9. Hill R. Families under stress. Westport, CT: Greenwood; 1949. [Google Scholar]
  10. Hill R. Social stresses on the family: Generic features of families under stress. Social Casework. 1949;39:139–150. [Google Scholar]
  11. Hu LT, Bentler P. Evaluating model fit. In: Hoyle RH, editor. Structural Equation Modeling Concepts, Issues, and Applications. London: Sage; 1995. pp. 76–99. [Google Scholar]
  12. Jackson SL, Hafemeister TL. Financial abuse of elderly people vs other forms of elder abuse: Assessing their dynamics, risk Factors, and society's response (National Institute of Justice Final Report [2011]) 2013 Retrieved from https://ssrn.com/abstract=2279695.
  13. Jang MH, Park CG. Risk factors influencing probability and severity of elder abuse in community-dwelling older adults: Applying zero-inflated negative binomial modeling of abuse count data. Journal of Korean Academy of Nursing. 2012;42(6):819–832. doi: 10.4040/jkan.2012.42.6.819. [DOI] [PubMed] [Google Scholar]
  14. Johannesen M, LoGiudice D. Elder abuse: A systematic review of risk factors in community-dwelling elders. Age and Ageing. 2013;42:292–298. doi: 10.1093/ageing/afs195. [DOI] [PubMed] [Google Scholar]
  15. Kim Y. A qualitative study on the experiences of abused older adults regarding abuse recurrence and adult protective services [in Korean] Health and social welfare review. 2015;35(2):315–346. [Google Scholar]
  16. Kim NH, Choi I. Effects of the physical and social characteristics of elderly women on self-esteem and life satisfaction [in Korean] Journal of the Korea contents association. 2011;11(11):241–252. [Google Scholar]
  17. Korea Elder Protection Agency. Annual report: Current status of elder abuse 2015. 2016 Retrieved from http://noinboho.or.kr/
  18. Korean Institute for Health and Social Affairs. Senior survey. 2014 Retrieved from http://www.mohw.go.kr.
  19. Knussen C, Tolson D, Swan IRC, Stott DJ, Brogan CA. Stress proliferation in caregivers: The relationships between caregiving stressors and deterioration in family relationships. Psychology and Health. 2005;20(2):207–221. doi: 10.1080/08870440512331334013. [DOI] [Google Scholar]
  20. Kwon JD. A study on the causal factors of elderly abuse [in Korean] Journal of the Korea Gerontological Society. 2004;24(1):1–19. [Google Scholar]
  21. Lavee Y, McCubbin HI, Patterson JM. The double ABCX model of family stress and adaptation: An empirical test by analysis of structural equations with latent variables. Journal of Marriage and Family. 1985;47(4):811–825. [Google Scholar]
  22. Lee M. Caregiver stress and elder abuse among Korean family caregivers of older adults with disabilities. Journal of Family Violence. 2008;23:707–712. doi: 10.1007/s10896-008-9195-2. [DOI] [Google Scholar]
  23. Lee M. The impact of elder abuse risk factors and characteristics on the effects of elder abuse [in Korean] Journal of the Korea Gerontology. 2003;23(2):105–123. [Google Scholar]
  24. Lee H, Moon Y. A delphi analysis of enhancement plans for long-term care insurance coverage [in Korean] Journal of the Korean Social Security Association. 2016;32(2):219–243. [Google Scholar]
  25. Lim YM, Ahn YH, Ahn JY. Multidimensional caregiving burden of female family caregivers in Korea. Clinical Nursing Research. 2016;25(6):665–682. doi: 10.1177/1054773815591472. [DOI] [PubMed] [Google Scholar]
  26. Luo Y, Waite LJ. Mistreatment and psychological well-being among older adults: Exploring the role of psychosocial resources and deficits. Journals of Gerontology: Psychological Sciences and Social Sciences. 2011;66(2):217–229. doi: 10.1093/geronb/gbq096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. McCubbin HI, Patterson JM. The family stress process: The Double ABCS Model of adjustment and adaptation. Marriage and Family Review. 1983;6:7–37. [Google Scholar]
  28. McCubbin HI, Patterson JM. Systematic assessment of family stress, resources and coping: Tools for research, education and clinical intervention. St. Paul, MN: Family Social Science; 1981. [Google Scholar]
  29. McCubbin MA, McCubbin HI. Theoretical orientations to family stress and coping. In: Figley CR, editor. Treating stress in families. New York, NY: Crunner/Mazel, Inc.; 1989. pp. 3–43. [Google Scholar]
  30. McKenry PC, Price SJ. Families and change: Coping with stressful events and transitions. 3rd. Thousand Oaks, CA: Sage Publications, Inc; 2005. [Google Scholar]
  31. Momtaz YA, Hamid TA, Ibrahim R. Theories and measures of elder abuse. Psychogeriatrics. 2009;13:182–188. doi: 10.1111/psyg.12009. [DOI] [PubMed] [Google Scholar]
  32. Monserud MA, Peek MK. Functional limitations and depressive symptoms: A longitudinal analysis of older Mexican American couples. Journals of Gerontology: Psychological Sciences and Social Sciences. 2014;69(5):743–762. doi: 10.1093/geronb/gbu039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Muthén LK, Muthén BO. Mplus user's guide. 5th. Los Angeles, CA: Muthén & Muthén; 2007. [Google Scholar]
  34. Oh J, Kim HS, Martins D, Kim H. A study of elder abuse in Korea. International Journal of Nursing Studies. 2006;43:203–214. doi: 10.1016/j.ijnurstu.2005.03.005. [DOI] [PubMed] [Google Scholar]
  35. Rosenberg M. Society and the adolescent self-image. Princeton, NJ: Princeton University Press; 1965. [Google Scholar]
  36. Schofield MJ, Powers JR, MMedStats, Loxton D. Mortality and disability outcomes of self-reported elder abuse: A 12-year prospective investigation. Journal of the American Geriatrics Society. 2013;61(5):679–685. doi: 10.1111/jgs.12212.Epub2013Apr16. [DOI] [PubMed] [Google Scholar]
  37. Steinmetz SK. Battered parents. Society. 1978;15:54–55. [Google Scholar]
  38. Steinmetz SK. Duty bound: Elder abuse and family care. Newbury Park, CA: Sage; 1988. [Google Scholar]
  39. Taylor SE, Stanton AL. Coping resources, coping processes, and mental health. Annual Review of Clinical Psychology. 2007;3:377–401. doi: 10.1146/annurev.clinpsy.3.022806.091520. [DOI] [PubMed] [Google Scholar]
  40. Von Heydrich L, Schiamberg LB, Chee G. Social-relational risk factors for predicting elder physical abuse: An ecological bi-focal model. International Journal of Aging and Human Development. 2012;75(1):71–94. doi: 10.2190/AG.75.1.f. doi: http://dx.doi.org/10.2190/AG.75.1.f. [DOI] [PubMed] [Google Scholar]
  41. Wolf RS. Elder abuse and neglect: Causes and consequences. Journal of Geriatric Psychiatry. 1997;30(1):153–174. [Google Scholar]

RESOURCES