Association between elder abuse and use of ED: findings from the Chicago Health and Aging Project

American Journal of Emergency Medicine 31 (2013) 693–698

Contents lists available at SciVerse ScienceDirect

American Journal of Emergency Medicine journal homepage: www.elsevier.com/locate/ajem

Original Contribution

Association between elder abuse and use of ED: findings from the Chicago Health and Aging Project☆ XinQi Dong MD, MPH a,⁎, Melissa A. Simon MD, MPH b a b

Rush Institute for Healthy Aging, Rush University Medical Center, Chicago, IL 60612, USA Department of Obstetrics and Gynecology, Northwestern University, Chicago, IL 60601, USA

a r t i c l e

i n f o

Article history: Received 24 October 2012 Received in revised form 18 December 2012 Accepted 21 December 2012

a b s t r a c t Purpose: This study aims to quantify the relationship between overall elder abuse and specific subtypes of elder abuse and rate of emergency department (ED) utilization in a community-dwelling population. Methods: A population-based study is conducted in Chicago of community-dwelling older adults who participated in the Chicago Health and Aging Project. Of the 6674 participants in the Chicago Health and Aging Project, 106 participants were reported to a social services agency for suspected elder abuse. The primary predictor was elder abuse reported to a social services agency. The outcome of interest was the annual rate of ED utilization obtained from the Center for Medicare and Medicaid Services. Poisson regression models were used to assess these longitudinal relationships. Results: The average annual rate of ED visits for those without elder abuse was 0.7(1.4) and, for those with reported elder abuse, was 2.1(3.2). After adjusting for sociodemographics, socioeconomic variables, medical comorbidities, cognitive and physical function, and psychosocial wellbeing, older adults who have been abused had higher rates of ED utilization (RR, 2.33 [1.60-3.38]). Psychological abuse (RR, 1.98[1.29-3.00]), financial exploitation (RR, 1.59 [1.01-2.52]) and caregiver neglect (RR, 2.04 [1.38-2.99]) were associated with increased rates of ED utilization, after considering the same confounders. Interaction terms suggest the association between elder abuse and ED utilization is not mediated through medical comorbidities, cognitive and functional impairment, or psychosocial distress. Conclusion: Elder abuse was associated with increased rates of ED utilization in this community population. Specific subtypes of elder abuse had differential association with increased rate of ED utilization. © 2013 Elsevier Inc. All rights reserved.

1. Introduction Elder abuse includes physical abuse, sexual abuse, psychological abuse, caregiver neglect, and financial abuse. The National Research Council [1] defines elder abuse as “intentional actions that cause harm or create a serious risk of harm, whether or not intended, to a vulnerable elder by a caregiver or other person who stands in a trust relationship to the elder; or failure by a caregiver to satisfy the elder's basic needs or to protect the elder from harm.” Available data suggests that one out of ten US elderly persons experience abuse each year, and many of them experience it in multiple forms [2,3]. In addition, recent data from US Adult Protective Services Agencies depict an increasing trend in the reporting of elder abuse [4]. This trend is particularly

☆ This work was supported by National Institute on Aging grant (R01 AG042318, R01 MD006173, R01 AG11101 & RC4 AG039085), Paul B. Beeson Award in Aging (K23 AG030944), The Starr Foundation, John A. Hartford Foundation and The Atlantic Philanthropies. ⁎ Corresponding author. Tel.: +1 312 942 3350; fax: +1 312 942 2861. E-mail address: [email protected] (X. Dong). 0735-6757/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ajem.2012.12.028

alarming as elder abuse is associated with increased risk of morbidity and mortality [5-7]. National Research Council has urgently called for rigorous research on all aspects of elder abuse, especially through population-based epidemiological studies [1], as our current understanding of the consequences for elder abuse in the general population remains limited. Moreover, elder abuse has great relevance not only to health care professionals and social services agencies, but also to public health professionals, community organizations and other relevant disciplines. As our aging population increases, elder abuse will likely become an even more pervasive issue across all sociodemographic and socioeconomic strata. Despite recent advances in our knowledge of elder abuse, there remain gaps in our understanding in the health services utilization among those who are victimized. Prior case reports suggest frequent health services utilization among those elder abuse victims [8,9]. Recent epidemiological studies have provided conflicting results in health services utilization among those reported to the adult protective services [10,11]. Older adult victims often are put in situation that may threaten their health and safety, which further

694

X. Dong, M.A. Simon / American Journal of Emergency Medicine 31 (2013) 693–698

race, exact home address, zip codes, and the home phone number. This resulted in 106 older CHAP participants who matched a social service agency record. If a CHAP participant was found to be reported more than once, we selected the first report.

predisposes their likelihood to have more frequent emergency department (ED) utilization. Despite contribution of recent work, we are not aware of any epidemiological study that has systematically examined the association between elder abuse and rate of ED use a community-dwelling population. Improved understanding of important factors that predict ED use could inform strategies for social services practice, public policy as well as clinical practice. In this study, we aim to quantify: (1) the relationship between elder abuse and the rate of ED use within a population-based cohort; and (2) the relationship between specific subtypes of elder abuse and ED use in the same cohort. We hypothesized that older adults who are abused have increased rates of ED use, even after controlling for potential confounders, and that the rate of ED use differs by the different subtypes of elder abuse.

2.2.1. Emergency department utilization ED use records were abstracted from the Medicare Standard Analytic Files (SAFs) obtained from the CMS. The CMS has approved the Study Protocol and Data Use Agreement with the CHAP study to obtain CMS data. CHAP study has successfully linked participants and their CMS claims data for the Medical Denominator Files and the SAF files which contains the record of ED utilization. For each participant, we have abstracted and summarized SAF files on their number of ED use during the study period.

2. Methods

2.3. Covariates

2.1. Setting

Demographic variables include age (in years), sex (men or women), race (self-reported: non-Hispanic black versus non-Hispanic white), income categories, and education (years of education completed). The parent CHAP study also collected self-reported medical conditions of hypertension, diabetes mellitus, stroke, coronary artery disease, hip fracture, and cancer. A battery of four cognitive function tests was administered: the Mini-Mental State Examination (MMSE) [14], immediate and delayed recall of brief stories in the East Boston Memory Test [15], and the Symbol Digit Modalities Test [16]. To assess global cognitive function with minimal floor and ceiling artifacts, we constructed a summary measure for global cognition based on all 4 tests. Physical function was assessed using the Katz Index of Activities of Daily Living, which measured limitations in an individual's ability to perform basic self-care tasks [17]. Physical function was also assessed by direct performance testing, which provided a comprehensive objective and detailed assessment of certain abilities [18]. Lowerextremity performance tests consisted of measures of tandem stand, timed walk, tandem walk, and ability to rise to a standing position from a chair. Most of these performance tests were used in the Established Populations for Epidemiologic Studies of the Elderly project [18] and in other large-scale studies of disability. Summary measures of these above tests were created as physical performance test scores (range, 0-15). Psychosocial factors included assessment of depressive symptoms, social network and social engagement. Symptoms of depression were measured using a modified version [19] of the Center for Epidemiologic Studies of Depression Scale [20]. Social network was summarized as the total number of children, relatives, and friends seen at least monthly [21]. Social engagement was assessed by asking how often older adults participate in social activities outside of house, such as religious activities, museums, library, and senior centers.

The Chicago Health and Aging Project (CHAP) began in 1993 to examine risk factors for Alzheimer's disease. Its participants include residents of 3 adjacent neighborhoods on the south side of Chicago. More in depth details of the study design of CHAP have been previously published [12,13]. Data collection occurred in cycles, each lasting 3 years, with each cycle ending as the succeeding cycle began. Each cycle consisted of in-person interviews in the participants' homes. In this study, we used CHAP data as the parent cohort from which elder abuse data was matched from an external source, in order to set the platform for our current analyses. 2.2. Participants In the current study, participants include those who were enrolled between 1993 and 2010 and had valid data on ED utilization history (N = 6674) obtained from the Center for Medicare and Medicaid Services (CMS). From this cohort, we identified a subset of participants (N = 106) who were reported to social services agencies for elder abuse. Suspected elder abuse cases were reported by friends, neighbors, family, social workers, city workers, health care professionals, and others. The reports were usually triggered by concerns for the health and safety of the older adult in their environment, which would initiate a number of different services to help the affected person. All CHAP participants received structured, standardized in-person interviews that included assessment of health history. Written informed consent was obtained, and the study was approved by the institutional review board at the Rush University Medical Center. In the Illinois Adult Protective Services (APS), physical abuse is defined as inflicting physical pain or injury. Sexual abuse is nonconsensual touching, fondling, intercourse, or any other sexual activity. Psychological abuse involves verbal assaults, threat of abuse, harassment, or intimidation. Confinement is restraining or isolating an older adult, other than for medical reasons. Neglect is a caregiver's failure to provide an older adult with life's necessities. Willful deprivation is defined as willfully denying an older adult medication, medical care, shelter, food, a therapeutic device or other physical assistance. Financial exploitation includes the misuse or withholding of an older adult's resources by another. After APS investigation, confirmed elder abuse implied that there were evidence of elder abuse. However, unconfirmed elder abuse cases do not necessarily mean there was no evidence of elder abuse, as sometime APS staff could not gain access to older adults or older adult victim refused to cooperate or participate in services. We matched data from CHAP participants to elder abuse cases reported to social services agencies. Matching was based on an algorithm that compared the following information: date of birth, sex,

2.4. Analytic approach Descriptive characteristics were provided by elder abuse groups across the sociodemographic variables, socioeconomic variables, medical conditions, cognitive function, physical function and psychosocial factors. Our independent variables of interest were reported elder abuse, confirmed elder abuse, and different subtypes of elder abuse. Our outcome of interest was annual rate of ED use, which was summarized for those with and without elder abuse. We used the ztest to compare differences in the rate of ED use between groups. Poisson regression models were used to quantify the relation between elder abuse and rate of ED use. We used a series of models to consider these relationships, taking into consideration the potential confounders. In our core model (Model A), we included age and sex. In addition, we added to prior model the variables of race, education, and income to quantify the association of elder abuse

X. Dong, M.A. Simon / American Journal of Emergency Medicine 31 (2013) 693–698

and ED use outcomes (Model B). Moreover, we added to the prior model common medical comorbidities of hypertension, coronary artery disease, stroke, hip fracture, cancer, and diabetes as well as levels of cognitive function and physical function (Model C). Finally, models were repeated controlling for additional psychological and social measures (Model D). We also repeated the prior models A to D to examine the association between confirmed elder abuse and rate of ED use. In addition, we examined the association between elder abuse subtypes of psychological abuse, physical abuse, caregiver neglect, and financial exploitation and rate of ED use by repeating Models A-D. Lastly, we conducted analyses comparing those with multiple forms of elder abuse (physical abuse and psychological abuse, or caregiver neglect and financial exploitation, etc) adjusting for the same confounding factors. Due to the low rate, we did not analyze sexual abuse cases. Rate ratio (RR), 95% CI, standardized-parameter estimates (PE), standard error (SE) and P values were reported for the regression models. Analyses were carried out using SAS, Version 9.2 (SAS Institute Inc, Cary, NC). 3. Results 3.1. Baseline characteristics There were a total of 6,674 CHAP participants in this study and 106 participants were identified by social services agencies for suspected elder abuse from 1993 to 2010 (confirmed = 56; psychological abuse = 45, physical abuse = 19, caregiver neglect = 50, and financial exploitation = 65). The mean age of those with reported elder abuse was 72.9 years (SD = 5.9 years) and those without elder abuse was 72.9 (6.9). Those with reported elder abuse were more likely to be women, black older adults, and have lower levels of education and income (Table 1).

695

Table 2 Elder abuse and rate of ED use

No elder abuse (N = 6568) Reported elder abuse (N = 106) Confirmed elder abuse (N = 56) Specific elder abuse subtypes Caregiver neglect (N = 50) Psychological abuse (N = 45) Physical abuse (N = 19) Financial exploitation (N = 65) Multiple types of elder abuse 1 type (N = 51) 2 or more types (N = 55)

Mean

SD

Median

IQR

z-test

P

0.74 2.11 2.22

1.43 3.18 2.95

0.33 0.89 0.91

0.84 2.49 3.06

4.29 2.98

b.001 .003

2.89 2.01 2.27 1.67

3.88 2.94 3.26 2.60

1.38 0.67 0.78 0.67

3.29 2.57 2.85 1.79

4.09 2.07 2.06 2.41

b.001 .038 .039 .016

2.04 2.19

3.49 2.92

0.85 1.04

1.89 3.00

2.58 3.39

.005 b.001

for caregiver neglect, 1.89 (3.88); and for financial exploitation, 1.67 (2.60). For elder abuse victims who suffered 1 type of elder abuse, the annual rate of ED use was 2.04 (3.49), and for 2 or more types of elder abuse, was 2.19 (2.92). In the initial Poisson regression model adjusting for age and sex, we found that reported elder abuse strongly predicted the increased rate of ED use (RR, 2.15; 95% CI, 1.67-2.78) (Table 3, Model A). After adding race, education, and income, the association diminished slightly (RR, 1.89; 95% CI, 1.45-2.48) (Model B). Next, we added common chronic medical conditions of hypertension, diabetes, stroke, cancer, thyroid disease, myocardial infarction, cognitive function, and physical function to the model (Model C); the association remained significant (RR, 1.71; 95% CI, 1.28-2.31). In the last model (Model D), after adjusting for psychological and social well-being factors, reported elder abuse remained an independent predictor of increased rate of ED use (RR, 1.76, 95% CI; 1.31-2.37). For confirmed elder abuse, in the fully adjusted model, our data indicate that elder abuse is associated with increased rate of ED use (RR, 2.33; 95% CI, 1.60-3.38).

3.2. Elder abuse and rate of ED utilization The annual rate of ED use for those no reported for elder abuse was 0.74 (1.43) and for those with reported elder abuse was 2.11 (3.18) (z-test, 4.29, P b .001) (Table 2). Similar results were found for confirmed elder abuse. In addition, the annual rate of ED use for psychological abuse was 2.01 (2.94); for physical abuse, 2.27 (3.26);

Table 1 Characteristics of elder abuse and no elder abuse in a community-dwelling population

Age, years, mean, (SD) Women, no. (%) Black, no. (%) Education, years, mean, (SD) Income categories, mean (SD) Medical conditions Coronary artery disease, no. (%) Stroke, no. (%) Cancer, no. (%) Hypertension, no. (%) Diabetes, no. (%) Hip Fracture, no. (%) Global cognitive function, mean (SD) MMSE, mean (SD) Katz, mean (SD) Physical performance testing, mean (SD) Depressive symptoms, mean (SD) Social engagement, mean (SD) Social network, mean (SD)

Elder abuse (N = 106)

No elder abuse (N = 6,568)

P

72.9 76 95 11.3 4.2 0.9 10 14 17 51 11 2 −0.03

72.9 (6.9) 3819 (58.2) 3661 (55.7) 12.5 (3.5) 5.6 (2.6) 0.9 (0.9) 849 (12.9) 582 (8.9) 1222 (18.6) 3216 (49.2) 411 (6.3) 217 (3.3) 0.24 (0.80)

.05 .005 b.001 b.001 b.001 .99 .28 .12 .49 .82 .08 .42 b.001

25.3 (4.8) 0.3 (0.9) 9.4 (3.7)

26.4 (4.9) 0.3 (1.1) 10.5 (3.7)

b.001 .68 .003

1.9 (2.4) 2.3 (1.6) 6.9 (5.8)

1.4 (1.9) 2.4 (1.7) 7.6 (6.5)

.15 .28 .70

(5.9) (71.7) (89.6) (3.1) (1.9) (0.9) (9.4) (13.2) (16.0) (48.1) (10.4) (1.9) (0.75)

3.3. Specific subtypes of elder abuse and ED utilization To quantify the relation between specific subtypes of elder abuse and ED use, we used same Poisson regression model adjusting for similar factors as above (Table 4). In the fully adjusted model, psychological abuse (RR, 1.98; 95% CI, 1.29-3.00), financial exploitation (RR, 1.59; 95% CI, 1.01-2.52), and caregiver neglect (RR, 2.04; 95% CI, 1.38-2.99) were independently associated with increased rate of ED use. For physical abuse, although the core model adjusting for age and sex, physical abuse is associated with increased rate of ED use (RR, 1.85; 95% CI, 1.01-3.41), after controlling for socioeconomic, health-related, and psychosocial factors, the association was no longer statistically significant (RR, 1.39; 95% CI, 0.74-2.63). Moreover, for elder abuse victims who suffered 2 or more forms of abusive acts, there was a significantly greater rate of ED use (RR, 2.07; 95% CI, 1.43-2.99).

3.4. Potential mediating factors between elder abuse and ED utilization Lastly, we examined the potential mediating factors which might have modified the relationship between elder abuse and ED use outcomes (Table 5). In this analysis, we used similar Poisson model as previously with addition of interaction terms (elder abuse × medical conditions, elder abuse × physical disability, and etc.). In the fully adjusted models, we found that the independent association between reported and confirmed elder abuse and ED use was not mediated by sociodemographic, socioeconomic, health-related factors or psychosocial factors. This finding was similar for all subtypes of elder abuse as well as multiple types of elder abuse.

696

X. Dong, M.A. Simon / American Journal of Emergency Medicine 31 (2013) 693–698

Table 3 Reported and confirmed elder abuse and the risk of ED use Reported elder abuse RR, 95% CI

Age Men Black Education Income Medical conditions Cognitive function Physical function Depressive symptom Social engagement Social network ED utilization Age Men Black Education Income Medical conditions Cognitive function Physical function Depressive symptom Social engagement Social network ED use

Model A

Model B

Model C

Model D

1.04 (1.03-1.04) 0.98 (0.92-1.04)

1.03 1.04 1.04 0.98 1.27

1.01 1.05 0.87 0.99 0.96 1.22 0.82 0.95

2.15 (1.67-2.78) Confirmed elder abuse RR, 95% CI

1.89 (1.45-2.48)

1.71 (1.28-2.31)

1.01 (1.01-1.02) 1.05 (0.98-1.12) 0.88 (0.81-0.95) 1.00 (0.99-1.01) 0.96 (0.95-1.98) 1.21 (1.17-1.26) 0.82 (0.78-0.87) 0.96 (0.95-0.97) 1.05 (1.03-1.07) 0.98 (0.96-1.00) 1.01 (1.00-1.01) 1.76 (1.31-2.37)

1.04 (1.03-1.04) 0.98 (0.92-1.04)

1.03 1.05 1.04 0.98 0.95

1.01 1.05 0.87 0.99 0.96 1.22 0.82 0.95

2.74 (1.99-3.79)

2.55 (1.84-3.52)

4. Discussion In this population-based study of 6864 older adults from an urban geographically defined community, we found that reported and confirmed elder abuse was independently associated with the increased risk of ED use. Moreover, specific subtypes of elder abuse had differential associations with the rate of ED use.

Table 4 Elder abuse subtypes and ED use

Psychological abuse

Physical abuse

Financial exploitation

Caregiver neglect

1 Type of elder abuse

2 or more types of elder abuse

Models

RR

95% CI

P

A B C D A B C D A B C D A B C D A B C D A B C D

2.37 2.14 2.01 1.98 1.85 1.68 1.41 1.39 1.85 1.59 1.49 1.59 2.88 2.54 2.15 2.04 1.82 1.61 1.35 1.39 2.64 2.33 2.13 2.07

1.61-3.50 1.42-3.22 1.33-3.05 1.29-3.00 1.01-3.41 0.89-3.15 0.77-2.58 0.74-2.63 1.28-2.69 1.06-2.39 0.94-2.38 1.01-2.52 2.04-4.07 1.81-3.57 1.47-3.13 1.38-2.99 1.25-2.67 1.08-2.39 0.82-2.21 0.85-2.30 1.89-3.769 1.64-3.29 1.49-3.06 1.43-2.99

b.001 b.001 b.001 .001 .047 .105 .272 .297 .001 .024 .089 .047 b.001 b.001 b.001 b.001 .002 .020 .237 .193 b.001 b.001 b.001 b.001

Models: A, adjusted for age and sex; B, Adjust for A + race, education, and income; C, Adjusted for B + hypertension, diabetes, stroke, cancer, hip fracture, coronary artery disease, MMSE, East Boston Memory Test, East Boston Delayed Recall, and Symbol Digit Modality Test, physical function; D: Adjusted for C + depressive symptoms, social engagement, social network.

(1.03-1.04) (0.98-1.12) (0.97-1.12) (0.97-0.99) (1.09-1.48)

(1.03-1.04) (0.98-1.12) (0.97-1.12) (0.97-0.99) (0.93-0.96)

(1.01-1.02) (0.98-1.13) (0.81-0.94) (0.98-1.01) (0.95-0.98) (1.18-1.27) (0.78-0.87) (0.94-0.97)

(1.01-1.02) (0.98-1.13) (0.81-0.94) (0.98-1.01) (0.95-0.98) (1.18-1.27) (0.78-0.87) (0.94-0.97)

2.36 (1.64-3.39)

1.01 (1.01-1.02) 1.05 (0.98-1.12) 0.87 (0.81-0.94) 1.00 (0.99-1.01) 0.96 (0.95-0.98) 1.21 (1.17-1.26) 0.82 (0.78-0.87) 0.96 (0.95-0.97) 1.05 (1.03-1.07) 0.98 (0.96-1.00) 1.01 (1.00-1.01) 2.33 (1.60-3.38)

Our findings expand the results of prior work of elder abuse and health services utilization. Prior study has matched the Connecticut Social Services Agency data to the Established Populations for the Epidemiologic Studies in the Elderly to identify those who have encounters with adult protective services [11], which had increased risk of nursing home placement. Other studies have also indicated that older adults who have encounters with Adult Protective Services agencies have higher rate of health services utilizations [22,23]. However, a retrospective case-control study of 131 adult protective services cases found no significant differences in health care utilizations compared the matched controls [10]. Our findings expand on the results of prior studies of elder abuse and health services utilization. First, our study is the largest population-based study to demonstrate a significant association between elder abuse increased rate of ED use. The study population is racially/ethnically and socioeconomically diverse and has been well characterized for more than 17 years, which contribute to the potential generalizability of our study findings. Second, our study considered a wide range of potential confounders as older adults who visit ED tend to be older, have lower levels of socioeconomic status, have more medical comorbidities, have lower levels of cognitive and physical health, and have greater levels of psychosocial distress. However, adjusting for these factors did not ameliorate the significant association between elder abuse and rate of ED use. Third, the present study is the first population-based study able to examine the specific subtypes of elder abuse with respect to rate of ED use. Improved understanding of the potential elder abuse–specific associations would contribute to our understanding to the potential causal association between elder abuse and health services utilization. Lastly, ED use is expensive and in part responsible for the soaring health care cost. Screening for elder abuse in the ED could help health care professionals to understand the social context of their emergency presentations and intervene before the deterioration occurs in extremes. As we enter the era of health care reform, improved understanding of factors that increases ED use could also have significant implications for public policy and clinical practice. The causal mechanisms between elder abuse and ED use remain incomplete. We considered a wide range of confounders. However,

X. Dong, M.A. Simon / American Journal of Emergency Medicine 31 (2013) 693–698 Table 5 Interactions terms analyses of elder abuse and potential mediators and ED utilization Parameter estimate

SE

P

Reported elder abuse Age x Gender x Race x Education x Income x Medical conditions x Cognitive function x Physical function x Depressive symptoms x Social network x Social engagement x

−0.04 −0.01 0.03 0.04 0.04 0.07 0.17 0.04 −0.07 −0.03 0.04

0.02 0.39 0.43 0.03 0.07 0.13 0.17 0.04 0.06 0.03 0.08

.061 .987 .947 .308 .518 .609 .305 .257 .271 .171 .656

Confirmed elder abuse Age x Gender x Race x Education x Income x Medical conditions x Cognitive function x Physical function x Depressive symptoms x Social network x Social engagement x

0.01 0.36 −0.67 −0.02 −0.02 0.13 0.11 0.03 −0.05 −0.04 0.02

0.03 0.38 0.49 0.04 0.08 0.17 0.24 0.07 0.07 0.04 0.09

.715 .347 .175 .568 .819 .443 .661 .641 .450 .337 .822

Psychological abuse Age x Gender x Race x Education x Income x Medical conditions x Cognitive function x Physical function x Depressive symptoms x Social network x Social engagement x

−0.01 −0.63 −0.64 −0.03 −0.02 0.27 0.84 0.01 −0.07 −0.02 −0.03

0.03 0.82 0.58 0.06 0.11 0.18 0.47 0.04 0.09 0.03 0.13

.688 .445 .273 .660 .858 .145 .077 .734 .473 .558 .840

Physical abuse Age x Gender x Race x Education x Income x Medical conditions x Cognitive function x Physical function x Depressive symptoms x Social network x Social engagement x

−0.07 0.22 0.13 −0.08 0.08 −0.03 −0.27 0.01 −0.25 −0.02 −0.05

0.06 0.85 1.37 0.09 0.13 0.28 0.59 0.08 0.17 0.04 0.19

.185 .793 .924 .409 .507 .909 .640 .920 .139 .588 .794

Caregiver neglect Age x Gender x Race x Education x Income x Medical conditions x Cognitive function x Physical function x Depressive symptoms x Social network x Social engagement x

−0.09 0.49 0.61 0.01 0.01 −0.09 0.33 0.05 −0.08 −0.03 0.06

0.04 0.48 0.74 0.05 0.11 0.14 0.19 0.04 0.08 0.03 0.164

.009 .298 .407 .875 .946 .509 .100 .197 .321 .242 .713

Financial exploitation Age x Gender x Race x Education x Income x Medical conditions x Cognitive function x Physical function x

0.01 0.38 0.08 0.04 0.08 0.27 −0.01 0.09

0.03 0.51 0.34 0.04 0.09 0.18 0.27 0.06

.901 .453 .833 .417 .350 .128 .989 .113

697

Table 5 (continued) Parameter estimate

SE

P

Depressive symptoms x Social network x Social engagement x

−0.07 −0.06 0.12

0.09 0.05 0.09

.505 .165 .218

1 Type of elder abuse Age x Gender x Race x Education x Income x Medical conditions x Cognitive function x Physical function x Depressive symptoms x Social network x Social engagement x

−0.05 −0.52 0.09 0.06 0.02 0.04 0.09 0.04 −0.05 −0.06 0.12

0.03 0.68 0.58 0.04 0.11 0.24 0.24 0.08 0.09 0.06 0.15

.088 .444 .873 .093 .868 .864 .682 .664 .644 .368 .413

2 or more types of elder abuse Age x Gender x Race x Education x Income x Medical conditions x Cognitive function x Physical function x Depressive symptoms x Social network x Social engagement x

−0.02 0.44 −0.63 −0.01 0.07 0.13 0.24 0.05 −0.08 −0.02 0.01

0.03 0.50 0.76 0.05 0.08 0.16 0.22 0.03 0.08 0.03 0.12

.479 .382 .405 .810 .439 .434 .278 .119 .267 .475 .954

Models adjusted for age, sex, and race, education and income, hypertension, diabetes, stroke, cancer, hip fracture, coronary artery disease, MMSE, East Boston Memory Test, East Boston Delayed Recall, Symbol Digit Modality Test, physical performance testing, depressive symptoms, social network and social engagement.

adjustments for these factors did not change the relationship between elder abuse and rate of ED use. Metabolic abnormalities, nutritional deficiencies, infections, injuries, or trauma may be other factors that could interact with consequences of elder abuse, which could account for the association between elder abuse and rate of ED use. Moreover, severity of chronic medical conditions (ie, ejection fraction of patient with congestive heart failure, or Force Expiratory Volume of a patient with emphysema, etc) could another important factor in determining the causal mechanisms between elder abuse and ED use. However, we do not have measures in our existing data to further elucidate these relations. Future studies are needed to explore the interactions of these variables with elder abuse with respect to ED use. Our study also has limitations. First, our study focused on the reporting of elder abuse as the primary predictor. Elder abuse is under-reported, although the precise rate of under-reporting is unknown, although evidence suggests that only 1 out of 10 cases are reported to APS. At the same time, an unconfirmed case of elder abuse does not exclusively mean there is no evidence of elder abuse, as sometimes caseworker are refused entry or unable to complete assessments. Future studies are needed to uniformly collect elder abuse indicators in a representative population to rigorously examine these associations. Second, ascertainment of ED use may not be complete. A limitation of using CMS data is selective underdetection of some services, including the use of Veterans Administration facilities and some managed care episodes. This underdetection of our outcomes of interest tends to underestimate the strength of association between elder abuse and ED use. Third, our sample size for the subtypes of elder abuse was relatively small, which might have limited more detailed analyses of elder abuse types and ED use. Fourth, there are likely to be additional factors that may account for the increased ED use (substance abuse, infection, injury/trauma, etc). In addition, improved understanding of the interaction between elder abuse and the severities of medical conditions and other health-related factors could contribute toward the causal mechanisms between elder abuse

698

X. Dong, M.A. Simon / American Journal of Emergency Medicine 31 (2013) 693–698

and rate of ED use. Regrettably, we do not have data to consider these additional confounders in our analyses, which may, in part, account for the findings in this report. However, this study sets the foundation for future study of elder abuse to fully examine these issues. Our findings have clinical implications for health care professionals in prevention, detection, and management of elder abuse. In the health care setting, health care professionals focus on preventive care and rigorous management of chronic medical conditions in order to avoid unnecessary ED use. Health care professionals should consider screening for elder abuse among older patients who may have frequent encounters with ED. In addition, health care professionals should be educated on the importance of screening elder abuse and could be integrated into the routine history taking for older patients in clinical settings. Vigilant monitoring of potential elder abuse victims could help clinicians to more closely monitor the patients and devise strategies to prevent unnecessary ED use. Moreover, our findings could have important implications for other disciplines that work with older adults with abuse. In addition to geriatricians, other relevant medical disciplines, nursing, social workers, and social services agencies who work with elder abuse victims or who are at risk for elder abuse could be in unique positions to further monitor factors that may exacerbate the unnecessary ED use. In addition, it is important for all relevant disciplines to monitor the severity or the progression of abusive behaviors towards older adults. Early identification of earlier signs of elder abuse and devising targeted prevention and intervention strategies could prevent deterioration of abusive acts into more severe forms, which in turn could potentially decrease the unnecessary utilization of ED services. Close monitoring and improved understanding of factors that may exacerbate abusive situations could also help clinicians leverage family members, social workers, health professionals, and public health and community organizations to create an interdisciplinary approach to comprehensively care for elder abuse victims. Our findings also have direct implication for future research, by providing the first evidence on elder abuse and rate of ED use in a representative population. Future research is needed to explore temporal associations of targeted risk/protective factors associated with elder abuse in community-dwelling populations. Future studies are needed to explore the longitudinal association of elder abuse to the rate as well as the intensity of other forms of health services utilization. Also needed are studies examining the role of interventions to prevent elder abuse and/or reduce elder abuse severity with respect to health services utilization outcomes in community populations. 5. Conclusion We conclude that elder abuse is independently associated with an increased rate of ED use in a community-dwelling population. In addition, different subtypes of elder abuse had a differential association with the rate of ED use. Future longitudinal investigations are needed to explore the potential causal mechanisms between elder abuse and its subtypes and health services utilization. Future studies

will also be necessary to adequately determine the specific causal mechanisms of the racial/ethnic and gender differences between elder abuse and health services utilizations in the general population. References [1] National Research Council, Elder Mistreatment: Abuse, neglect and exploitation in an Aging America. Washington, DC: The National Academies Press; 2003. [2] Beach SR, Schulz R, Castle NG, Rosen J. Financial exploitation and psychological mistreatment among older adults: differences between African Americans and non-African Americans in a population-based survey. Gerontologist 2010;50(6): 744–57. [3] Acierno R, Hernandez MA, Amstadter AB, Resnick HS, Steve K, Muzzy W, et al. Prevalence and correlates of emotional, physical, sexual, and financial abuse and potential neglect in the United States: The National Elder Mistreatment Study. Am J Public Health 2010;100(2):292–7. [4] Teaster PB, Dugar T, Mendiondo M, Abner EL, Cecil KA, Otto JM. The 2004 Survey of Adult Protective Services: Abuse of Adults 60 Years of Age and Older. www. elderabusecenter.org/pdf/research/apsreport030703.pdf; 2007. Accessed April 4th, 2007. [5] Dong X. Medical implications of elder abuse and neglect. Clin Geriatr Med 2005;21(2):293–313. [6] Dong XQ, Simon M, Mendes de Leon C, Fulmer T, Beck T, Hebert L, et al. Elder selfneglect and abuse and mortality risk in a community-dwelling population. JAMA 2009;302(5):517–26. [7] Dong X, Simon MA, Beck T, McCann J, Farran C, Laumann, et al. Elder abuse and mortality: the role of psychological and social wellbeing. Gerontology 2011;57: 549–58. [8] Burston GR. Granny Battering. BMJ 1975(iii):592. [9] Fitzsimmons S. Elder Abuse–a Case Study. Trends Health Care Law Ethics 1993;8(2):67, 26. [10] Franzini L, Dyer CB. Healthcare costs and utilization of vulnerable elderly people reported to adult protective services for self-neglect. J Am Geriatr Soc 2008;56(4): 667–76. [11] Lachs MS, Williams CS, O'Brien S, Pillemer KA. Adult protective service use and nursing home placement. Gerontologist 2002;42(6):734–9. [12] Bienias JL, Beckett LA, Bennett DA, Wilson RS, Evans DA. Design of the Chicago Health and Aging Project (CHAP). J Alzheimers Dis 2003;5(5):349–55. [13] Evans DA, Bennett DA, Wilson RS, Bienias JL, Morris MC, Scherr PA, et al. Incidence of Alzheimer disease in a biracial urban community: relation to apolipoprotein E allele status. Arch Neurol 2003;60(2):185–9. [14] Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12(3): 189–98. [15] Albert M, Smith LA, Scherr PA, Taylor JO, Evans DA, Funkenstein HH. Use of brief cognitive tests to identify individuals in the community with clinically diagnosed Alzheimer's disease. Int J Neurosci 1991;57(3–4):167–78. [16] Smith A. Symbol Digit Modalities Test Manual-Revised. Los Angeles: Western Psychological; 1984. [17] Katz S, Akpom CA. A measure of primary sociobiological functions. Int J Health Serv 1976;6(3):493–508. [18] Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol 1994;49(2):M85–94. [19] Kohout FJ, Berkman LF, Evans DA, Cornoni-Huntley J. Two Shorter Forms of the CES-D (Center for Epidemiological Studies Depression) depression symptoms index. J Aging Health 1993;5(2):179–93. [20] Radloff L. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385–401. [21] Cornoni-Huntley J, Brock DB, Ostfeld A, Taylor JO, Wallace RB. Established Populations for Epidemiological Studies of the Elderly Resource Data Book (Rep. No. NIH Publication No. 86–2443). Washington, DC: US Department of Health and Human Services; 1986. [22] Schonfeld L, Larsen RG, Stiles PG. Behavioral health services utilization among older adults identified within a state abuse hotline database. Gerontologist 2006;46(2):193–9. [23] Lachs MS, Williams CS, O'Brien S, Hurst L, Kossack A, Siegal A, et al. ED use by older victims of family violence. Ann Emerg Med 1997;30(4):448–54.