This article was downloaded by: [Michigan State University] On: 04 March 2015, At: 01:29 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Elder Abuse & Neglect Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/wean20
Longitudinal Prevalence and Correlates of Elder Mistreatment Among Older Adults Receiving Home Visiting Nursing a
b
Bruce Friedman PhD, MPH , Elizabeth J. Santos MD, MPH , Dianne V. c
b
b
Liebel RN, PhD, MSEd , Ann J. Russ PhD & Yeates Conwell MD a
Departments of Public Health Sciences and Psychiatry, University of Rochester, Rochester, New York, USA b
Department of Psychiatry, University of Rochester, Rochester, New York, USA c
Click for updates
School of Nursing, University of Rochester, Rochester, New York, USA Accepted author version posted online: 10 Sep 2014.Published online: 24 Nov 2014.
To cite this article: Bruce Friedman PhD, MPH, Elizabeth J. Santos MD, MPH, Dianne V. Liebel RN, PhD, MSEd, Ann J. Russ PhD & Yeates Conwell MD (2015) Longitudinal Prevalence and Correlates of Elder Mistreatment Among Older Adults Receiving Home Visiting Nursing, Journal of Elder Abuse & Neglect, 27:1, 34-64, DOI: 10.1080/08946566.2014.946193 To link to this article: http://dx.doi.org/10.1080/08946566.2014.946193
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Downloaded by [Michigan State University] at 01:29 04 March 2015
Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions
Journal of Elder Abuse & Neglect, 27:34–64, 2015 Copyright © Taylor & Francis Group, LLC ISSN: 0894-6566 print/1540-4129 online DOI: 10.1080/08946566.2014.946193
Longitudinal Prevalence and Correlates of Elder Mistreatment Among Older Adults Receiving Home Visiting Nursing
Downloaded by [Michigan State University] at 01:29 04 March 2015
BRUCE FRIEDMAN, PhD, MPH Departments of Public Health Sciences and Psychiatry, University of Rochester, Rochester, New York, USA
ELIZABETH J. SANTOS, MD, MPH Department of Psychiatry, University of Rochester, Rochester, New York, USA
DIANNE V. LIEBEL, RN, PhD, MSEd School of Nursing, University of Rochester, Rochester, New York, USA
ANN J. RUSS, PhD and YEATES CONWELL, MD Department of Psychiatry, University of Rochester, Rochester, New York, USA
The objectives of this study were to identify elder mistreatment (EM) prevalence among a cohort of older adults receiving visiting nurse care in their homes, determine EM subtypes, and identify factors associated with EM. EM data were collected by nurses during monthly home visits for up to 24 months. It took the nurses a mean of 10.5 visits to discern EM. Fifty-four (7.4%) of 724 patients were identified as mistreated, of which 33 had enough information to subtype the EM. Of these 33, 27 were victims of neglect, 16 of psychological abuse, and 10 of financial exploitation, and 17 suffered more than one type. Among the entire sample, 11 variables were positively correlated with EM presence. Nurses visiting older adults in their homes should be aware that their patients are, as a group, vulnerable to EM, and that the factors identified here may be specific markers of greater risk.
Address correspondence to Bruce Friedman, Division of Health Policy and Outcomes Research, Department of Public Health Sciences, University of Rochester Medical Center, 265 Crittenden Boulevard, CU 420644, Rochester, NY 14642, USA. E-mail: bruce_friedman@ urmc.rochester.edu 34
Elder Mistreatment Among Aged Home Care Patients
35
KEYWORDS elder mistreatment, elder abuse, home care, neglect, psychological abuse, financial exploitation
Downloaded by [Michigan State University] at 01:29 04 March 2015
INTRODUCTION The National Elder Mistreatment Study reported that 11.4% of adults age 60 years and older living in the community in 2008 reported emotional, physical, or sexual mistreatment in the past year or current potential neglect (Acierno et al., 2010). However, little is known about elder mistreatment (EM) prevalence among persons receiving the services of a visiting nurse in their home or about factors associated with EM among this population. Much visiting nurse care provided to older persons is delivered to patients who are experiencing difficulty with or are dependent in activities of daily living (ADL; basic activities such as bathing or dressing; Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963) or instrumental activities of daily living (IADL; more complex activities such as ordinary housework or meal preparation; Lawton & Brody, 1969). Each day, about 850,000 Americans age 65 years and older that have difficulty with or are dependent in one or more ADLs receive home health care (Caffrey, Sengupta, Moss, Harris-Kojetin, & Valverde, 2011). EM prevalence can be expected to be higher for this group than for the general population age 65 years and older, since a number of studies (but not all) have found ADLs to be significantly associated with EM (National Research Council, 2003). Because of the nature of and the amount of time spent in nurse-patient contact, visiting nurses are in a position to detect mistreatment and, potentially, to intervene (Anetzberger, 2005; Lachs & Pillemer, 2004). This notion is supported by an earlier study in which different categories of health care professionals were asked about elder abuse (Costa, 1984). Visiting nurses and home care personnel reported EM cases more often than other disciplines. Thus far, a small number of qualitative studies (Sandmoe & Kirkevold, 2011, 2013; Sandmoe, Kirkevold, & Ballantyne, 2011; Saveman, Hallberg, & Norberg, 1993; Saveman, Hallberg, Norberg, & Eriksson, 1993; Saveman, Norberg, & Hallberg, 1992) and a few quantitative studies (Buri, Daly, Hartz, & Jogerst, 2006; Ogioni et al., 2007; Post et al., 2010; Shugarman, Fries, Wolf, & Morris, 2003) of EM using visiting nurses or home care samples have been conducted. However, these studies have collected data at only a single point in time. To our knowledge, no research including information about visiting nurse interactions with older adults over extended periods of time has examined EM prevalence and correlates. EM data collected by nurses making monthly home visits for up to 24 months for the Medicare Primary and Consumer-Directed Care Demonstration were evaluated. The objectives in this secondary analysis of
36
B. Friedman et al.
Demonstration data were to: (a) identify EM prevalence among this longitudinal cohort of older adult patients with functional disabilities, (b) determine the prevalence of EM subtypes (i.e., physical abuse, psychological abuse, sexual assault, financial exploitation, and neglect), and (c) identify correlates associated with EM.
METHODS
Downloaded by [Michigan State University] at 01:29 04 March 2015
Sample The Medicare Primary and Consumer-Directed Care Demonstration (1994–2003, with enrolled Medicare patients during 1998–2002; Friedman, Wamsley, Liebel, Saad, & Eggert, 2009) was a randomized controlled trial designed to test whether utilizing either visiting nurses, a voucher that covers goods and services not traditionally covered by Medicare, or the combination of these two options would improve health care outcomes and reduce health care costs for patients with disabilities who are at high risk of having high health care expenditures. The voucher could not be used to purchase home visits by a nurse. It was restricted to nonskilled services. Patients were recruited from the practices of 307 primary care physicians in 19 counties: 8 in New York State, 6 in West Virginia, and 5 in Ohio. A total of 1,605 subjects was randomized to 1 of 4 intervention arms: treatment as usual, the intervention including visiting nurses, the voucher, or both nurse and voucher interventions (the combination group). All Demonstration participants were required to have a high level of disability, defined as needing or receiving help for at least two ADLs or three IADLs. They also had to have received recent significant health services use, defined as having been a hospital or nursing home inpatient, or having received Medicare home health care during the previous year, or having had at least two emergency room visits during the previous 6 months. Of the 1,605 patients enrolled, 20% died and an additional 13% dropped out or were disenrolled during the 24-month period each person was eligible to participate in the Demonstration. For the present study, only the 724 subjects aged ≥ 65 years who were enrolled in the Demonstration’s nurse only and combination (nurse plus voucher) intervention arms were included, because the study’s primary interest was in the role that visiting nurses play in EM detection and management.
Data Collection Three data sources from the Demonstration were utilized for the present study. Extensive questionnaires were administered to study subjects and/or their primary informal (nonpaid) caregivers at baseline. The study forms were administered by interviewers, primarily nurses, whose only job was
Downloaded by [Michigan State University] at 01:29 04 March 2015
Elder Mistreatment Among Aged Home Care Patients
37
to administer these forms. The interviewers received extensive training; their work was carefully assessed (including the review of recorded patient/caregiver interviews); and inter-rater reliability was determined. None of the interviewers were the visiting nurses who carried out the intervention. The interviewers administered a cognitive screen to each patient. The cognitive screen assessed word repetition (book, watch, and table), 5-minute word recall, and nonresponse to basic questions about subjective health status, functional status, and life satisfaction. If the patient failed the cognitive screen (110 of 724 patients, or 15.2%), the questionnaire was administered to the caregiver. The reliability of the data collected from such caregivers was very good. The Demonstration had a very careful, elaborate process to ensure valid and reliable data collection. The interviewers read the questions to the participants and/or their primary informal caregivers and recorded the responses on paper data collection forms. Upon completion, the forms were scanned into a database. There was no information about EM at baseline since the forms contained no EM questions. The nurses, who visited patients in their homes on average once a month, completed a computerized Nurse Activity List that noted specific characteristics of the older adults’ life, such as EM presence or absence, chronic medical conditions, need for help with ADLs, and health care utilization. During and after each patient visit, the nurses documented Nurse Contact Notes, including comments on EM, patients’ physical and mental health, and their social circumstances. The nurse identified the presence of EM as well as the perpetrators during the 24-month period after each patient entered the intervention phase of the Demonstration. (The first phase consisted of planning and development, the intervention phase was the second phase, and the third phase consisted of staff phase-down and data cleaning.) For the present study, perpetrators were determined on the basis of information in the Nurse Contact Notes.
Dependent Variable—Elder Mistreatment EM was defined in the Demonstration’s Nurse Procedure Manual as any of the following: (a) physical abuse as “inappropriate physical contact such as: striking, pinching, kicking, shoving, bumping, sexual molestation”; (b) mistreatment as “inappropriate use of medication, isolation, inappropriate use of physical restraints, and inappropriate use of chemical restraints”; or (c) neglect as “failure to provide timely, safe, consistent, adequate, and appropriate care and services to participants, such as: nutrition, medication, therapies, sanitary clothing and surroundings and daily living activities.” For each visit to the patient’s home, the nurse was required to enter, in the intervention’s computerized Nurse Activity List, whether or not the patient was being mistreated. The nurses determined EM through observation; by being told about EM by patients, family members, and neighbors; and by the discovery of
Downloaded by [Michigan State University] at 01:29 04 March 2015
38
B. Friedman et al.
physical evidence on examination of the older adults. Examples are a nurse entering the home and seeing the older person left alone without food or placed in a chair in the corner and ignored. While mistreatment was based on the definition in the Nurse Procedure Manual, the Nurse Activity List did not include EM types. In the present study the dependent variable for the presence of EM (1 = yes, 0 = no) used in the quantitative analyses discussed below was constructed by first identifying whether the Nurse Activity List for each patient indicated EM, and then confirming EM presence from information in the Nurse Contact Notes. While the Nurse Procedure Manual definition included self-neglect, when the EM cases were confirmed self-neglect was not included as part of EM. In order to determine the type(s) of EM endorsed in the Nurse Activity List, two of us (EJS, AJR) carefully reviewed the Nurse Contact Notes, recording presence or absence of each of five subtypes of mistreatment: physical abuse, psychological abuse, sexual assault, financial exploitation, and neglect. No cases of self-neglect or abandonment were included as part of neglect. That is, when the subtype neglect was identified in the Nurse Contact Notes it was defined as intentional or unintentional neglect, but not self-neglect. Nurse Contact Notes were also used to characterize the relationships that were implicated, that is, who were the perpetrators. The analysis of Nurse Contact Notes’ content elucidated details (a) regarding the relationships involved in instances endorsed for EM and (b) what precisely (i.e., what actions or inaction) was seen by nurses to constitute EM in those situations.
Independent Variables—Correlates of EM The National Research Council (NRC) Panel to Review Risk and Prevalence of Elder Abuse and Neglect (National Research Council, 2003) identified three categories of EM risk factors: (a) “risk factors validated by substantial evidence” (living arrangement, social isolation, dementia, intraindividual characteristics of abusers [mental illness, hostility, alcohol abuse], and abuser dependency); (b) “possible risk factors” (gender, relationship of victim to perpetrator, personality characteristics of victims, and race); and (c) “contested risk factors” (physical impairment of the older person, victim dependence and caregiver stress, and intergenerational transmission; National Research Council, 2003). Variables measuring many but not all of these factors were available, using information collected in the baseline interview. The following domains were used in the study’s quantitative analyses: sociodemographics, family and social factors, health insurance, physical health, mental health, disability status, and Demonstration features. Sociodemographics included age, gender, race, education, annual household income, and urban or rural residence. The income question and categories were the nine categories used in Medicare’s 1998 Health of Seniors
Downloaded by [Michigan State University] at 01:29 04 March 2015
Elder Mistreatment Among Aged Home Care Patients
39
Survey (Centers for Medicare and Medicaid Services, 1998). For the present study’s analyses these were collapsed into four categories. There were four family and social factors: marital status, lives with others, reports having 0 or 1 close friends, and reports having 0 or 1 close relatives. The present study included one variable for health insurance: Medicare supplemental insurance. Since all demonstration participants were enrolled in Medicare, we did not include a variable indicating Medicare enrollment. Supplemental Medicare coverage was included because we thought that people who were without such coverage might be of a lower socioeconomic class, and hence could be more likely to suffer from EM. Supplemental insurance was not used to extend the number of Medicare home health care visits. Physical health was measured by the number of chronic conditions (constructed from a list of 13 conditions the patient reported as having been diagnosed by a physician) and the SF-36 Physical Component Summary (PCS) score (Ware, Kosinski, & Keller, 1994). The SF-36 Health Survey has been used in thousands of studies (http://www.sf-36.org/tools/SF36.shtml). Its validity and reliability have been demonstrated in older and chronically ill populations similar to the sample of the present study (Ware, Snow, Kosinski, & Gandek, 1993; Ware, Kosinski, & Keller, 1994). Mental health was assessed by four measures: the Nun Study life satisfaction question (Greiner, Snowden, & Greiner, 1996), the SF-36 Mental Component Summary (MCS) score (Ware, Kosinski, & Keller, 1994), the Cognitive Performance Scale (CPS) (Morris et al., 1994), and the MiniInternational Neuropsychiatric Interview (MINI) Major Depressive Episode (MDE) module (Sheehan et al., 1998). The Minimum Data Set CPS (Morris et al., 1994) was developed for the U.S. Department of Health and Human Services and is used for all nursing home patients whose care is paid for by Medicare or Medicaid. It consists of five items. The CPS score consists of integers ranging from 0 (intact) to 6 (very severe impairment). It has been found to be valid for a home care population similar to the sample in the present study with an R2 = 0.81 between the CPS and Mini Mental State Exam scores (Landi et al., 2000). The CPS score is calculated using a decision-tree. Because of this, it is not a domain-sampling psychometric scale of a latent construct. Standard internal consistency reliability measures, therefore, do not apply. The MINI-MDE (Sheehan et al., 1998) has been shown to be valid and reliable overall when measured against the psychiatric patient version of the Structured Clinical Interview for DSM-III-R, with a kappa of 0.55 and a sensitivity and specificity of .77 and .79, respectively, for Major Depressive Disorder (Sheehan et al., 1997). The MINI-MDE consists of nine questions. If one or both of the first two questions are answered yes, seven additional yes-no questions are asked. A major depressive episode is identified if at least five of the nine questions are answered in the affirmative.
Downloaded by [Michigan State University] at 01:29 04 March 2015
40
B. Friedman et al.
Because the MINI-MDE uses a decision-tree, it is not a domain-sampling psychometric scale of a latent construct, and standard internal consistency reliability measures do not apply. Disability status was measured by both ADLs and IADLs. There were questions for both ADL/IADL difficulty and dependence. A simple additive scale with a range of 0–6 was constructed from the six ADLs in the ADL difficulty questions of the Health of Seniors (HOS) Survey (Centers for Medicare and Medicaid Services, 1998). (While we collected data for all 86 HOS questions, we only used a small number of these questions in the present study.) Internal consistency reliability (Cronbach’s alpha) for the six-item HOS ADL difficulty scale in the present sample = 0.754. A similar additive scale was made for the IADLs from the IADL difficulty items of the HOS Survey (Centers for Medicare and Medicaid Services, 1998). Its internal consistency reliability in the present sample was 0.743. Six-item additive ADL and IADL dependence scales were constructed from the ADL and IADL dependence questions in the Outcome and Assessment Information Set B (Shaughnessy, Crisler, & Schlenker, 1997). In the present sample, the Cronbach’s alphas of those two scales were 0.761 and 0.755, respectively. Three demonstration features were included among the independent variables: study site (West Virginia/Ohio or New York), intervention group (nurse only or combination), and number of study days (length of time the patient participated in the intervention phase of the demonstration). Table 1 presents the variable name, description, and instrument source of the independent variables.
Analyses Statistical analyses consisted of descriptive statistics, bivariate analysis, and logistic regression analysis. The descriptive statistics consisted of univariate analyses and are depicted in Table 2. Bivariate analyses were used to examine the association between EM (yes or no) and the independent variables presented above. These analyses consisted of the Chi Square test for categorical variables and the t-test or Analysis of Variance (ANOVA) for continuous variables. Four logistic regression models were estimated to examine the association between EM (yes or no) as the dependent variable and the independent variables. Models 1 and 2 included variables for ADL and IADL dependence while Models 3 and 4 included variables for ADL and IADL difficulty. Difficulty and dependence were analyzed separately because they are fundamentally different constructs measured with very different questions. In Models 1 and 3 the ADL and IADL variables were numbers (counts) of ADLs and IADLs, while Models 2 and 4 included individual ADLs and IADLs, such as dressing and shopping. A value of p < .05 was used to indicate significant association. Stata version 12 (Stata Corporation, College Station, TX) was employed for all statistical calculations.
41
Annual household income < $5,000 income $5,000–$9,999
High school graduate or GED Some college or 2-year degree 4-year college graduate More than a 4-year college degree > High school graduate
Education 8th grade or less Some high school, but did not graduate < High school graduate
SOCIODEMOGRAPHICS Age (in years) Female gender Race American Indian/Alaskan Native Asian or Pacific Islander Black or African American White Another race or multiracial Non-White race
Variable
TABLE 1 Independent Variables
= = = = = Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0
= = = = = No No No No No
= = = = Yes, Yes, Yes, Yes,
0 0 0 0
= = = = No No No No
1 = Yes, 0 = No 1 = Yes, 0 = No
1 1 1 1
1 = Yes, 0 = No 1 = Yes, 0 = No
1 1 1 1 1
65–99 1 = Yes, 0 = No
When Data Were Collected
In Regression Models
(Continued)
1 = some college or 2-year degree, 4-year college graduate, or more than a 4-year college degree, 0 = other
1 = 8th grade or less, or some high school, but did not graduate, 0 = other 1 = Yes, 0 = No
1 = American Indian/Alaskan Native, Asian or Pacific Islander, Black or African American, or another race or multiracial, 0 = White
65–99 1 = Yes, 0 = No
Variable Description
Downloaded by [Michigan State University] at 01:29 04 March 2015
42
Living arrangement Living alone Living in a household with spouse only Living in a household with spouse and others Living in a household with relatives
FAMILY AND SOCIAL FACTORS Marital status Married Widowed Divorced Separated Never married Other
Urban (metropolitan statistical area)
$10,000–$19,999 $20,000–$29,999 $30,000–$39,999 $40,000–$49,999 $50,000–$79,999 $80,000–$99,999 $100,000+ $30,000+
< $10,000 income
Variable
TABLE 1 (Continued)
= = = = = = = Yes, Yes, Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0 0 0
= = = = = = = No No No No No No No
= = = = = Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0
= = = = = No No No No No
1 = Yes, 0 = No
1 = Yes, 0 = No 1 = Yes, 0 = No 1 = Yes, 0 = No
1 1 1 1 1
1 = Yes, 0 = No
1 1 1 1 1 1 1
When Data Were Collected
In Regression Models
1 = divorced, separated, or never married, 0 = other
1 = Yes, 0 = No 1 = Yes, 0 = No
1 = $30,000–$39,999, $40,000–$49,999, $50,000–$79,999, $80,000–$99,999, or $100,000+, 0 = other 1 = Yes, 0 = No
1 = < $5,000 income or $5,000–$9,999, 0 = other 1 = Yes, 0 = No 1 = Yes, 0 = No
Variable Description
Downloaded by [Michigan State University] at 01:29 04 March 2015
43
PHYSICAL HEALTH Chronic conditions Hypertension or high blood pressure Angina or coronary artery disease Congestive heart failure Myocardial infarction or heart attack Other heart conditions Stroke Emphysema, asthma, or COPD Crohn’s disease, ulcerative colitis, or inflammatory bowel disease
HEALTH INSURANCE Medicare Supplemental Insurance
Number of friends that feels close to None One friend Two to five friends Six to nine friends Ten or more friends 0 or 1 close friend Number of relatives that feels close to None One relative Two to five relatives Six to nine relatives Ten or more relatives 0 or 1 close relative
Living in a household with nonrelatives only Other Lives with others
Yes, Yes, Yes, Yes, Yes,
= = = = =
1 1 1 1 1
0 0 0 0 0
0 0 0 0 0 = = = = =
= = = = = No No No No No
No No No No No
1 1 1 1 1 1 1 1
= = = = = = = = Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0 0 0 0
= = = = = = = = No No No No No No No No
1 = Yes, 0 = No
Yes, Yes, Yes, Yes, Yes,
= = = = =
1 1 1 1 1
1 = Yes, 0 = No 1 = Yes, 0 = No
1 = Yes, 0 = No
(Continued)
1 = none or one relative, 0 = other
1 = none or one friend, 0 = other
1 = living in a household with spouse only, spouse and others, relatives, or nonrelatives only, or other, 0 = living alone
Downloaded by [Michigan State University] at 01:29 04 March 2015
44
Major depressive episode
Definite cognitive impairment
Borderline cognitive impairment
SF-36 Mental Component Summary (MCS) score Cognitive impairment
High life satisfaction
MENTAL HEALTH Life satisfaction
Arthritis of hip or knee Arthritis of hand or wrist Sciatica Diabetes, high blood sugar, or sugar Any cancer (other than skin cancer) Number of chronic conditions SF-36 Physical Component Summary (PCS) score
Variable
TABLE 1 (Continued)
= = = = = Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0
= = = = = No No No No No
The 9 Mini-International Neuropsychiatric Interview (MINI) Major Depressive Episode (MDE) module questions (See Sheehan, et al., 1998)
The SF-36 questions (See Ware et al., 1993) The 5 Cognitive Performance Scale questions (See Morris et al., 1994)
1 = excellent, 2 = very good, 3 = good, 4 = fair, 5 = poor
The SF-36 questions (See Ware et al., 1993)
1 1 1 1 1
When Data Were Collected
In Regression Models
1 = Cognitive Performance Scale score of 1, 0 = other 1 = Cognitive Performance Scale score of 2–6, 0 = other 1 = Yes, 0 = No
1 = excellent, very good, or good, 0 = fair or poor 0 (worst)–100 (best)
0–13 0 (worst)–100 (best)
Variable Description
Downloaded by [Michigan State University] at 01:29 04 March 2015
45
Toileting dependence
Eating dependence
Dressing dependence
Number of ADLs with any difficulty Activities of daily living (ADL) dependence Bathing dependence
Walking difficulty
Transferring difficulty
Toileting difficulty
Eating difficulty
Dressing difficulty
DISABILITY STATUS Activities of daily living (ADL) difficulty Bathing difficulty
= has difficulty, 3 =
= has difficulty, 3 =
= has difficulty, 3 =
= has difficulty, 3 =
= has difficulty, 3 =
= has difficulty, 3 =
1 = independent, 2 = some human assistance, 3 = uses bedside commode, 4 = uses bedpan or urinal, 5 = totally dependent
1 = independent without assistive devices, 2 = independent with assistive devices, 3 = some human assistance, 4 = requires another person, 5 = totally by another person 1 = independent, 2 = clothing laid/handed out, 3 = some human sssistance, 4 = entirely dependent 1 = independent, 2 = intermittent assistance, 3 = assisted throughout meal, 4 = supplemented by feeding tube, 5 = fed entirely by tube/other
1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty or has
or has
or has
or has
or has
or has
(Continued)
1 = any dependence (clothing laid/ handed out, some human assistance, or entirely dependent), 0 = no dependence 1 = any dependence (intermittent assistance, assisted throughout meal, supplemented by feeding tube, or fed entirely by tube/other), 0 = no dependence 1 = any dependence (some human assistance, uses bedside commode, uses bedpan or urinal, or totally dependent), 0 = no dependence
1 = any dependence (some human assistance, requires another person, or totally by another person), 0 = no dependence
1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 0–6
Downloaded by [Michigan State University] at 01:29 04 March 2015
46
Number of IADLs with any difficulty
Telephone difficulty
Shopping difficulty
Money management difficulty
Medication management difficulty
Meal preparation difficulty
Number of ADLs with any dependence Instrumental activities of daily living (IADL) difficulty Ordinary housework difficulty
Walking dependence
Transferring dependence
Variable
TABLE 1 (Continued)
1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 0–6
1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty
some difficulty,
some difficulty,
some difficulty,
some difficulty,
some difficulty,
or great difficulty),
or great difficulty),
or great difficulty),
or great difficulty),
or great difficulty),
or great difficulty),
1 = any dependence (minimal assistance/device, can bear weight during transfer, unable to bear weight during transfer, bedfast—able to turn, or bedfast—unable to turn), 0 = no dependence 1 = any dependence (device/human assistance, walks only with assistance, chairfast—wheels self, chairfast—cannot wheel self, or bedfast), 0 = no dependence 0–6
1 = independent, 2 = minimal assistance/device, 3 = can bear weight during transfer, 4 = unable to bear weight during transfer, 5 = bedfast— able to turn, 6 = bedfast—unable to turn 1 = independent, 2 = device/human assistance, 3 = walks only with assistance, 4 = chairfast—wheels self, 5 = chairfast—cannot wheel self, 6 = bedfast
some difficulty,
In Regression Models
When Data Were Collected
Variable Description
Downloaded by [Michigan State University] at 01:29 04 March 2015
47
Number of study days
DEMONSTRATION FEATURES West Virginia/Ohio site Intervention combination group
Number of IADLs with any dependence 1 = yes, 0 = no 1 = combination (nurse + voucher) group, 0 = nurse only group 0–732 days in the Demonstration intervention phase
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
Telephone dependence
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
Money management dependence 1 = independent, 2 = some help, 3 = only with help, 4 = by others only
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
Medication management dependence
Shopping dependence
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
Meal preparation dependence
Instrumental activities of daily living (IADL) dependence Ordinary housework dependence
with
with
with
with
with
with
1 = yes, 0 = no 1 = combination (nurse + voucher) group, 0 = nurse only group 0–732 days in the Demonstration intervention phase
1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 0–6
Downloaded by [Michigan State University] at 01:29 04 March 2015
48 363 301 60 444 240 148 599
FAMILY AND SOCIAL FACTORS Marital status Widowed Married Other Lives with others Reports having 0 or 1 close friend (n = 718) Reports having 0 or 1 close relative (n = 718)
HEALTH INSURANCE Medicare Supplemental Insurance
42.3 30.2 27.5
306 219 199
82.7
50.1 41.6 8.3 61.3 33.4 20.6
32.0 34.8 19.2 14.0 28.7
(7.5) 69.9 3.3
%
79.8 506 24
232 252 139 101 208
N
Study Population (N = 724)
SOCIODEMOGRAPHICS Age—mean (SD) Female Non-White Education < High school graduate High school graduate > High school graduate Annual household income < $10,000 income $10,000–$19,999 $20,000–$29,999 $30,000+ Rural (non-MSA)
Variable
TABLE 2 Patient Baseline Characteristics (N = 724)
50
37 14 3 35 26 15
12 23 8 11 44
26 15 13
81.2 44 2
N
%
92.6
68.5 25.9 5.6 64.8 48.2 27.8
22.2 42.6 14.8 20.4 81.5
48.2 27.8 24.1
(7.9) 81.5 3.7
Mistreated (n = 54)
Downloaded by [Michigan State University] at 01:29 04 March 2015
549
326 287 57 409 214 133
220 229 131 90 472
280 204 186
79.6 462 22
N
81.9
48.7 42.8 8.5 61.0 32.2 20.0
32.8 34.2 19.6 13.4 70.4
41.8 30.4 27.8
(7.4) 69.0 3.3
%
Not Mistreated (n = 670)
.046
.584 .017 .176
.019
.085
.166
.140 .054 .868 .656
P Value
49
DISABILITY STATUS Activities of Daily Living (ADL) Number of ADLs with any difficulty—mean (SD) Any ADL difficulty Walking Transferring Bathing Dressing Toileting Eating Number of ADLs with any dependence—mean (SD) Any ADL dependence Walking Bathing Transferring Dressing Eating Toileting
MENTAL HEALTH High (ex., very good, good) life satisfaction SF-36 Mental Component Summary score—mean (SD) Borderline cognitive impairment (Cognitive Performance Scale score = 1) Definite cognitive impairment (Cognitive Performance Scale score = 2–6) Major depressive episode (n = 718)
PHYSICAL HEALTH SF-36 Physical Component Summary score—mean (SD) Number of chronic conditions—mean (SD)
26.7 15.3
193 110
80.9 63.7 59.9 50.0 30.9 19.9 (1.8) 64.1 47.2 40.6 28.6 19.5 11.3
586 461 434 362 224 144 2.3 464 342 294 207 141 82
(1.8)
41.7
302
3.0
59.4 (10.9)
(2.2)
4.5 420 47.8
(9.0)
30.6
34 24 18 11 11 5
42 37 28 27 17 12 2.1
3.0
11
21
23
39 48.6
4.9
30.0
63.0 44.4 33.3 20.4 20.4 9.3
77.8 68.5 51.8 50.0 31.5 22.2 (1.7)
(1.8)
20.4
38.9
42.6
72.2 (11.0)
(2.2)
(9.9)
Downloaded by [Michigan State University] at 01:29 04 March 2015
430 318 276 196 130 77
544 424 406 335 207 132 2.3
3.0
99
172
279
391 47.7
4.5
30.6
64.2 47.5 41.2 29.2 19.4 11.5
81.2 63.3 60.6 50.0 30.9 19.7 (1.8)
(1.8)
14.9
25.7
41.6
58.4 (10.9)
(2.2)
(9.0)
(Continued)
.858 .669 .258 .165 .863 .618
.539 .442 .207 1.000 .929 .655 .363
.883
.284
.035
.875
.046 .559
.262
.662
50
88.1 85.1 66.8 58.2 46.4 24.6 (1.8) 83.4 82.5 61.7 59.2 44.6 17.3
638 616 484 421 336 178 3.5 604 597 447 429 323 125 228 381 594
DEMONSTRATION FEATURES West Virginia/Ohio Site Intervention Combination group Number of study days—mean (SD)
31.5 52.6 (221)
(1.7)
%
3.7
N
Study Population (N = 724)
Instrumental Activities of Daily Living (IADL) Number of IADLs with any difficulty—mean (SD) Any IADL difficulty Ordinary housework Shopping Meal preparation Money management Medication management Telephone use Number of IADLs with any dependence—mean (SD) Any IADL dependence Shopping Household chores Meal preparation Money management Medication management Telephone use
Variable
TABLE 2 (Continued)
34 35 627
42 43 36 31 29 14
45 45 38 30 29 16 3.6
3.8
N
%
63.0 64.8 (190)
77.8 79.6 66.7 57.4 53.7 25.9
83.3 83.3 70.4 55.6 53.7 29.6 (2.0)
(1.9)
Mistreated (n = 54)
Downloaded by [Michigan State University] at 01:29 04 March 2015
194 346 591
562 554 411 398 294 111
593 571 446 391 307 162 3.5
3.7
N
29.0 51.6 (223)
83.9 82.7 61.3 59.4 43.9 16.6
88.5 85.2 66.6 58.4 45.8 24.2 (1.7)
(1.7)
%
Not Mistreated (n = 670)
High school graduate $10,000–$19,999 annual income $20,000–$29,999 annual income $30,000+ annual income Urban residence FAMILY AND SOCIAL FACTORS Widowed Other marital status Lives with others Reports having 0 or 1 close friend Reports having 0 or1 close relative HEALTH INSURANCE Medicare Supplemental Insurance PHYSICAL HEALTH SF-36 PCS score SF-36 PCS score squared SF-36 PCS score cubed Number of chronic conditions MENTAL HEALTH High (Ex-VG-Gd) life satisfaction SF-36 MCS score
Variable 0.94−1.05 0.84−3.90 0.25−13.94 0.42−2.25 0.34−2.00 1.38−8.58 0.65−9.26 1.18−13.30 0.61−5.05 2.08−11.63 0.77−23.31 1.15−8.07 1.25−4.84 0.81−3.94 0.98−8.72 0.25−0.63 1.01−1.04 1.00−1.00 0.98−1.28 1.05−5.40 0.80−1.20
4.92 4.25 3.04 2.46 1.78 2.93 0.40 1.03 1.00 1.12 2.38 0.98
OR 95% CI
0.99 1.81 1.86 0.97 0.83 3.44 2.46 3.97 1.76
OR
2.09 −0.18
.037 .857
Journal of Elder Abuse & Neglect Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/wean20
Longitudinal Prevalence and Correlates of Elder Mistreatment Among Older Adults Receiving Home Visiting Nursing a
b
Bruce Friedman PhD, MPH , Elizabeth J. Santos MD, MPH , Dianne V. c
b
b
Liebel RN, PhD, MSEd , Ann J. Russ PhD & Yeates Conwell MD a
Departments of Public Health Sciences and Psychiatry, University of Rochester, Rochester, New York, USA b
Department of Psychiatry, University of Rochester, Rochester, New York, USA c
Click for updates
School of Nursing, University of Rochester, Rochester, New York, USA Accepted author version posted online: 10 Sep 2014.Published online: 24 Nov 2014.
To cite this article: Bruce Friedman PhD, MPH, Elizabeth J. Santos MD, MPH, Dianne V. Liebel RN, PhD, MSEd, Ann J. Russ PhD & Yeates Conwell MD (2015) Longitudinal Prevalence and Correlates of Elder Mistreatment Among Older Adults Receiving Home Visiting Nursing, Journal of Elder Abuse & Neglect, 27:1, 34-64, DOI: 10.1080/08946566.2014.946193 To link to this article: http://dx.doi.org/10.1080/08946566.2014.946193
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Downloaded by [Michigan State University] at 01:29 04 March 2015
Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions
Journal of Elder Abuse & Neglect, 27:34–64, 2015 Copyright © Taylor & Francis Group, LLC ISSN: 0894-6566 print/1540-4129 online DOI: 10.1080/08946566.2014.946193
Longitudinal Prevalence and Correlates of Elder Mistreatment Among Older Adults Receiving Home Visiting Nursing
Downloaded by [Michigan State University] at 01:29 04 March 2015
BRUCE FRIEDMAN, PhD, MPH Departments of Public Health Sciences and Psychiatry, University of Rochester, Rochester, New York, USA
ELIZABETH J. SANTOS, MD, MPH Department of Psychiatry, University of Rochester, Rochester, New York, USA
DIANNE V. LIEBEL, RN, PhD, MSEd School of Nursing, University of Rochester, Rochester, New York, USA
ANN J. RUSS, PhD and YEATES CONWELL, MD Department of Psychiatry, University of Rochester, Rochester, New York, USA
The objectives of this study were to identify elder mistreatment (EM) prevalence among a cohort of older adults receiving visiting nurse care in their homes, determine EM subtypes, and identify factors associated with EM. EM data were collected by nurses during monthly home visits for up to 24 months. It took the nurses a mean of 10.5 visits to discern EM. Fifty-four (7.4%) of 724 patients were identified as mistreated, of which 33 had enough information to subtype the EM. Of these 33, 27 were victims of neglect, 16 of psychological abuse, and 10 of financial exploitation, and 17 suffered more than one type. Among the entire sample, 11 variables were positively correlated with EM presence. Nurses visiting older adults in their homes should be aware that their patients are, as a group, vulnerable to EM, and that the factors identified here may be specific markers of greater risk.
Address correspondence to Bruce Friedman, Division of Health Policy and Outcomes Research, Department of Public Health Sciences, University of Rochester Medical Center, 265 Crittenden Boulevard, CU 420644, Rochester, NY 14642, USA. E-mail: bruce_friedman@ urmc.rochester.edu 34
Elder Mistreatment Among Aged Home Care Patients
35
KEYWORDS elder mistreatment, elder abuse, home care, neglect, psychological abuse, financial exploitation
Downloaded by [Michigan State University] at 01:29 04 March 2015
INTRODUCTION The National Elder Mistreatment Study reported that 11.4% of adults age 60 years and older living in the community in 2008 reported emotional, physical, or sexual mistreatment in the past year or current potential neglect (Acierno et al., 2010). However, little is known about elder mistreatment (EM) prevalence among persons receiving the services of a visiting nurse in their home or about factors associated with EM among this population. Much visiting nurse care provided to older persons is delivered to patients who are experiencing difficulty with or are dependent in activities of daily living (ADL; basic activities such as bathing or dressing; Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963) or instrumental activities of daily living (IADL; more complex activities such as ordinary housework or meal preparation; Lawton & Brody, 1969). Each day, about 850,000 Americans age 65 years and older that have difficulty with or are dependent in one or more ADLs receive home health care (Caffrey, Sengupta, Moss, Harris-Kojetin, & Valverde, 2011). EM prevalence can be expected to be higher for this group than for the general population age 65 years and older, since a number of studies (but not all) have found ADLs to be significantly associated with EM (National Research Council, 2003). Because of the nature of and the amount of time spent in nurse-patient contact, visiting nurses are in a position to detect mistreatment and, potentially, to intervene (Anetzberger, 2005; Lachs & Pillemer, 2004). This notion is supported by an earlier study in which different categories of health care professionals were asked about elder abuse (Costa, 1984). Visiting nurses and home care personnel reported EM cases more often than other disciplines. Thus far, a small number of qualitative studies (Sandmoe & Kirkevold, 2011, 2013; Sandmoe, Kirkevold, & Ballantyne, 2011; Saveman, Hallberg, & Norberg, 1993; Saveman, Hallberg, Norberg, & Eriksson, 1993; Saveman, Norberg, & Hallberg, 1992) and a few quantitative studies (Buri, Daly, Hartz, & Jogerst, 2006; Ogioni et al., 2007; Post et al., 2010; Shugarman, Fries, Wolf, & Morris, 2003) of EM using visiting nurses or home care samples have been conducted. However, these studies have collected data at only a single point in time. To our knowledge, no research including information about visiting nurse interactions with older adults over extended periods of time has examined EM prevalence and correlates. EM data collected by nurses making monthly home visits for up to 24 months for the Medicare Primary and Consumer-Directed Care Demonstration were evaluated. The objectives in this secondary analysis of
36
B. Friedman et al.
Demonstration data were to: (a) identify EM prevalence among this longitudinal cohort of older adult patients with functional disabilities, (b) determine the prevalence of EM subtypes (i.e., physical abuse, psychological abuse, sexual assault, financial exploitation, and neglect), and (c) identify correlates associated with EM.
METHODS
Downloaded by [Michigan State University] at 01:29 04 March 2015
Sample The Medicare Primary and Consumer-Directed Care Demonstration (1994–2003, with enrolled Medicare patients during 1998–2002; Friedman, Wamsley, Liebel, Saad, & Eggert, 2009) was a randomized controlled trial designed to test whether utilizing either visiting nurses, a voucher that covers goods and services not traditionally covered by Medicare, or the combination of these two options would improve health care outcomes and reduce health care costs for patients with disabilities who are at high risk of having high health care expenditures. The voucher could not be used to purchase home visits by a nurse. It was restricted to nonskilled services. Patients were recruited from the practices of 307 primary care physicians in 19 counties: 8 in New York State, 6 in West Virginia, and 5 in Ohio. A total of 1,605 subjects was randomized to 1 of 4 intervention arms: treatment as usual, the intervention including visiting nurses, the voucher, or both nurse and voucher interventions (the combination group). All Demonstration participants were required to have a high level of disability, defined as needing or receiving help for at least two ADLs or three IADLs. They also had to have received recent significant health services use, defined as having been a hospital or nursing home inpatient, or having received Medicare home health care during the previous year, or having had at least two emergency room visits during the previous 6 months. Of the 1,605 patients enrolled, 20% died and an additional 13% dropped out or were disenrolled during the 24-month period each person was eligible to participate in the Demonstration. For the present study, only the 724 subjects aged ≥ 65 years who were enrolled in the Demonstration’s nurse only and combination (nurse plus voucher) intervention arms were included, because the study’s primary interest was in the role that visiting nurses play in EM detection and management.
Data Collection Three data sources from the Demonstration were utilized for the present study. Extensive questionnaires were administered to study subjects and/or their primary informal (nonpaid) caregivers at baseline. The study forms were administered by interviewers, primarily nurses, whose only job was
Downloaded by [Michigan State University] at 01:29 04 March 2015
Elder Mistreatment Among Aged Home Care Patients
37
to administer these forms. The interviewers received extensive training; their work was carefully assessed (including the review of recorded patient/caregiver interviews); and inter-rater reliability was determined. None of the interviewers were the visiting nurses who carried out the intervention. The interviewers administered a cognitive screen to each patient. The cognitive screen assessed word repetition (book, watch, and table), 5-minute word recall, and nonresponse to basic questions about subjective health status, functional status, and life satisfaction. If the patient failed the cognitive screen (110 of 724 patients, or 15.2%), the questionnaire was administered to the caregiver. The reliability of the data collected from such caregivers was very good. The Demonstration had a very careful, elaborate process to ensure valid and reliable data collection. The interviewers read the questions to the participants and/or their primary informal caregivers and recorded the responses on paper data collection forms. Upon completion, the forms were scanned into a database. There was no information about EM at baseline since the forms contained no EM questions. The nurses, who visited patients in their homes on average once a month, completed a computerized Nurse Activity List that noted specific characteristics of the older adults’ life, such as EM presence or absence, chronic medical conditions, need for help with ADLs, and health care utilization. During and after each patient visit, the nurses documented Nurse Contact Notes, including comments on EM, patients’ physical and mental health, and their social circumstances. The nurse identified the presence of EM as well as the perpetrators during the 24-month period after each patient entered the intervention phase of the Demonstration. (The first phase consisted of planning and development, the intervention phase was the second phase, and the third phase consisted of staff phase-down and data cleaning.) For the present study, perpetrators were determined on the basis of information in the Nurse Contact Notes.
Dependent Variable—Elder Mistreatment EM was defined in the Demonstration’s Nurse Procedure Manual as any of the following: (a) physical abuse as “inappropriate physical contact such as: striking, pinching, kicking, shoving, bumping, sexual molestation”; (b) mistreatment as “inappropriate use of medication, isolation, inappropriate use of physical restraints, and inappropriate use of chemical restraints”; or (c) neglect as “failure to provide timely, safe, consistent, adequate, and appropriate care and services to participants, such as: nutrition, medication, therapies, sanitary clothing and surroundings and daily living activities.” For each visit to the patient’s home, the nurse was required to enter, in the intervention’s computerized Nurse Activity List, whether or not the patient was being mistreated. The nurses determined EM through observation; by being told about EM by patients, family members, and neighbors; and by the discovery of
Downloaded by [Michigan State University] at 01:29 04 March 2015
38
B. Friedman et al.
physical evidence on examination of the older adults. Examples are a nurse entering the home and seeing the older person left alone without food or placed in a chair in the corner and ignored. While mistreatment was based on the definition in the Nurse Procedure Manual, the Nurse Activity List did not include EM types. In the present study the dependent variable for the presence of EM (1 = yes, 0 = no) used in the quantitative analyses discussed below was constructed by first identifying whether the Nurse Activity List for each patient indicated EM, and then confirming EM presence from information in the Nurse Contact Notes. While the Nurse Procedure Manual definition included self-neglect, when the EM cases were confirmed self-neglect was not included as part of EM. In order to determine the type(s) of EM endorsed in the Nurse Activity List, two of us (EJS, AJR) carefully reviewed the Nurse Contact Notes, recording presence or absence of each of five subtypes of mistreatment: physical abuse, psychological abuse, sexual assault, financial exploitation, and neglect. No cases of self-neglect or abandonment were included as part of neglect. That is, when the subtype neglect was identified in the Nurse Contact Notes it was defined as intentional or unintentional neglect, but not self-neglect. Nurse Contact Notes were also used to characterize the relationships that were implicated, that is, who were the perpetrators. The analysis of Nurse Contact Notes’ content elucidated details (a) regarding the relationships involved in instances endorsed for EM and (b) what precisely (i.e., what actions or inaction) was seen by nurses to constitute EM in those situations.
Independent Variables—Correlates of EM The National Research Council (NRC) Panel to Review Risk and Prevalence of Elder Abuse and Neglect (National Research Council, 2003) identified three categories of EM risk factors: (a) “risk factors validated by substantial evidence” (living arrangement, social isolation, dementia, intraindividual characteristics of abusers [mental illness, hostility, alcohol abuse], and abuser dependency); (b) “possible risk factors” (gender, relationship of victim to perpetrator, personality characteristics of victims, and race); and (c) “contested risk factors” (physical impairment of the older person, victim dependence and caregiver stress, and intergenerational transmission; National Research Council, 2003). Variables measuring many but not all of these factors were available, using information collected in the baseline interview. The following domains were used in the study’s quantitative analyses: sociodemographics, family and social factors, health insurance, physical health, mental health, disability status, and Demonstration features. Sociodemographics included age, gender, race, education, annual household income, and urban or rural residence. The income question and categories were the nine categories used in Medicare’s 1998 Health of Seniors
Downloaded by [Michigan State University] at 01:29 04 March 2015
Elder Mistreatment Among Aged Home Care Patients
39
Survey (Centers for Medicare and Medicaid Services, 1998). For the present study’s analyses these were collapsed into four categories. There were four family and social factors: marital status, lives with others, reports having 0 or 1 close friends, and reports having 0 or 1 close relatives. The present study included one variable for health insurance: Medicare supplemental insurance. Since all demonstration participants were enrolled in Medicare, we did not include a variable indicating Medicare enrollment. Supplemental Medicare coverage was included because we thought that people who were without such coverage might be of a lower socioeconomic class, and hence could be more likely to suffer from EM. Supplemental insurance was not used to extend the number of Medicare home health care visits. Physical health was measured by the number of chronic conditions (constructed from a list of 13 conditions the patient reported as having been diagnosed by a physician) and the SF-36 Physical Component Summary (PCS) score (Ware, Kosinski, & Keller, 1994). The SF-36 Health Survey has been used in thousands of studies (http://www.sf-36.org/tools/SF36.shtml). Its validity and reliability have been demonstrated in older and chronically ill populations similar to the sample of the present study (Ware, Snow, Kosinski, & Gandek, 1993; Ware, Kosinski, & Keller, 1994). Mental health was assessed by four measures: the Nun Study life satisfaction question (Greiner, Snowden, & Greiner, 1996), the SF-36 Mental Component Summary (MCS) score (Ware, Kosinski, & Keller, 1994), the Cognitive Performance Scale (CPS) (Morris et al., 1994), and the MiniInternational Neuropsychiatric Interview (MINI) Major Depressive Episode (MDE) module (Sheehan et al., 1998). The Minimum Data Set CPS (Morris et al., 1994) was developed for the U.S. Department of Health and Human Services and is used for all nursing home patients whose care is paid for by Medicare or Medicaid. It consists of five items. The CPS score consists of integers ranging from 0 (intact) to 6 (very severe impairment). It has been found to be valid for a home care population similar to the sample in the present study with an R2 = 0.81 between the CPS and Mini Mental State Exam scores (Landi et al., 2000). The CPS score is calculated using a decision-tree. Because of this, it is not a domain-sampling psychometric scale of a latent construct. Standard internal consistency reliability measures, therefore, do not apply. The MINI-MDE (Sheehan et al., 1998) has been shown to be valid and reliable overall when measured against the psychiatric patient version of the Structured Clinical Interview for DSM-III-R, with a kappa of 0.55 and a sensitivity and specificity of .77 and .79, respectively, for Major Depressive Disorder (Sheehan et al., 1997). The MINI-MDE consists of nine questions. If one or both of the first two questions are answered yes, seven additional yes-no questions are asked. A major depressive episode is identified if at least five of the nine questions are answered in the affirmative.
Downloaded by [Michigan State University] at 01:29 04 March 2015
40
B. Friedman et al.
Because the MINI-MDE uses a decision-tree, it is not a domain-sampling psychometric scale of a latent construct, and standard internal consistency reliability measures do not apply. Disability status was measured by both ADLs and IADLs. There were questions for both ADL/IADL difficulty and dependence. A simple additive scale with a range of 0–6 was constructed from the six ADLs in the ADL difficulty questions of the Health of Seniors (HOS) Survey (Centers for Medicare and Medicaid Services, 1998). (While we collected data for all 86 HOS questions, we only used a small number of these questions in the present study.) Internal consistency reliability (Cronbach’s alpha) for the six-item HOS ADL difficulty scale in the present sample = 0.754. A similar additive scale was made for the IADLs from the IADL difficulty items of the HOS Survey (Centers for Medicare and Medicaid Services, 1998). Its internal consistency reliability in the present sample was 0.743. Six-item additive ADL and IADL dependence scales were constructed from the ADL and IADL dependence questions in the Outcome and Assessment Information Set B (Shaughnessy, Crisler, & Schlenker, 1997). In the present sample, the Cronbach’s alphas of those two scales were 0.761 and 0.755, respectively. Three demonstration features were included among the independent variables: study site (West Virginia/Ohio or New York), intervention group (nurse only or combination), and number of study days (length of time the patient participated in the intervention phase of the demonstration). Table 1 presents the variable name, description, and instrument source of the independent variables.
Analyses Statistical analyses consisted of descriptive statistics, bivariate analysis, and logistic regression analysis. The descriptive statistics consisted of univariate analyses and are depicted in Table 2. Bivariate analyses were used to examine the association between EM (yes or no) and the independent variables presented above. These analyses consisted of the Chi Square test for categorical variables and the t-test or Analysis of Variance (ANOVA) for continuous variables. Four logistic regression models were estimated to examine the association between EM (yes or no) as the dependent variable and the independent variables. Models 1 and 2 included variables for ADL and IADL dependence while Models 3 and 4 included variables for ADL and IADL difficulty. Difficulty and dependence were analyzed separately because they are fundamentally different constructs measured with very different questions. In Models 1 and 3 the ADL and IADL variables were numbers (counts) of ADLs and IADLs, while Models 2 and 4 included individual ADLs and IADLs, such as dressing and shopping. A value of p < .05 was used to indicate significant association. Stata version 12 (Stata Corporation, College Station, TX) was employed for all statistical calculations.
41
Annual household income < $5,000 income $5,000–$9,999
High school graduate or GED Some college or 2-year degree 4-year college graduate More than a 4-year college degree > High school graduate
Education 8th grade or less Some high school, but did not graduate < High school graduate
SOCIODEMOGRAPHICS Age (in years) Female gender Race American Indian/Alaskan Native Asian or Pacific Islander Black or African American White Another race or multiracial Non-White race
Variable
TABLE 1 Independent Variables
= = = = = Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0
= = = = = No No No No No
= = = = Yes, Yes, Yes, Yes,
0 0 0 0
= = = = No No No No
1 = Yes, 0 = No 1 = Yes, 0 = No
1 1 1 1
1 = Yes, 0 = No 1 = Yes, 0 = No
1 1 1 1 1
65–99 1 = Yes, 0 = No
When Data Were Collected
In Regression Models
(Continued)
1 = some college or 2-year degree, 4-year college graduate, or more than a 4-year college degree, 0 = other
1 = 8th grade or less, or some high school, but did not graduate, 0 = other 1 = Yes, 0 = No
1 = American Indian/Alaskan Native, Asian or Pacific Islander, Black or African American, or another race or multiracial, 0 = White
65–99 1 = Yes, 0 = No
Variable Description
Downloaded by [Michigan State University] at 01:29 04 March 2015
42
Living arrangement Living alone Living in a household with spouse only Living in a household with spouse and others Living in a household with relatives
FAMILY AND SOCIAL FACTORS Marital status Married Widowed Divorced Separated Never married Other
Urban (metropolitan statistical area)
$10,000–$19,999 $20,000–$29,999 $30,000–$39,999 $40,000–$49,999 $50,000–$79,999 $80,000–$99,999 $100,000+ $30,000+
< $10,000 income
Variable
TABLE 1 (Continued)
= = = = = = = Yes, Yes, Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0 0 0
= = = = = = = No No No No No No No
= = = = = Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0
= = = = = No No No No No
1 = Yes, 0 = No
1 = Yes, 0 = No 1 = Yes, 0 = No 1 = Yes, 0 = No
1 1 1 1 1
1 = Yes, 0 = No
1 1 1 1 1 1 1
When Data Were Collected
In Regression Models
1 = divorced, separated, or never married, 0 = other
1 = Yes, 0 = No 1 = Yes, 0 = No
1 = $30,000–$39,999, $40,000–$49,999, $50,000–$79,999, $80,000–$99,999, or $100,000+, 0 = other 1 = Yes, 0 = No
1 = < $5,000 income or $5,000–$9,999, 0 = other 1 = Yes, 0 = No 1 = Yes, 0 = No
Variable Description
Downloaded by [Michigan State University] at 01:29 04 March 2015
43
PHYSICAL HEALTH Chronic conditions Hypertension or high blood pressure Angina or coronary artery disease Congestive heart failure Myocardial infarction or heart attack Other heart conditions Stroke Emphysema, asthma, or COPD Crohn’s disease, ulcerative colitis, or inflammatory bowel disease
HEALTH INSURANCE Medicare Supplemental Insurance
Number of friends that feels close to None One friend Two to five friends Six to nine friends Ten or more friends 0 or 1 close friend Number of relatives that feels close to None One relative Two to five relatives Six to nine relatives Ten or more relatives 0 or 1 close relative
Living in a household with nonrelatives only Other Lives with others
Yes, Yes, Yes, Yes, Yes,
= = = = =
1 1 1 1 1
0 0 0 0 0
0 0 0 0 0 = = = = =
= = = = = No No No No No
No No No No No
1 1 1 1 1 1 1 1
= = = = = = = = Yes, Yes, Yes, Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0 0 0 0
= = = = = = = = No No No No No No No No
1 = Yes, 0 = No
Yes, Yes, Yes, Yes, Yes,
= = = = =
1 1 1 1 1
1 = Yes, 0 = No 1 = Yes, 0 = No
1 = Yes, 0 = No
(Continued)
1 = none or one relative, 0 = other
1 = none or one friend, 0 = other
1 = living in a household with spouse only, spouse and others, relatives, or nonrelatives only, or other, 0 = living alone
Downloaded by [Michigan State University] at 01:29 04 March 2015
44
Major depressive episode
Definite cognitive impairment
Borderline cognitive impairment
SF-36 Mental Component Summary (MCS) score Cognitive impairment
High life satisfaction
MENTAL HEALTH Life satisfaction
Arthritis of hip or knee Arthritis of hand or wrist Sciatica Diabetes, high blood sugar, or sugar Any cancer (other than skin cancer) Number of chronic conditions SF-36 Physical Component Summary (PCS) score
Variable
TABLE 1 (Continued)
= = = = = Yes, Yes, Yes, Yes, Yes,
0 0 0 0 0
= = = = = No No No No No
The 9 Mini-International Neuropsychiatric Interview (MINI) Major Depressive Episode (MDE) module questions (See Sheehan, et al., 1998)
The SF-36 questions (See Ware et al., 1993) The 5 Cognitive Performance Scale questions (See Morris et al., 1994)
1 = excellent, 2 = very good, 3 = good, 4 = fair, 5 = poor
The SF-36 questions (See Ware et al., 1993)
1 1 1 1 1
When Data Were Collected
In Regression Models
1 = Cognitive Performance Scale score of 1, 0 = other 1 = Cognitive Performance Scale score of 2–6, 0 = other 1 = Yes, 0 = No
1 = excellent, very good, or good, 0 = fair or poor 0 (worst)–100 (best)
0–13 0 (worst)–100 (best)
Variable Description
Downloaded by [Michigan State University] at 01:29 04 March 2015
45
Toileting dependence
Eating dependence
Dressing dependence
Number of ADLs with any difficulty Activities of daily living (ADL) dependence Bathing dependence
Walking difficulty
Transferring difficulty
Toileting difficulty
Eating difficulty
Dressing difficulty
DISABILITY STATUS Activities of daily living (ADL) difficulty Bathing difficulty
= has difficulty, 3 =
= has difficulty, 3 =
= has difficulty, 3 =
= has difficulty, 3 =
= has difficulty, 3 =
= has difficulty, 3 =
1 = independent, 2 = some human assistance, 3 = uses bedside commode, 4 = uses bedpan or urinal, 5 = totally dependent
1 = independent without assistive devices, 2 = independent with assistive devices, 3 = some human assistance, 4 = requires another person, 5 = totally by another person 1 = independent, 2 = clothing laid/handed out, 3 = some human sssistance, 4 = entirely dependent 1 = independent, 2 = intermittent assistance, 3 = assisted throughout meal, 4 = supplemented by feeding tube, 5 = fed entirely by tube/other
1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty 1 = unable to do, 2 has no difficulty or has
or has
or has
or has
or has
or has
(Continued)
1 = any dependence (clothing laid/ handed out, some human assistance, or entirely dependent), 0 = no dependence 1 = any dependence (intermittent assistance, assisted throughout meal, supplemented by feeding tube, or fed entirely by tube/other), 0 = no dependence 1 = any dependence (some human assistance, uses bedside commode, uses bedpan or urinal, or totally dependent), 0 = no dependence
1 = any dependence (some human assistance, requires another person, or totally by another person), 0 = no dependence
1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 1 = any difficulty (unable to do difficulty), 0 = no difficulty 0–6
Downloaded by [Michigan State University] at 01:29 04 March 2015
46
Number of IADLs with any difficulty
Telephone difficulty
Shopping difficulty
Money management difficulty
Medication management difficulty
Meal preparation difficulty
Number of ADLs with any dependence Instrumental activities of daily living (IADL) difficulty Ordinary housework difficulty
Walking dependence
Transferring dependence
Variable
TABLE 1 (Continued)
1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 1 = any difficulty (some 0 = no difficulty 0–6
1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty 1 = no difficulty, 2 = 3 = great difficulty
some difficulty,
some difficulty,
some difficulty,
some difficulty,
some difficulty,
or great difficulty),
or great difficulty),
or great difficulty),
or great difficulty),
or great difficulty),
or great difficulty),
1 = any dependence (minimal assistance/device, can bear weight during transfer, unable to bear weight during transfer, bedfast—able to turn, or bedfast—unable to turn), 0 = no dependence 1 = any dependence (device/human assistance, walks only with assistance, chairfast—wheels self, chairfast—cannot wheel self, or bedfast), 0 = no dependence 0–6
1 = independent, 2 = minimal assistance/device, 3 = can bear weight during transfer, 4 = unable to bear weight during transfer, 5 = bedfast— able to turn, 6 = bedfast—unable to turn 1 = independent, 2 = device/human assistance, 3 = walks only with assistance, 4 = chairfast—wheels self, 5 = chairfast—cannot wheel self, 6 = bedfast
some difficulty,
In Regression Models
When Data Were Collected
Variable Description
Downloaded by [Michigan State University] at 01:29 04 March 2015
47
Number of study days
DEMONSTRATION FEATURES West Virginia/Ohio site Intervention combination group
Number of IADLs with any dependence 1 = yes, 0 = no 1 = combination (nurse + voucher) group, 0 = nurse only group 0–732 days in the Demonstration intervention phase
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
Telephone dependence
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
Money management dependence 1 = independent, 2 = some help, 3 = only with help, 4 = by others only
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
Medication management dependence
Shopping dependence
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
1 = independent, 2 = some help, 3 = only with help, 4 = by others only
Meal preparation dependence
Instrumental activities of daily living (IADL) dependence Ordinary housework dependence
with
with
with
with
with
with
1 = yes, 0 = no 1 = combination (nurse + voucher) group, 0 = nurse only group 0–732 days in the Demonstration intervention phase
1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 1 = any dependence (some help, only help, or by others only), 0 = no dependence 0–6
Downloaded by [Michigan State University] at 01:29 04 March 2015
48 363 301 60 444 240 148 599
FAMILY AND SOCIAL FACTORS Marital status Widowed Married Other Lives with others Reports having 0 or 1 close friend (n = 718) Reports having 0 or 1 close relative (n = 718)
HEALTH INSURANCE Medicare Supplemental Insurance
42.3 30.2 27.5
306 219 199
82.7
50.1 41.6 8.3 61.3 33.4 20.6
32.0 34.8 19.2 14.0 28.7
(7.5) 69.9 3.3
%
79.8 506 24
232 252 139 101 208
N
Study Population (N = 724)
SOCIODEMOGRAPHICS Age—mean (SD) Female Non-White Education < High school graduate High school graduate > High school graduate Annual household income < $10,000 income $10,000–$19,999 $20,000–$29,999 $30,000+ Rural (non-MSA)
Variable
TABLE 2 Patient Baseline Characteristics (N = 724)
50
37 14 3 35 26 15
12 23 8 11 44
26 15 13
81.2 44 2
N
%
92.6
68.5 25.9 5.6 64.8 48.2 27.8
22.2 42.6 14.8 20.4 81.5
48.2 27.8 24.1
(7.9) 81.5 3.7
Mistreated (n = 54)
Downloaded by [Michigan State University] at 01:29 04 March 2015
549
326 287 57 409 214 133
220 229 131 90 472
280 204 186
79.6 462 22
N
81.9
48.7 42.8 8.5 61.0 32.2 20.0
32.8 34.2 19.6 13.4 70.4
41.8 30.4 27.8
(7.4) 69.0 3.3
%
Not Mistreated (n = 670)
.046
.584 .017 .176
.019
.085
.166
.140 .054 .868 .656
P Value
49
DISABILITY STATUS Activities of Daily Living (ADL) Number of ADLs with any difficulty—mean (SD) Any ADL difficulty Walking Transferring Bathing Dressing Toileting Eating Number of ADLs with any dependence—mean (SD) Any ADL dependence Walking Bathing Transferring Dressing Eating Toileting
MENTAL HEALTH High (ex., very good, good) life satisfaction SF-36 Mental Component Summary score—mean (SD) Borderline cognitive impairment (Cognitive Performance Scale score = 1) Definite cognitive impairment (Cognitive Performance Scale score = 2–6) Major depressive episode (n = 718)
PHYSICAL HEALTH SF-36 Physical Component Summary score—mean (SD) Number of chronic conditions—mean (SD)
26.7 15.3
193 110
80.9 63.7 59.9 50.0 30.9 19.9 (1.8) 64.1 47.2 40.6 28.6 19.5 11.3
586 461 434 362 224 144 2.3 464 342 294 207 141 82
(1.8)
41.7
302
3.0
59.4 (10.9)
(2.2)
4.5 420 47.8
(9.0)
30.6
34 24 18 11 11 5
42 37 28 27 17 12 2.1
3.0
11
21
23
39 48.6
4.9
30.0
63.0 44.4 33.3 20.4 20.4 9.3
77.8 68.5 51.8 50.0 31.5 22.2 (1.7)
(1.8)
20.4
38.9
42.6
72.2 (11.0)
(2.2)
(9.9)
Downloaded by [Michigan State University] at 01:29 04 March 2015
430 318 276 196 130 77
544 424 406 335 207 132 2.3
3.0
99
172
279
391 47.7
4.5
30.6
64.2 47.5 41.2 29.2 19.4 11.5
81.2 63.3 60.6 50.0 30.9 19.7 (1.8)
(1.8)
14.9
25.7
41.6
58.4 (10.9)
(2.2)
(9.0)
(Continued)
.858 .669 .258 .165 .863 .618
.539 .442 .207 1.000 .929 .655 .363
.883
.284
.035
.875
.046 .559
.262
.662
50
88.1 85.1 66.8 58.2 46.4 24.6 (1.8) 83.4 82.5 61.7 59.2 44.6 17.3
638 616 484 421 336 178 3.5 604 597 447 429 323 125 228 381 594
DEMONSTRATION FEATURES West Virginia/Ohio Site Intervention Combination group Number of study days—mean (SD)
31.5 52.6 (221)
(1.7)
%
3.7
N
Study Population (N = 724)
Instrumental Activities of Daily Living (IADL) Number of IADLs with any difficulty—mean (SD) Any IADL difficulty Ordinary housework Shopping Meal preparation Money management Medication management Telephone use Number of IADLs with any dependence—mean (SD) Any IADL dependence Shopping Household chores Meal preparation Money management Medication management Telephone use
Variable
TABLE 2 (Continued)
34 35 627
42 43 36 31 29 14
45 45 38 30 29 16 3.6
3.8
N
%
63.0 64.8 (190)
77.8 79.6 66.7 57.4 53.7 25.9
83.3 83.3 70.4 55.6 53.7 29.6 (2.0)
(1.9)
Mistreated (n = 54)
Downloaded by [Michigan State University] at 01:29 04 March 2015
194 346 591
562 554 411 398 294 111
593 571 446 391 307 162 3.5
3.7
N
29.0 51.6 (223)
83.9 82.7 61.3 59.4 43.9 16.6
88.5 85.2 66.6 58.4 45.8 24.2 (1.7)
(1.7)
%
Not Mistreated (n = 670)
High school graduate $10,000–$19,999 annual income $20,000–$29,999 annual income $30,000+ annual income Urban residence FAMILY AND SOCIAL FACTORS Widowed Other marital status Lives with others Reports having 0 or 1 close friend Reports having 0 or1 close relative HEALTH INSURANCE Medicare Supplemental Insurance PHYSICAL HEALTH SF-36 PCS score SF-36 PCS score squared SF-36 PCS score cubed Number of chronic conditions MENTAL HEALTH High (Ex-VG-Gd) life satisfaction SF-36 MCS score
Variable 0.94−1.05 0.84−3.90 0.25−13.94 0.42−2.25 0.34−2.00 1.38−8.58 0.65−9.26 1.18−13.30 0.61−5.05 2.08−11.63 0.77−23.31 1.15−8.07 1.25−4.84 0.81−3.94 0.98−8.72 0.25−0.63 1.01−1.04 1.00−1.00 0.98−1.28 1.05−5.40 0.80−1.20
4.92 4.25 3.04 2.46 1.78 2.93 0.40 1.03 1.00 1.12 2.38 0.98
OR 95% CI
0.99 1.81 1.86 0.97 0.83 3.44 2.46 3.97 1.76
OR
2.09 −0.18
.037 .857
