Utilization of Formal Support Services for Elder Abuse: Do Informal Supporters Make a Difference?

David Burnes, PhD; Risa Breckman, LCSW; Charles R. Henderson Jr., MA; Mark S. Lachs, MD, MPH; Karl Pillemer, PhD

Gerontologist. 2019;59(4):619-624.

Abstract and Introduction

Abstract

Background and Objectives: Few elder abuse (EA) victims ever seek or receive assistance from formal support services designed to mitigate risk and harm of revictimization. This study examined whether the presence of third-party "concerned persons" in victims' personal social networks plays a role in enabling formal support service utilization.

Research Design and Methods: A representative population-based survey administered to adults (n = 800) in New York State identified 83 EA cases from the past year. Penalized likelihood logistic regression was used to examine the relationship between availability of a concerned person and victim formal support services usage.

Results: EA victims who had a concerned person in their personal life were significantly more likely to use formal EA support services than victims without a concerned person. EA victims who lived with their perpetrator were significantly less likely to use formal services.

Discussion and Implications: Third-party concerned persons represent a critical population to target in efforts designed to promote EA victim help-seeking.

1 2 3 4 5

Next Section

Table 1. Descriptive Characteristics of EA Cases
Characteristics	Cases of EA
Characteristics	(n = 83) n (%)
Predisposing
Victim age
60–74	31 (37.3)
75–84	33 (39.8)
85 or above	18 (21.7)
Victim gender
Male	25 (30.1)
Female	58 (69.9)
Victim race/ethnicity
White/non-Hispanic	52 (62.7)
White/Hispanic	9 (10.8)
African American	13 (15.7)
Other	8 (9.6)
Enabling
Concerned person present
No	18 (21.7)
Yes	62 (74.7)
Victim–perpetrator relationship
Offspring (child or grandchild)	12 (14.5)
Other family	24 (28.9)
Nonfamily	45 (54.2)
Victim–perpetrator living arrangement
Victim lives without perpetrator	50 (60.2)
Victim lives with perpetrator	31 (37.3)
Need
Primary EA type
Financial abuse	34 (41.0)
Emotional abuse	26 (31.3)
Physical abuse	8 (9.6)
Sexual abuse	0 (0)
Neglect	15 (18.1)
EA severity (± SD)	6.8 (2.4)
Dependent Variable
Use of formal EA services
No	54 (65.1)
Yes	21 (25.3)

Note: EA = Elder abuse.

Table 2. Penalized Likelihood Logistic Regression Predicting Use of Formal Support Services Among EA Victims
Independent variables	Odds ratio	95% confidence interval
Predisposing
Victim age (ref. 60–74)
75–84	0.35	0.06–1.98
85 or above	0.14	0.02–0.98
Victim gender (ref. male)
Female	2.90	0.64–13.05
Victim race/ethnicity (ref. White/non-Hispanic)
White/Hispanic	4.41	0.57–34.21
African American	2.17	0.35–13.35
Other	0.97	0.12–8.01
Enabling
Concerned person present (ref. no)
Yes	30.13*	1.94–467.41
Victim–perpetrator relationship (ref. offspring)
Other family	0.73	0.09–6.25
Nonfamily	0.22	0.02–2.29
Victim–perpetrator living arrangement (ref. lives without perpetrator)
Victim lives with perpetrator	0.17*	0.03–0.94
Need
Primary EA type (ref. financial abuse)
Emotional abuse	1.66	0.30–9.20
Physical abuse	3.77	0.31–45.70
Neglect	1.98	0.26–14.97
EA severity (±SD)	1.15	0.82–1.60

Note: EA = Elder abuse. Independent variables demonstrated tolerance (≥0.68) and variance inflation factor (≤1.48) diagnostics indicative of little concern for multicollinearity.
*p < .05, **p < .01, ***p < .001.

Supplemental File: Appendix A. Data Separability – Concerned Person Presence * Formal Support Service Utilization
	Presence of Concerned Person
Frequency Matrix
Use of Formal Support Services	No	Yes
No	15 (20.5%)	38 (52.1%)
Yes	0 (0%)^a	20 (27.4%)

Notes:
^aData separability occurs when there is a cell with no observation within the distribution of responses across independent and dependent variables. It is indicative of an unusually strong association or risk factor (Heinze & Schemper, 2002).
Heinze, G., & Schemper, M. (2002). A solution to the problem of separation in logistic regression. Statistics in Medicine, 21, 2409–2419. doi:10.1002/sim.1047.

Supplemental File: Appendix B. Types of Concerned Persons
Concerned Person	N (%)
Relation of Concerned Person to Victim
Son	12 (14.5)
Daughter	12 (14.5)
Child-in-law	0 (0)
Grandson	2 (2.4)
Granddaughter	0 (0)
Sibling	4 (4.8)
Nephew	3 (3.6)
Niece	1 (1.2)
Husband	0 (0)
Wife	2 (2.4)
Friend	8 (9.6)
Neighbor	5 (6.0)
Other	12 (14.5)