NRC in 2003 and make a meaningful research and policy investment to enhance well-being in the years we have gained.
Terry T. Fulmer, Ph.D., R.N.
Bouve College of Health Sciences, Northeastern University
XinQi Dong, M.D., M.P.H.
Chinese Health, Aging, and Policy Program,
Rush Institute for Healthy Aging
Rush University Medical Center
Elder Neglect Review: The Research
Inadequate progress has been made in our understanding of elder mistreatment and particularly the subtype of elder neglect. Three areas of investigation in elder mistreatment research have made advances in knowledge, improving our understanding in areas including the incidence and prevalence of elder mistreatment (Cisler et al., 2010), self-neglect (Dyer et al., 2007; Dong et al., 2009, 2010a,b), and resident-to-resident mistreatment (Lachs et al., 2007). However, there continues to be negligible advancement in our understanding of neglect of older people by other individuals. The purpose of this paper is to review studies that have data specific to neglect by others, document trends in the data, underscore the concern related to the limited number of scientists engaged in this important area, and begin to determine how intervention studies can be developed to reduce and eliminate neglect to older individuals. Several investigators have documented that elder neglect is a potentially fatal syndrome and that of all individuals in the elder mistreatment category, those in the neglect category have significant risk for morbidity and mortality. A systematic review of all research databases was conducted to ascertain the state of the science related to elder neglect, and those papers with original data are reported here. The study of neglect appears to have three main approaches: examination of cases using prospective longitudinal cohort studies against other databases, screening for cases in practice settings, and postmortem analysis.
Elder neglect has been defined by the National Research Council as “an omission by responsible caregivers that constitutes neglect under applicable federal or state law” (NRC, 2003, p. 39). Others have defined it as “the
refusal or failure by those responsible to provide food, shelter, health care, or protection for a vulnerable elder” (National Center on Elder Abuse, 2013).
Elder Neglect Literature Review
Early studies with data on neglect include the elder abuse and neglect findings from Three Model Projects (Godkin et al., 1989), in which a case comparison study was completed that examined 62 individuals from Worcester, Massachusetts, and 65 from Boston, Massachusetts. Active neglect was found in 20 percent of the cases. Neglect had the strongest relationship to dependency needs of the victims and neglect cases had significant problems for cognitive and physical functioning. Individuals who were neglected were less likely to be burdensome and stressful, but caregivers had significant stressors in their own lives.
In a 1984 study, Fulmer and Cahill conducted elder mistreatment assessments on all patients arriving at a busy emergency room using screening by the nurses on that floor who had been trained to use the particular elder assessment instrument (Fulmer and Cahill, 1984). In this work, investigators chose a “good, fair, poor” scale to elicit nursing concerns related to suspected elder mistreatment. This strategy failed in that in the majority of cases, nurses selected the midline (fair) and were reticent to make judgments about suspected mistreatment. Subsequently, a body of work used a 4-point scale of “very good, good, poor, very poor” and was found to be much preferred in that nurses would lean to the positive or negative in order to better understand suspected cases of elder abuse and neglect. In each of these screens, neglect was by far the most predominant selected category (over abuse, neglect, exploitation, or abandonment).
Pillemer and Finkelhor (1988) conducted the first large-scale random sample survey of elder mistreatment involving 2,020 elderly persons from Boston. Of those individuals, 63 persons were maltreated and 7 out of 63 came out of the category of neglect. Thirty-two per 1,000 older individuals in this study were labeled as mistreated, with 4 in 1,000 labeled as neglected. The perpetrators in the neglect cases were husband–wife (29 percent), daughter–mother (29 percent), and other (42 percent), which is a strong indicator that the phenomena was not well explained in this study. The most common characteristics of neglect in this study included victims who were female, were divorced, had a spouse and children, were in poor health, and were without a helper. Neglect victims were commonly in poor health and reported that they did not have close contacts on whom they could rely. This work has proven extremely valuable in anchoring subsequent research.
Fulmer and Ashley examined 107 cases of suspected neglect and referred to an elder mistreatment team compared with 146 non-referred cases (Fulmer and Ashley, 1989). An expert panel was asked to select indicators of neglect from the Elder Assessment Instrument (EAI). Nine neglect indicators were selected. A factor analysis of those items indicated that nutritional deficits, alterations in skin integrity, and alterations in elimination patterns were the three major clusters constituting neglect symptoms.
In a study of patients presenting to the emergency department over a 6-month period (Fulmer and Degutis, 1992; Fulmer et al., 1992), 3,153 recorded visits for older adults represented 1,975 older individuals (a significant recidivism to the emergency room). Of those, 56 percent were female with a mean age of 76.5 years. The group was predominantly white. Of those included in the group, 126 individuals (4 percent) were recorded as having some sort of elder mistreatment. Of those, 55 percent were recorded as neglect, further evidence that neglect is the most prevalent category of mistreatment. In that study, associations with the outcome of mistreatment include non-white, not married, and without insurance, as well as with some form of delirium or dementia. These associations with mistreatment represent vulnerable older adults. Associations with abuse and neglect included those who were non-white and non-married. However, it is interesting that neglect was associated with delirium and dementia, while abuse was not. Again, this speaks to the vulnerable older adult.
Using a prospective longitudinal cohort database (Lachs et al., 1996; Harrell et al., 2002), Lachs and others compared APS records and determined that in a group of 2,812 individuals, 184 cases were suspected of some form of mistreatment, with 81 cases substantiated. Of those, 30 individuals (64 percent) experienced neglect by another party. In a follow-up study on mortality of elder mistreatment, Lachs and colleagues (1997) documented that of 176 older adults seen by APS over 9 years, 30 (17 percent) were for cases of neglect.
In an emergency room pilot feasibility study conducted over a 3-week period, 180 patients ages 70 and older were screened using the EAI to determine how well it worked in busy practice environments. Of the 180 patients who met the age criteria, 36 agreed to be screened and 7 screened positive for neglect. The purpose of this pilot was to determine if nurses were able to screen with accuracy versus an elder mistreatment team. There was 70 percent accuracy for detecting neglect between the nurses and the expert team (Fulmer et al., 2000).
Pavlik and colleagues (2001) used a statewide database to describe the case reports received in 1997 by the Texas Department of Protective and Regulatory Services-Adult Protective Services Division (TDPRS-APS). They documented more than 62,000 allegations of adult mistreatment and neglect. Neglect accounted for 80 percent of the allegations. The incidence
of cases in the TDPRS-APS increased sharply after age 65. The prevalence was 1,310 individuals/100,000 for all abuse types. Dyer et al. (2007) asked APS workers to describe elder neglect in an effort to define a working case definition. Using a structured interview, the authors asked APS workers in Texas to identify indicators of elder neglect that they see in practice. The most common indicator (36 percent) was cited to be derangement in the client’s environment. This involved the condition of the home, including debris and food left on the floors, as well as the outside environment of the home. The second most identified factor (18.4 percent) was personal hygiene. The workers were also asked to define both self-neglect and caregiver neglect. The majority distinguished caregiver neglect as simply having a caregiver present. The study concluded that elder neglect could be described by deficiencies in environmental cleanliness, cognition, and personal hygiene. Physicians and other home care workers should consider the home environment as well as personal hygiene when screening for elder neglect (Harrell et al., 2002; Bitondo Dyer et al., 2005).
Chokkanathan and Lee conducted a probability study in urban India to determine cases of mistreatment. Of 400 older adults surveyed, 56 (14 percent) reported mistreatment, and of the 14 percent, 4.3 percent reported neglect versus 5 percent physical abuse. This again underscores the presence of neglect across cultures (Chokkanathan and Lee, 2005). Oh et al. (2006) studied the prevalence of elder abuse and neglect in South Korea, and they interviewed 15,230 adults ages 65 and older. The percentage of older adults who experienced any type of elder abuse was 6.3 percent. An abusive act was defined as occurring more than once and occurring more than two or three times per month. Of participants, 2.4 percent reported neglect. Furthermore, they identified the main abuser for each type of abuse. For neglect, the main abuser was identified as the daughter-in-law at 38 percent.
Dong et al. (2007) investigated the prevalence of elder abuse and neglect in an urban Chinese population. A cross-sectional study was performed in a major urban medical center in Nanjing, China. A total of 412 participants completed the survey and 145 participants (35 percent) screened positive for elder abuse and neglect. The mean age of the victims was 69 years, and 59 percent were male. Caregiver neglect was the most common form of abuse, followed by financial exploitation, psychological abuse, physical abuse, sexual abuse, and abandonment. Thirty-six percent of the victims suffered multiple forms of abuse and neglect. In the logistic regression analyses of the data, female gender, lower education, and
lower income were demographic risk factors associated with elder abuse and neglect. Dong and colleagues reported that a better understanding of these and additional risk factors associated with elder abuse and neglect in older Chinese people is needed. They further examined depression as a risk factor for elder abuse and neglect in this population. Depression was assessed using the Geriatric Depression Scale and direct questions were asked regarding abuse and neglect experienced by the elderly since the age of 60; 412 patients completed the survey. The mean age of the participants was 70; 34 percent were female. Depression was found in 12 percent of the participants, and elder abuse and neglect were found in 35 percent. After multiple logistic regression, feeling of dissatisfaction with life (OR, 2.92; 95 percent CI, 1.51-5.68, p < 0.001), often being bored (OR, 2.91; CI, 1.53-5.55, p < 0.001), often feeling helpless (OR, 2.79; CI, 1.35-5.76, p < 0.001), and feeling worthless (OR, 2.16; CI, 1.10-4.22, p < 0.001) were associated with increased risk of elder abuse and neglect. Multiple logistic regression modeling showed that depression is independently associated with elder abuse and neglect (OR, 3.26; CI, 1.49-7.10, p < 0.003). One limitation of this study is that there was not sufficient power to clearly separate out abuse from neglect and the two forms of mistreatment were reported together. However, these findings suggest that depression is a significant risk factor associated with elder abuse and neglect among Chinese elderly. Subsequent studies should address the two forms independently (Dong, 2008).
The prevalence of physical and financial abuse among Turkish elderly in two different socioeconomic strata was reported. Keskinoglu et al. (2007) selected a district of low socioeconomic status, and documented physical and financial abuse prevalence was 1.5 and 2.5 percent, respectively, while among the elderly in the district of high socioeconomic status, it was 2 and 0.3 percent, respectively. However, the prevalence of elder neglect in the two districts was 27.4 and 11.2 percent, respectively. Prevalence of neglect was associated with infrequent contact with relatives, little or no income, and fewer years of education among the elderly in the low socioeconomic district. In the high socioeconomic district, neglect was associated with fewer years of education, poor health status, and having chronic status. They concluded that the prevalence of abuse among the elderly living in the two different districts was low, but nearly one-fifth of elderly people were exposed to neglect. Here, again the phenomenon of neglect holds up as construct internationally. This is important because it has been postulated that neglect may be identified in some cultures and not in others. However, there is a growing body of evidence that the construct of neglect holds across cultures.
Clinical Assessment Approaches: Use of Autopsies
Obviously, neglect takes place in the context of the older adult and some other person. Using a dyadic approach would allow the investigators to interview both the elder and the caregiver (Fulmer et al., 2005a). Fulmer and colleagues (2005a) examined vulnerability and risk profiling for elder neglect. In that study, 5,159 older adults were approached for inclusion and 3,669 were screened and recruited through five emergency departments in New York City and Tampa, Florida. Of those screened, 405 met the inclusion criteria (11 percent) and agreed to participate. Of the 405 older adults and caregivers who were eligible to participate, 165 (41 percent) completed the face-to-face, in-home interview.
In all, the demographics of the neglect versus the non-neglect groups were the same, except that the neglect group had fewer people living in the home (p < .04), were more likely to be Hispanic or Latino by self-report (p < .02), had a health problem that limited activities (p < .05), and were more likely to be Medicare recipients (p < .03). Caregivers of the neglect group were more likely to be Hispanic or Latino by self-report (p < .04), were less likely to have health problems that required a doctor’s attention (p < .02), and were less likely to be on Medicare (p < .01). The mean ages of the two caregiver groups were not significantly different. The data supported the idea that vulnerable older adults who have cognitive impairment have increased functional needs and were more likely to suffer from neglect. Furthermore, the absence of social support, childhood trauma, as well as the personality trait “openness” on the NEO personality inventory scale (McCrae and Costa, 1987) created the association with the neglect outcome (Fulmer et al., 2005b). In another study that enrolled patients from busy primary care clinics, including a dental clinic, older adults were screened with the EAI, and the prevalence of mistreatment was 4 percent, with neglect as the most prevalent subcategory. In that study, associations were documented between cognitive function and caregiver neglect. Furthermore, this study demonstrated that elder neglect can be screened for and examined in busy primary care clinics (Russell et al., 2012).
Akaza et al. (2003) documented neglect and abuse by examining 15 autopsy cases in Gifu Prefecture, Japan, between 1990 and 2000. Using a retrospective approach of all the cases in which the victim was 65 years or older and autopsied, 15 victims were classified as elder abuse victims: 5 men and 10 women. The victims ranged in age from 66 to 87 years (mean age, 74.5 years). The types of abuse were as follows: physical abuse, 13 cases; emotional abuse, 5 cases; neglect, 4 cases; and financial abuse, 3 cases. In eight cases, the victims were subjected to two or more types of abuse. The cause of death of the victims varied with the type of abuse. In the physical abuse cases, subdural hemorrhage was the most common cause, followed
by other violence-related deaths and hypothermia. In the neglect cases, the victims died of either starvation or suffocation after the aspiration of food into the airway. In the domestic abuse cases, one of the victim’s sons was the most common perpetrator, and little or no income was considered to be a risk factor for perpetrators. In the neglect cases, dementia and difficulty in performing activities of daily living were considered to be risk factors for victims, in addition to living in social isolation.
Collins and Presnell (2007) also used autopsy cases in a retrospective review of individuals 65 years and older over a 20-year period, and determined that eight cases of suspected neglect resulted in death. The causes of death included sepsis due to extreme dehydration and severe decubitus ulcers. Unfortunately there were no follow-up studies to this analysis, but the work and the approach represent another important way to research neglect and its potential outcomes. Interestingly, the dehydration and decubitus ulcers often occur due to the conditions of nutritional status and skin integrity (Fulmer and Ashley, 1989).
In another example of postmortem analysis, Shields and colleagues (2004) conducted a retrospective 10-year study of elder abuse and neglect cases in Kentucky and Indiana. They reviewed postmortem records and also reviewed examinations under the Clinical Forensic Medicine Program. The authors described 74 postmortem cases, of which 22 deaths were suspicious for neglect. Of those, the most common cause of death was bronchopneumonia. They identified 22 living victims, and of those 3 were identified as suffering from neglect. The authors concluded that a multidisciplinary team approach is crucial to identifying cases of elder neglect, and the importance of the forensic team and Medical Examiner’s office should not be ignored (Shields et al., 2004).
National Elder Mistreatment Study: United States
In 2010, a study was published that has set the stage for future research, and has created a new baseline for research in the field. Acierno et al. (2010), in the National Elder Mistreatment Study from 2010, used a definition of neglect as “failing to meet the following needs: transportation, household needs (i.e., cooking and cleaning), taking care of financial matters, and obtaining medication.” They reported on a representative sample using random digit dialing across geographic strata, analyzing 5,777 respondents. They reported a prevalence of 4.6 percent for emotional abuse, 1.6 percent for physical abuse, 0.6 percent for sexual abuse, 5.1 percent for potential neglect, and 5.2 percent for financial abuse (Acierno et al., 2010). Approximately 1 in 10 respondents reported some mistreatment in the past year. They found that the risk for potential neglect was predicted by low income, lack of social support, racial minority status, and poor health. In
general, the study found that the most significant predictors of abuse and neglect were low social support and exposure to a traumatic event. This aligns with data from a Fulmer study in 2005 (Fulmer et al., 2005a). The associations between cognitive function and elder abuse and neglect were documented previously by others using the Chicago Health and Aging Project database (Dong et al., 2012c). Neglect is second only to financial abuse in this report and clearly these data should trigger investigators to respond to the important unanswered questions on the subject of neglect by others.
Summary of the Evidence
In summary, evidence since the 1980s would indicate that the overall prevalence of mistreatment is approximately 4 to 10 percent, and neglect is the most prevalent subcategory. Neglect can have dire clinical consequences and increase the mortality in older adults. Few investigators have taken on the topic of neglect by others in a caregiving relationship, and the need for additional research on this topic is clear. Assessing for elder neglect in the clinical setting should be a standard of care. Of concern, Cooper et al. (2008) conducted a systematic review of the prevalence of elder abuse and neglect using multiple databases and independent raters and identified 49 studies meeting the inclusion criteria, of which only 7 used measures for which reliability and validity had been assessed and established. Clinicians need guidance in addressing this phenomenon and clinical assessment instrumentation must improve. Several of the current instruments are either flawed psychometrically or require disrobing of the patient for full assessment, which is impractical in many settings.
Likely, most cases of neglect go undetected, and currently we have no intervention studies, except for Wiglesworth et al.’s (2006) work related to bruising and Dyer et al.’s (1999) team approach to cases of abuse and neglect. To this point, research on elder neglect has predominantly taken the form of documenting the presence in longitudinal cohort studies, and screening for incidence and prevalence of neglect and reviewing autopsy cases. We argue that it is essential for investigators to take on the extremely challenging topic of elder neglect by trusted others, whether they are formal or informal caregivers, and begin to understand how we can create intervention strategies to prevent neglect. By doing so, it is highly likely that we can decrease the morbidity and mortality associated with this phenomenon.