| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a Kent School of Social Work, University of Louisville, KY
b Virginia Commonwealth University School of Social Work, Richmond
c Department of Health Science, Towson University, MD
Correspondence: Ruth Huber, PhD, Kent School of Social Work, University of Louisville, Louisville, KY 40292. E-mail: ruth.huber{at}louisville.edu.
Laurence G. Branch, PhD
 |
Abstract |
|---|
 TOP Abstract Elderly Persons at Risk...MethodsResultsDiscussionReferences |
|---|
Key Words: Elder abuse • Race • Gender • Nursing homes
The Long-Term Care Ombudsman Program was established by the Congress in the early 1970s as a mandate of the Older Americans Act. Referred to as a "community presence in long-term care facilities" (Cherry 1993
, p. 336) ombudsmen identify, investigate, and resolve individual and systems level complaints that affect long-term care residents. As the program has grown, some states have expanded their ombudsmen's scope of practice to include board and care homes as well as home and community-based services. Ombudsmen include both paid staff and volunteers located under various agency and community auspices (Huber, Netting, and Kautz 1996). Interest in the ombudsman program has increased over the years (American Association of Retired Persons 1994; Cherry 1991, Cherry 1993; Connor and Winkelpleck 1990; Huber, Netting, and Paton 1993; Monk, Kaye, and Litwin 1984; Nathanson and Eggleton 1993; National Association of State Units on Aging 1993; Nelson 1995; Nelson, Huber, and Walter 1995; Nelson, Pratt, Carpenter, and Walter 1995; Office of the Inspector General 1993; Schiman and Lordeman 1989a, Schiman and Lordeman 1989b).
Less than a decade ago, the Long-Term Care Ombudsman Program did not have comprehensive reporting systems in every state, even though this was part of its federal mandate. In 1992, Netting, Paton, and Huber found that the Administration on Aging (AoA) received dissimilar reports from the states and that a standardized complaint reporting form was only recommended. In the early 1990s, hand-tabulated data, inconsistent coding, and the impossibility of developing a meaningful report to Congress led researchers and policymakers to push for a more viable reporting system for the program (Kautz, 1990; Kusserow 1991a; Netting et al. 1992; Testimony of E. Chelimsky, 1991). These pleas for accountability were reinforced by the Institute of Medicine's (IOM) study of the Long Term Care Ombudsman Program (Harris-Wehling, Feasley, and Estes 1995).
In 1994 AoA awarded a 2-year grant to our research team to begin working with five pilot states to develop OmTrak, software for the ombudsman program. Since 1996 this and other initiatives have led to tremendous change in the development of computerized ombudsman databases in various states (Huber, Borders, Netting, and Kautz 1997). Among the changes was a new reporting form containing 133 types of complaints (the National Ombudsman Reporting System, or NORS). Now more comprehensive and meaningful reports are being released by AoA, in which aggregated statistics across states provide an overview of ombudsman program activities nationally (Administration of Aging [AoA], 1999). Despite this progress, caution must still be taken in interpreting data within and across states (Huber, Borders, Netting, and Kautz 2000).
Although the new form did not include residents' demographics, six states that had joined the OmTrak effort to computerize their data using the same software agreed to collect residents' gender and race. Researchers working with these states argued that the gender and race of these older persons should be known, even though these data were not required by AoA. Persons responsible for the form's development held that demographic statistics about long-term care residents are readily available and that their collection would be redundant. Researchers replied that existing statistics would not reveal which residents were using ombudsman services and whether there were differences in the types of complaints reported by gender and race.
In this article, we report data from six states that used OmTrak to track data. We focus on the relationship between types of complaints, especially abuse, and race and gender to illustrate that collecting basic demographic data can provide important information about events in the nation's long-term care facilities.
|
Elderly Persons at Risk for Abuse |
|---|
|
TOPAbstractElderly Persons at Risk...MethodsResultsDiscussionReferences |
|---|
). No national statistics exist on abuse in long-term care. The result is that elder abuse is more than likely underreported (Kusserow 1991b). Despite Pillemer and Finkelhor 1988 finding that neither race nor gender is a significant predictor of abuse in long-term care, a small but growing body of evidence calls into question the so-called leveling hypothesis, which holds that old age distributes inequity and vulnerability more equally among the races and genders than is the case in youth and middle age. Instead, previous research seems to support the rival double-jeopardy hypothesis—that old age exacerbates the harms of social inequality, so that racial minorities, and perhaps women, face greater adversity in their advanced age than do Caucasians and men (McPherson 1983).
That gender- and race-related problems surface in long-term care is suggested in a scarce but provocative literature. Evidence suggests that minority racial status is a risk factor for long-stay facility access problems (Falcone and Broyles 1994), increased rates of accidental injury (Weinberg 1998), and placement in troubled, impoverished facilities (Kosberg 1973) that are further away from home settings than is the case for Caucasians (Kosberg and Tobin 1972). Gender is also problematic, but the data are harder to interpret. For example, it has recently been suggested that men in nursing homes may receive better care and more staff attention than women (Claridge, Rowell, Duffy, and Duffy 1995) and that women are more vulnerable to sexual and physical abuse than men (Kosberg and Nahmiash 1996). There is also evidence that women are subjected to more extreme forms of abuse than men (Pillemer and Finkelhor 1988). Griffin and Aitken 1999 maintained that institutional abuse is routine and is overwhelmingly perpetrated by poor women against old women. They concluded that institutional abuse remains hidden because this very non-nurturing, nonfeminine pattern runs contrary to culturally based gender expectations. As such, it is profoundly underreported and underinvestigated. Payne and Cikovic 1995, on the other hand, found that men were more likely to be victims of abuse than women. This is consistent with other analysts writing about elder abuse, but drawing conclusions relevant to institutional settings, who argued that men may be more susceptible to physical abuse than women because men are more likely to exhibit combative behaviors (Tatara, cited in Kosberg and Nahmiash 1996) and may be excluded from "screening protocols" (Kosberg and Nahmiash 1996, p. 34). Regardless, physical abuse seems to be widespread. Pillemer and Moore 1989(pp. 315–316) found that 36% of the 577 nursing home staff they interviewed reported seeing at least one incident of physical abuse, and 81% had seen at least one incident of psychological abuse over the last year. Ten percent of those surveyed had themselves committed acts of physical abuse, and 40% admitted to committing an act of psychological abuse (Pillemer and Moore 1989, p. 317).
Given this history of sketchy evidence of abuse that occurs in nursing homes, we determined that an examination of types of complaints by race and gender could contribute to the literature in two ways: (a) to demonstrate the utility of knowing the race and gender of nursing home residents and (b) to learn whether abuse and neglect are among the five most frequently reported complaints that were investigated by ombudsmen.
Ombudsmen receive complaints before they are referred to the criminal justice system. The very nature of their work puts them in local long-term care facilities on an ongoing basis, as potential troubleshooters and as advocates for vulnerable older residents. The data reported in this article reveal that race may put residents at greater risk of abuse, regardless of gender.
|
Methods |
|---|
|
TOPAbstractElderly Persons at Risk...MethodsResultsDiscussionReferences |
|---|
Definitions
Abuse.
The willful (a) infliction of injury, unreasonable confinement, intimidation, or cruel punishment with resulting physical harm, pain, or mental anguish or (b) deprivation by a person, including a caregiver, of goods or services that are necessary to avoid physical harm, mental anguish, or mental illness (Administration of Aging [AoA], 1998, p. 13).
Complaint.
A problem brought to the attention of the ombudsman for investigation and action. Complaints may be lodged by (or on behalf of) one or more residents of a long-term care facility. One or more complaints—from cold coffee to physical or sexual abuse—constitute a case (AoA, 1998, p. 1).
Origins of Complaints.
Long-term care ombudsmen identify, investigate, and work to resolve complaints that come to their attention in a number of ways. First, residents of long-term care facilities (usually nursing homes) tell ombudsmen about problems they are experiencing when ombudsmen visit their facilities. These complaints can be said to be lodged by the residents themselves. Second, other people may call the ombudsman and file a complaint. Such complaints are said to be lodged on behalf of residents. Third, in most states, ombudsmen routinely visit nursing homes at varying times to observe the facilities and the nature of the care being provided. These complaints are then lodged by the ombudsmen themselves. Exactly when ombudsmen visit facilities varies among states and even among local programs within a state. Ombudsmen may visit on a fairly routine schedule unless they begin to hear reports of, for example, neglect on the night shift. They may then make an unannounced visit during that shift to gain a different perspective. Most states do not dictate when facility visits should occur. An Oregon Administrative Rule, for example, states that "the ombudsman or designee may obtain access to a long term care facility at any time considered necessary and reasonable by the ombudsman or designee for the purpose of performing the duties of the ombudsman or designee" (Oregon Administrative Rule 114, 1989). Other ombudsman programs even make it their policy not to establish predictable visiting routines.
Verified Complaints.
After investigation (e.g., phone calls, visits) the complaint is determined to be generally true or accurate. This verification, in most states, is not held to a legal standard such as "beyond reasonable doubt."
Disposition.
Each complaint is assigned a disposition code when the case is closed. 1 = resolution required legal or legislative action that the ombudsman was unable to provide; 2 = not resolved; 3 = the complaint was referred to another agency, (e.g., the health department), but no action was taken; 4 = the complaint was referred to another agency and no response was received; 5 = no action needed/appropriate; 6 = partially resolved; and 7 = the complaint was fully resolved.
It is important to stress that ombudsman complaint data are not duplicated by any other governmental body, except perhaps for the small percentage of the complaints that are referred to regulators for investigation or punitive action. However, even in states where ombudsmen are mandated to report abuse, this number is reported to be less than 20% (Nelson et al. 1995), which includes all referrals, not just governmental. Consequently, it can be safely assumed that up to 85% or more of all ombudsman complaints are not recorded elsewhere.
Moreover, although exact numbers are not available, a large percentage of ombudsman complaints arise from a process of routine visits, often during odd hours and weekends when other government agents are rarely seen. It is fair to say, then, that ombudsman data may represent a more accurate picture of routine problems and issues in long-stay institutions than the annual snapshots, for example, that make up the state health department surveys, which constitute the only other major source of problem identification available to the public.
Data Analyzed in This Study
Six state ombudsmen using OmTrak agreed that we could analyze their 1996 data as long as state names were not revealed, given the highly political climate in which these programs operate. Although we realize that readers might be interested in knowing the states used, our purpose in this article is to demonstrate the value of collecting resident demographics, not to compare data across states.
Data from these six states were sent to us on computer disks. Considerable data cleaning was required before all six databases were ready for merging into a single Statistical Package for the Social Sciences file for analysis. The names of residents had been removed before data were sent to us, and we also removed the names of individual facilities. Data entry errors were corrected if the information was available and eliminated from current analyses if not.
OmTrak is a relational database, which means that data can be collected on two different levels: case and the specific complaints within the case. On the case level, information is obtained about the resident and complainant. On the complaint level, one complainant may lodge several complaints relative to one case. In this article we are concerned with these two levels of data: (a) case specific and (b) complaint specific.
These data manipulations completed, we began data analyses to answer the question: What are the relationships between race and gender, and types of complaints lodged, verified, and fully resolved (for residents of long-term care facilities in six states using NORS)? The findings we discuss later show why it is important to collect these resident characteristics.
|
Results |
|---|
|
TOPAbstractElderly Persons at Risk...MethodsResultsDiscussionReferences |
|---|
|
|
Second, the five most frequently lodged complaints were the same for both Caucasian men and women, albeit in different rank order. Nearly equal percentages of the complaints for minority men, however (about 25%), pertained to physical abuse and loss of dignity and respect. Further examination of the column for minority men also shows that gross neglect (20%) and appeared in their five most frequently reported complaints but did not appear in the top five for minority women or Caucasians. A connection between staff being unresponsive (16%) and gross neglect (20% for minority men) and 15% each of staff unresponsive and unattended symptoms for minority women is not hard to make (Table 1 ).
Third, the second highest percentage of the five most frequently reported complaints lodged for minority women pertained to physical abuse (16%, Table 1 ), which was about the same percentage as for Caucasian women. Wandering (16%) also appeared among the six (due to the tie) most frequently reported complaints for minority women, but for none of the other groups. One can imagine a common theme in the five types of complaints that appeared for minority men (gross neglect and staff unresponsive) and minority women (staff unresponsive, symptoms unattended, and wandering) that did not appear in the five most frequent complaints for either Caucasian men or women: one of inattention to minority residents (Table 1 ).
Lastly, loss of dignity/respect, accidents, and physical abuse were found in the five most frequently reported complaints lodged for all four groups of elderly persons (Table 1 ). Certainly the data from this study raise additional questions about gender and race that have not hitherto been addressed.
Verification and Disposition of Complaints by Race and Gender
There were additional differences by race and gender. After complaints are lodged ombudsmen investigate circumstances to verify the complaints (or not). Of all complaints lodged on behalf of residents (for whom race and gender data were available), higher percentages of complaints for minorities were verified (Fig. 2). Conversely, even though significantly higher percentages of complaints lodged for minorities were verified, significantly lower percentages of those verified complaints were fully resolved for minorities (Fig. 3).
|
|
It is somewhat troubling that although higher percentages of complaints lodged on behalf of minorities were verified, lower percentages of those verified complaints were fully resolved to the satisfaction of the residents. Even complaints for which race and gender were not available resulted in a higher percentage fully resolved than complaints for minority residents. Future research should examine these phenomena to discern whether there is some form of racial or gender discrimination at play in these areas. Huber and colleagues 2000 made the point that it is vitally important to have a working analyst-practitioner link when analysts are attempting to understand what data mean. States report data differently for different reasons: state ombudsmen's philosophies, state statutes, political climate, and so forth. Even though the emphasis in this article is on the benefits of reporting demographic information, the point must be made that data must be interpreted with great care.
|
Discussion |
|---|
|
TOPAbstractElderly Persons at Risk...MethodsResultsDiscussionReferences |
|---|
Practice Implications
What can an ombudsman do with these data to improve services? Having demographic data enables local and state ombudsmen to not only react to complaints, but also to watch for known patterns and trends from the data to prevent further maltreatment of residents in long-term care facilities. For example, if the local ombudsman knows that minority men, as a group, are most likely to be neglected, the ombudsman can watch for early signs and preventive measures can be taken. Similarly, if local data indicate that minority women are at higher risk for having their symptoms go unattended, perhaps even in a certain facility, prevention measures can bring symptoms to the attention of the facility staff and administration before they escalate into dangerous stages, for example, bed sores. Analyses of types of complaint by demographic information can add the advantage of known indicators of maltreatment in time to prevent it.
In keeping with the ombudsman program's community education mandate, local programs might develop culturally/racially sensitive materials to help identify special minority risk factors and to encourage abuse reporting among groups who are fearful of retaliation or skeptical of government reporting. Moreover, because racial differences between staff and patients may be linked to less demonstrative interactions (Kahana, Kahana, and Young 1985) and are no doubt the cause of some "personality conflicts," which are, in turn, linked to "abuse interactions" (Kosberg and Nahmiash 1996, p. 32), ombudsman programs in communities with strong minority or immigrant presence might benefit from diversity/sensitivity training.
Policy Implications
As it stands now, both the general public and policymakers have a poor grasp of the influence of race and gender on the quality of life and care in the various long-stay settings. Any coordinated effort to identify and assess protective needs and objectives on the basis of race and gender is now virtually impossible. Consequently, we make the following recommendations. First, the Assistant Secretary for Aging should publish guidance documents encouraging states with large minority populations to collect race data for purposes of educational advocacy. Second, the Assistant Secretary should direct all states to assess the feasibility of collecting race-based data by the end of this decade in anticipation of important demographic shifts. Third, the Assistant Secretary should require all states to collect gender data. Fourth, the Assistant Secretary should take steps to support gender- and race-based research.
Determining Whether to Collect Data on Race and Gender
Among the six states that were willing to collect data on race and gender, there are still many complaints about which the gender and race of the older resident are not known. It is not surprising that gender is missing less often than race, given that gender is more easily discerned when complaints are received. Residents' names or pronouns used (he or she) by the complainant often indicate gender. Race, on the other hand, requires an ombudsman to inquire or see the resident, and there is the potential for an inaccurate response given that the person lodging the complaint may not be correct about a resident's race. We do not contend that tracking race is easy, but we are suggesting that knowing race is valuable in determining social injustices that may disproportionately impact elderly people in minority groups.
Some ombudsmen have argued that collecting racial data in states with smaller numbers of minorities is not helpful because there are so few persons served. Certainly arguments that racial minorities make up a small percentage of the nation's nursing homes can be used to minimize the importance of examining race as a data element. Moreover, it is true that some states have so little racial diversity within their long-term care systems that requiring the collection of this variable would currently render very few situations in which long-term care residents are not Caucasian, although gender would certainly be important. However, to rely on this argument as a reason for not collecting racial data is shortsighted at best. The near-future America will have a different racial hue. Non-Hispanic Caucasians will become just another minority exceeded in number by other racial groups that will make up America's fast-evolving racial mosaic.
Can policymakers rationally ignore race in developing protocols and programs for tomorrow's increasingly burdened long-term care system? At the very least, it would seem that, even now, in states serving larger numbers of racially diverse elderly persons the unit responsible for overseeing the ombudsman program at the state level would want to have these data collected. At this point, few states collect information on gender or race, regardless of whom they serve or the racial diversity of their populations.
On the other hand, in states serving large numbers of racially diverse clients, not collecting these data could actually mask the fact that serious complaints such as physical abuse are more likely to occur for racial minority elderly persons, particularly men. If ombudsmen are to perform advocacy roles, it would seem imperative that they know when certain groups of elderly persons are disproportionate targets for complaints such as physical abuse and gross neglect.
The ombudsman reporting system has the potential to identify specific facilities in which large numbers of complaints are being filed. Without demographic data to reveal resident characteristics, it is not possible to identify those facilities that may be particularly insensitive to persons of different races and gender. The data presented in this article reveal that different types of complaints may be associated with different subgroups of elderly persons. They also indicate that fully resolving complaints for some groups may be more difficult than for others. Knowing this information would allow a state ombudsman to target certain types of complaints for careful study, to determine why these discrepancies exist. However, if these data are not collected, there is no way to document what is happening. Reliance on anecdotal data is not enough to effect change.
Conclusion: Whose Responsibility Is It to Collect These Data?
In the last 5 years tremendous progress has been made in the development of NORS. As recently as 1994, it was not possible to conduct cross-state analyses of complaint data. AoA aggregated data from states willing to participate and reported the results to Congress. Today, AoA requires only aggregated data from states, but data are more reliable, standard reporting forms are used, and individual states have the capacity to analyze their respective databases.
We do not contend that AoA should begin asking for raw data and become analysts of a national ombudsman database. With decentralization and the diversity among states, creation of a national database may not be feasible. Even if AoA were to ask for raw data, it lacks sufficient staff and interest to fully analyze data across states. This has never been AoA's intent and it is not likely to change now. We do believe, however, that AoA is responsible for raising consciousness, for providing a broad vision of the issues, and for advocating for increasingly diverse cohorts of older Americans. AoA is in a leadership position to amplify state and regional analyses that need to be brought to national attention. At a minimum, AoA could raise the consciousness of state leaders so that ombudsmen in states serving large numbers of racial minority residents in long-term care would be aware of the importance of knowing how these residents fare in comparison with their Caucasian counterparts. Similarly, it would be important to know what happens to women and men, whether they are treated differently, and to what problems they might be particularly susceptible. If gender and race are indeed tied to particular complaint types, ombudsmen need to be vigilant in monitoring what happens to older residents.
For researchers interested in older persons, ombudsman state-level databases represent a rich resource for analysis. The role of the academy is to work with state and local ombudsmen to conduct data analyses that AoA cannot (and is not expected to) do. For example, researchers can team with state ombudsmen to clean raw data, which can explain small discrepancies in data sent to AoA. Researchers and ombudsmen can work together to allow states to review statewide trends and then compare differences across substate regions, including other variables (such as gender and race) that are not required by AoA but that are important to understanding diversity within a particular state. Researchers and ombudsmen can share data with one another to see if there are trends within the state, the region, or even within local facilities, for example, on what shifts abuse is occurring. These trends could be shared with other state agencies that are attempting to find reliable data on the events in the lives of older residents in long-term care facilities, particularly with an eye to gender and racial sensitivity.
|
Acknowledgments |
|---|
Received for publication February 10, 2000. Accepted for publication August 23, 2000.
|
References |
|---|
|
TOPAbstractElderly Persons at Risk...MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. L. Estes, D. M. Zulman, S. C. Goldberg, and D. D. Ogawa State Long Term Care Ombudsman Programs: Factors Associated With Perceived Effectiveness Gerontologist, February 1, 2004; 44(1): 104 - 115. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. D. Allen, W. C. Klein, and C. Gruman Correlates of Complaints Made to the Connecticut Long-Term Care Ombudsman Program: The Role of Organizational and Structural Factors Research on Aging, November 1, 2003; 25(6): 631 - 654. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
2 (4, N = 23,787) = 339.351, p < .001.