Functional disability and associated factors among older Zuni Indians

Journal of Cross-Cultural Gerontology 19: 1 12, 2004. C 2004 Kluwer Academic Publishers. Printed in the United States. 1 Functional disability and associated factors among older Zuni Indians MARGARET P. MOSS, 1 YVETTE D. ROUBIDEAUX, 2 CLEMMA JACOBSEN, 3 DEDRA BUCHWALD 3 & SPERO MANSON 4 1 University of Minnesota School of Nursing, Minneapolis, MN; 2 University of Arizona College of Public Health, Tucson, AZ; 3 University of Washington School of Medicine, Seattle, WA; 4 University of Colorado Health Sciences Center, Denver, CO. Abstract. Few studies have focused on American Indian elderly and functional disability, and none have explored potential moderating or mediating factors that may lend themselves to subsequent intervention. The purpose of this study was to describe the extent of functional disability in elders and to determine which factors were associated with a higher number of Activities of Daily Living (ADL) limitations. The study was a secondary data analysis of an existing survey of American Indian elders in one southwest tribe. Functional disability was defined as limitations in ADLs and was measured by the percent of respondents reporting specific limitations and by the mean total ADL limitations. Multiple linear regression analyses were used to determine the demographic, socioeconomic and health factors associated with ADL limitations. In the 90 elders surveyed, 40 percent of respondents reported a limitation with bathing, 31 percent with walking, and 22 percent with dressing. Only 6 percent of the elders surveyed, however, reported their health status as poor on a 5-point scale. Factors associated with more ADL limitations included poorer health status, less frequent exercise, and more elder care services used. Rates of functional disability in this tribe were higher than those found in the U.S. for all races. Further studies are needed to understand functional disability in American Indian elders and their need for long-term care services. Keywords: activities of daily living, aged, Indians, North American. Introduction Chronic health conditions are the major cause of illness, disability, and death among all races in the U.S. Among the elderly, chronic disease appears to result in functional disability that, in turn, impacts health-related quality of life (Fried 1988). Functional status can be described as the path from the onset of chronic disease to disability (Nagi 1976; WHO 1980; Pope 1991). The intermediate points include pathologic changes, functional impairments, limitations, and disability. Functional disability in this report is defined as the inability to perform one or more Activities of Daily Living (ADLs) (U.S. Census 2001).

2 MARGARET P. MOSS ET AL. Functional disability in American Indian elders has only been studied in a few of the over 500 U.S. tribes and urban groups (Chapleski 1995 & 1997; Kramer 1999). These studies have found that American Indian elderly have increased levels of diabetes and other chronic diseases, which lead to increased levels of disability (USDHHS 2001); (AOA 1996). Since each tribal community varies in terms of its culture, socioeconomic context and resources for healthcare, measuring the extent of functional disability in individual tribes can reveal important and unique information. Our study focused on the reservation-based Zuni tribe whose members are culturally and geographically distinct from even their close Puebloan neighbors (Moss 2000). Our specific aims were to measure the proportion and extent of functional disability in Zuni elders, and to examine which demographic, health-related, and cultural factors were associated with functional disability. Methods Setting This secondary data analysis of an existing survey of American Indian elders focused on elders from the Zuni tribe. The Zuni Pueblo, a Southwest pueblo tribe, is located approximately 150 miles west of Albuquerque, New Mexico. Zuni is a rural reservation with no nearby towns or services, and most Zuni people reside on the Pueblo. Zuni tribal members maintain traditional beliefs and practices in many domains, including medicine, religion, food, and family roles (Moss 2000). According to the Pueblo of Zuni Tribal Census Office, Zuni s total population was 10,028 in 1996. Of this figure, 833 people (8 percent) were aged 56 106 years, consistent with averages for American Indian groups nationally (AOA 1996). While this figure is the only available age breakdown, 55 and over is a frequent service age for elderly in American Indian programs. In American Indian culture, elder status is often not solely a function of chronological age and is conferred earlier than in the majority culture. In addition, impairments associated with aging may occur 20 years earlier than in the general population, and lower age eligibility criteria exist for some federal programs (Buchwald 2001). The Pueblo of Zuni maintains a Senior Center that offers meals-on-wheels, congregate meals, senior center activities, and an adult day care serving up to 20 elders. The Indian Health Service hospital at Zuni provides a 45-bed general medical hospital and a full range of outpatient services. Patients must travel to Gallup or Albuquerque for higher levels of service (e.g. surgery, emergencies, and intensive care). American Indian elders do not have access to assisted living or nursing home facilities on the pueblo.

DISABILITY AMONG OLDER AMERICAN INDIANS 3 Survey instrument/sample This study used data collected by the University of North Dakota National Resource Center on Native American Aging community assessment surveys completed in 2000. This ongoing study funded by the Administration on Aging provides materials and resources for tribal communities to conduct needs assessment surveys among their elder populations. Elders were recruited to participate in each community, often through elderly services and nutrition programs. Senior program staff and other trained volunteers assisted elders in filling out the survey. Results were provided to the individual tribal communities on completion of the survey analysis. Nationally, a total of 8,560 elders over 55 years completed the survey, representing 85 tribes ( National Resource Center on Native American Aging 2002). The larger aggregate dataset with all tribes surveyed was unavailable for analysis. The Zuni sample includes 128 elders, age 55 years and older, or about 15 percent of the elders reported in the Zuni tribal census. In the Zuni sample, elders were recruited largely in local elder care settings and therefore, represent a convenience sample of service users. After excluding observations with missing data, we were left with 90 subjects for this analysis. Human subjects approvals were obtained from the Zuni tribe and the University of Minnesota. The Zuni tribe reviewed the manuscript and its conclusions and approved it for publication. Measures The National Resource Center on Native American Aging instrument included items on sociodemographics, health status indicators, and functional ability. Sociodemographic variables available for analysis included age (categories 55 64, 65 74, 75 84, 85+ years), gender, living arrangement (alone or with others) and retired status. Self reported health status indicators included health status (excellent, very good, good, fair, poor), number and type of chronic conditions (arthritis, diabetes, congestive heart failure, stroke, vision problems, hearing problems, hypertension, any cancer, asthma), and number of times exercised per week. Health services utilization indicators included number of hospitalizations in the last year, number and type of elder care services used (dietary and nutritional services, occupational/vocational therapy, speech/audiology therapy, meals on wheels, transportation, respite care, personal care, skilled nursing services, physician services, social services, and physical therapy). Functional disability was assessed by the elder indicating the need for assistance in one or more of the following ADLs: 1) bathing, 2) dressing, 3) eating, 4) getting in and out of bed, 5) walking, and 6) toileting.

4 MARGARET P. MOSS ET AL. Other relevant variables in the dataset were excluded due to high numbers of missing values. Analysis We determined the demographic and socioeconomic characteristics of the Zuni sample by calculating percent and mean responses for each question. The extent of functional disability was evaluated using two measures: 1) the percent of respondents selecting each ADL limitation; and 2) the mean number of total ADL limitations. Total ADL limitations were used as the functional disability outcome in this analysis. We used multiple linear regression analyses to determine whether selected demographic, socioeconomic or health status factors were associated with ADL limitations. We tested sociodemographic factors (age, female gender, living alone vs. with family, retired status), and health status factors (poorer perceived health status, number of chronic diseases, number of times of reported exercise per week, number of hospitalizations in the past year, and number of services for elders used) for association with total ADL limitations. Factors were included in the final model if they showed an association with the outcome in preliminary univariate analyses, or if they were associated with the outcome in the literature. Due to the small sample size and the desire to minimize the risk of eliminating potentially important variables from the analysis, we chose a p = 0.10 threshold as the univariate analysis criterion for variables to be included in the final model. We considered factors with a p 0.05 to be significantly associated with total ADL limitations in the final model, but also reported factors with a p 0.10. Results are presented as odds-ratios with 95 percent confidence intervals. All analyses were performed using SPSS version 11.0. Results Demographics The 90 subjects were distributed approximately evenly among age and gender categories (Table 1). Only 8 percent lived alone and 31 percent were retired. The majority (58 percent) of the elders experienced good to excellent health, and only 6 percent had poor health. The mean number of chronic conditions was 1.8, with hypertension (56 percent), vision problems (54 percent), hearing problems (51 percent), and diabetes (42 percent) being most common. Overall,

DISABILITY AMONG OLDER AMERICAN INDIANS 5 Table 1. Demographic, Socioeconomic, and Health Status Characteristics of 90 Zuni Elders Variable N (%) Demographic: Age, years, N (%) 55 64 18 (20) 65 74 24 (27) 75 84 18 (20) 85 and over 30 (33) Female gender, N (%) 52 (58) Socioeconomic: Live alone, N (%) 7 (8) Retired, N (%) 28 (31) Health: Health status, N (%) Poor 5 (6) Fair 33 (37) Good 33 (37) Very good 15 (17) Excellent 4 (4) 1 Overnight hospital stay N (%) 28 (31) Chronic conditions, N (%) Hypertension 50 (56) Vision problem 49 (54) Hearing problem 46 (51) Diabetes 38 (42) Arthritis 20 (22) Congestive heart failure 11 (12) Stroke 10 (11) Any cancer 7 (8) Asthma 1 (1) Total chronic conditions, mean (SD) 1.8 (1) Number of times exercised/week, mean (SD) 1.4 (2) Service use Meals on Wheels 44 (49) Transportation 21 (23) Dietary and nutritional services 14 (16) Personal care 12 (13) Physical therapy 8 (9)

6 MARGARET P. MOSS ET AL. Table 1. (Continued) Variable N (%) Respite care 7 (8) Skilled nursing services 5 (6) Physician services 4 (4) Speech/Audiology 4 (4) Social services 2 (2) Occupational/Vocational therapy 0 (0) Other services 10 (11) Total services used, mean (SD) 1.5 (2) Functional disability ADLs, N (%) Bathing 36 (40) Dressing 20 (22) Eating 10 (11) Getting in and out of bed 16 (18) Walking 28 (31) Using the toilet 18 (20) Total ADL limitations, mean (SD) 1.4 (2) 31 percent of elders reported one or more hospitalizations in the past year and the mean number of elder care services used was 1.5, with meals on wheels (49 percent) and transportation (23 percent) being most common. The mean number of times respondents exercised per week was 1.4. Functional disability The extent of functional disability in this sample of elders, as measured by limitations in ADLs, was high (Table 1). The mean number of total ADL limitations was 1.4. At least 11 percent of respondents reported limitation in each ADL category, with many respondents reporting limitations in bathing (40 percent) and walking (31 percent). Factors associated with higher ADL limitations in preliminary univariate analyses (results not shown) included higher age, poorer health status, one or more overnight hospital stays in the past year, fewer times exercised per week, higher total number of chronic conditions, and higher total number of services used. Gender and living alone did not show a univariate association with total number of ADLs, but have been associated with functional disability in other studies and were included in the final model.

DISABILITY AMONG OLDER AMERICAN INDIANS 7 Table 2. Adjusted Linear Regression on Total Activities of Daily Living Limitations for Selected Demographic, Socioeconomic, and Health Status Factors in Zuni Elders Factor Change a (CI) b p Age category 0.2 ( 0.2, 0.5) 0.35 Female gender 0.6 ( 1.3, 0.0) 0.06 1 hospital stays in year 0.6 ( 0.1, 1.4) 0.08 Number of times exercised/week 0.3 ( 0.6, 0.1) <0.01 Total number of services used 0.3 (0.1, 0.5) <0.01 Living alone 0.6 (.1.8, 0.7) 0.38 Poorer health c 0.4 (0.0, 0.8) 0.04 Total number of chronic conditions 0.2 ( 0.1, 0.4) 0.20 a Change represents a one-unit the estimated increase or decrease in total limitations associated with a one-unit increase the factor of interest. Binary variables are presented as associated category vs. reference category. b CI = 95% confidence interval. c Higher values correspond to poorer health on a 5 point scale (5 = poor, 1 = excellent). Table 2 shows the results from the final adjusted linear regression model. The relative change in the number of total ADL limitations was estimated for a one-unit increase in each factor of interest. Variables associated with higher total ADL limitations at the p 0.05 significance level included fewer number of times exercised per week, use of more services, and poorer health. Variables associated with higher total ADL limitations at the p 0.10 significance level also included male gender and one or more overnight hospital stays in the past year. Age, living alone, and total number of chronic conditions were not associated with total ADL limitations in the adjusted linear regression analysis. Discussion In this secondary analysis of a community based needs assessment of Zuni elders, the extent of functional disability was found to be high, with specific limitations ranging from 11 40 percent. In NHANES III, non-hispanic black and Mexican-American men and women aged 60 and older reported more disability than non-hispanic whites in all categories, and the percent of non- Hispanic whites reporting difficulty in each category was lower than the levels reported in this study (Ostchega 2000).

8 MARGARET P. MOSS ET AL. The high levels of ADL limitations found in this study were consistent with the results of other studies in American Indian communities. Chapleski (1995) reviewed service use in 206 American Indian elders in Michigan in rural (both on and off reservation) and urban sites. Functional disability was analyzed for association with increased or decreased eldercare service use. Scores for functional status ranged from one to 15 (additive of ADLs and Instrumental Activities of Daily Living) and resulted in a mean of 1.8 limitations per elder. A study of 294 urban elders in California reported percentages of those reporting any impairment in ADLs, with toileting (4.9 percent) being the least frequent and mobility (13.1 percent) the most frequent (Kramer 1999). Although limited in its report on ADLs, a study of Yaqui elderly in Arizona (Paz 1998) found one-quarter of the sample was retired on disability and 61 percent experienced severely limited personal transportation. The elders in this study had a high level of chronic medical conditions, including many that are often associated with disability (diabetes 46 percent, vision/hearing problems 54 51 percent, arthritis 22 percent), consistent with other studies of American Indian elders. One study of 21 chronic conditions among Great Lakes American Indians aged 55 years reported that the mean number of conditions per elder was 3.9; 25 percent had 1 or 2 conditions, over 30 percent had 3 4 conditions, and approximately 40 percent reported 5 chronic conditions with only 7.1 percent reporting no conditions. At least one ADL limitations was present in 22 percent of their sample. Moreover, their measure of the severity of comorbidity was a significant predictor of functional abilities (Chapleski 1997). Despite the extent of ADL limitations and chronic disease in this Zuni sample, the mean use of elder care services seemed low (1.5), with meals on wheels and transportation used most commonly. Increasing numbers of ADL limitations has been shown to result in a greater need for long term care, with more than two ADL limitations seen as severely disabled and requiring a higher level of care (McBride 1989). However, long-term care services are limited on the Zuni Pueblo, and elders may not have been able to access needed services. Additionally, those requiring a higher level of care were likely not living on the reservation, and therefore unlikely to have participated in this survey. On the Zuni Pueblo, community practices and the environment may play a role in the higher reported ADL limitations in these elders. Elders may require assistance more commonly than elders who do not live in isolated rural conditions due to the more active and strenuous level of ADLS. Even though the elders in this study reported a low level of exercise, their lifestyles in this rural setting require a great deal of daily activity which may lead to functional challenges manifesting themselves at an earlier stage. For instance, heavy

DISABILITY AMONG OLDER AMERICAN INDIANS 9 housework may include chopping wood since most homes have outside kiva ovens that are widely used, especially by the women, and/or use wood to heat and cook inside. Frequently, improvements such as grab bars for bathrooms may be difficult to install due to the ancient adobe walls that will not accommodate such installations. Additionally, elderly may use outhouses. These factors may play a role in the high rate of toileting problems. Walking is often done on uneven dirt roads, and shopping can entail traveling an 80 mile round trip to the nearest large town. However, this sample represents a convenience sample of service users, so their higher reported ADL limitations may also be due to selection bias. Before any programmatic changes are implemented based on the results, these strikingly high levels of ADL limitations and functional disability should be verified by a more rigorous scientific survey that includes measures that are more culturally relevant to this population, such as a more accurate assessment of daily activity levels and regular exercise. Our findings also highlight the need to measure the functional status of elders with measures such as the ADL, since measuring just whether they exercise may not provide enough information on their true level of functioning throughout the day. The finding that 58 percent of these elders reported good to excellent health is surprising given the high level of ADL limitations and chronic disease, but these findings are self-reported and are dependent on the frame of reference from which the elder reports their health, which may be different in this unique community compared to the U.S. general population or even to other American Indian communities. These results are consistent with a previous study of American Indian elders that found that 51.2 percent of elders reported good or excellent health (Chapleski 1995). The association of poorer health status with greater ADL limitations is consistent with other studies (Kennedy 2001). In this study, factors associated with greater functional disability included poorer health status, less frequent exercise and more elder care services used. Previous studies have shown that ADL limitation increases with age, female gender, being unmarried, living in certain regions of the country, and the number of chronic health conditions (McBride 1989; Weiner 1989; Lueckenotte 2000). In our study, the number of chronic health conditions was not found to be associated with ADL limitations in the final linear regression model. However, the factors with significant association (less exercise, more services used, and poorer health) all relate to an increased medical burden, which has been shown to be a better predictor of ADL limitations than the absolute number of chronic health conditions (Chapleski 1997). Our results apply only to elders in the Zuni Pueblo, and are not generalizable to the entire American Indian population. Recruitment for this study

10 MARGARET P. MOSS ET AL. often occurred in venues where elders participated in established services, which may have biased the sample towards those elders who have greater degrees of functional disability and co-morbid conditions and thus require more services, although reported use of services was low in this analysis. Also, this community may be more concerned about their elder services and may not represent all Indian communities. In the Zuni community, the association of male gender with greater ADL limitations may be due to a lower utilization of health care services by men, which was not measured in this survey. While this study relies on self-report of functional disability and does not include provider assessments, previous studies have shown self-report to be more accurate than physician report (Calkins 1991). Although certain limitations exist in using functional status measures, such as variability in assessment, terminologies, and degree of loss or limitation, no better determination of frailty and disability in the geriatric patient is available (Maxted 1999). Precise assessment of an individual s functional health status can assist in their appropriate placement in either the home and communitybased or institutional-based long-term care systems if needed. A foremost contributor to the need for long-term care is functional decline resulting from physical frailty and impairment (Lueckenotte 2000). Regarding assessment for long-term care, proper placement, and continued planning, physical functioning is conceivably the most essential measure required in long-term care (Kane 1981). An increasing number of ADL limitations results in a greater need for long term care, with more than two ADL limitations seen as severely disabled and requiring a higher level of care (McBride 1989). In this study, the mean number of ADL losses (1.4) defines this group of elders as fitting the definition of severe disability (U.S. Census, 2001). The results of this community-specific study can inform substantive as well as programmatic issues of considerable importance. Measurement of elderly functional health status in a distinct community may contribute to understanding its long-term care system needs. Understanding not only characteristics of the individual but also the support system and community resources such as formal and informal systems, presence of long-term care institutions, and the characteristics of those institutions, are all factors affecting the need for nursing home admission (Ouslander 1991). The older person with functional disability will eventually require an accessible, appropriate placement along the long-term care continuum. Given the variety of options for long term care of elders with functional disability and the increasing role of tribes in managing their own healthcare services, assessment such as this study can help a tribal community better understand the needs of its elders with disability and better provide appropriate services for both home-based and institution-based care as needed.

DISABILITY AMONG OLDER AMERICAN INDIANS 11 The goal of community-based estimates of functional assessment can be seen as: 1) estimating care requirements, 2) planning placement, and 3) choosing types of specific care. While these three goals were conceived for individuals (Feinstein 1986), their use can extend toward community planning for policy and programs. Further studies are needed to assess the extent of functional disability in other American Indian communities and the extent of use and access to needed long-term care services. Given the significant underfunding of health care in the Indian Health Service, and the lack of long term care funding for tribes, more data is needed to determine needs and to justify increased resources. Acknowledgements The authors wish to thank the Zuni tribe for its approval of and assistance with this study, and to the National Resource Center on Native American Aging for their initial survey. References AOA (1996). Home and community-based long-term care for native Americans: Final report. Washington, DC: US DHHS. Buchwald, D., Furman, R., Ashton, S. & Manson, S. (2001). Preventive care of older urban American Indians and Alaska Natives in primary care, Journal of Internal Medicine 16: 257 261. Calkins, D.R., Rubenstein, L.V., Cleary, P.D., Davies, A.R., Jette, A.M., Fink, A., Kosecoff, J., Young, R.T., Brook, R.H. & Delbanco, T.L. (1991). Failure of physicians to recognizefunctional disability in ambulatory patients, Annals of Internal Medicine 114: 451 454. Chapleski, E.E. & Dwyer, J.W. (1995). The effects of on- and off-reservation residence on in-home service use among Great Lakes American Indians, Journal of Rural Health 11: 204 216. Chapleski, E.E., Lichtenberg, P.A., Dwyer, J.W. & Tsai, P.F. (1997). Morbidity and comorbidity among Great Lakes American Indians: Predictors of functional ability, Gerontologist 37: 588 597. Feinstein, A.R. (1986). Scientific and clinical problems in indexes of functional disability, Annals of Internal Medicine 105: 413 420. Fried, L.P. (1988). Morbidity as a focus of preventative health care in the elderly, Epidemiology Review 10: 48 64. Kane, R.A. & Kane, R.L. (1981). Measuring the elderly. Lexington, MA: Lexington Books. Kennedy, J. (2001). Unmet and undermet need for activities of daily living and instrumental activities of daily living assistance among adults with disability estimates from the 1994 and 1995 disability follow-back surveys, Medical Care 39: 1305 1312. Kramer, J.B. (1999). The health status of urban Indian elders. The IHS Primary Care Provider 24: 69 73. Lueckenotte, A.G. (2000). Gerontologic nursing. St. Louis: Mosby.

12 MARGARET P. MOSS ET AL. Maxted, G. (1999). Functional assessment in the elderly, IHS Primary Care Provider 23: 149 152. McBride, T.D. (1989). Measuring the disability of the elderly: Empirical analysis and projections into the 21st century. Washington, DC: The Urban Institute. Moss, M.P. (2000). Zuni elders: Ethnography of American Indian Aging. Doctoral Dissertaation, University of Texas-Houston, HSC. Nagi, S.Z. (1976). An epidemiology of disability among adults in the United States, Milbank Memorial Fund Quarterly 6: 493 508. NRCNAA (2002). Research overview. http://www.und.edu/dept/nrcnaa/research.htm, accessed 2003. Ostchega, Y., Harris, R., Hirsch, R., Parson, V.L. & Kington, R. (2000). The prevalence of functional limitations and disability in older persons in the US: Data from the National Health and Nutrition Examination Survey III, JAGS 48: 1132 1135. Ouslander, J., Osterweil, D. & Morley, J. (1991). Medical care in the nursing home. New York: McGraw-Hill. Paz, J. & Aleman, S. (1998). The Yaqui elderly of old Pasqua. In M. Delgado (ed.), Latino elders and the twenty-first century: Issues and challenges for culturally competent research and practice (pp. 47 59). Binghamton, NY: Haworth Press. Pope, A.M. & Tarlov, A.R. (1991). A model for disability and disability prevention. In IOM (ed.), Disability in America (pp. 76 108). Washington, DC: The National Academy Press. US Census (2001). Census brief. http://www.census.gov/prod/3/97pubs/cenbr975.pdf, accessed, 2003. USDHHS (2001). Trends in Indian health 1998-1999. Rockville: Indian Health Service, Division of Program Statistics. Weiner, J.M. & Hanley, R.J. (1989). Measuring the activities of daily living among the elderly: A guide to national surveys. The International Forum on Aging-Related Statistics and the Brookings institution. Washington, DC. WHO (1980). International classification of impairments, disabilities, and handicaps. Geneva: World Health Organization. Address for correspondence: Margaret P. Moss, University of Minnesota, School of Nursing, WDH 6-101, 308 Harvard St. SE, Minneapolis, MN 55455, USA Phone: 612-359-8545; Fax: 612-626-2359; E-mail: mossx015@umn.edu