Differences in predictors of self-rated health among people with and without a disability.
Self-rated health (SRH) is widely recognized as a significant measure of the overall global health of an individual and has been researched for over 40 years. One of the more consistent findings in SRH research is its predictive ability of mortality and survival (Benyamini & Idler, 1999; Idler & Benyamini, 1997). Even after taking into consideration objective measures of health status and health condition, SRH still emerges as an additional independent predictor, thereby highlighting the significance of the subjective view of one's health. Much of the research on SRH has focused predominantly on older adults and in identifying the numerous predictors of SRH. However, there is a lack of studies that include special groups and different ages, and this has been identified as a limitation in the literature and the represents the next stage of research in this field (Idler & Benyamini, 1997).
One group not adequately studied in regards to SRH is people of all ages with an early to mid-life physical disability onset such as cerebral palsy (CP), polio, or spinal cord injury (SCI). We cannot assume that the existence of a long term physical disability automatically leads to a lower SRH rating. Despite functional limitations resulting from a disability, people with physical impairments may vary widely in how they view their health, with many even reporting good to excellent health. Data from the Health and Retirement study (Finnegan, Marion, & Cox, 2005) documented that 34% of people with chronic conditions reported their SRH as either very good or excellent. Also, the issues of health and wellness for people with a disability have become a significant focus in the national health agenda (U.S. Department of Health and Human Services, 2000) and thus knowledge about how people with a disability perceive their health is also relevant (Putnam, Geenen, & Powers, 2003).
For older adults without a disability, studies on SRH have found that functional impairment and frailty were often associated with lower SRH (Han, Small, & Haley, 2001; Idler & Benyamini, 1997; Idler, Hudson, & Leventhal, 1999). However, the impact of physical impairment and any resulting functional impairment on SRH may operate differently for people with a permanent disability. Disability is often a consequence of permanent physical impairment and people with such impairments learn to adapt over time. Functional impairment that results from a disability becomes a permanent part of the experience of people with a disability and is not considered evidence of poor health. These individuals may separate their physical condition from their general health status and not consider their impairment when forming their subjective health perceptions. They may form a health perception that takes other variables into consideration, such as the debilitating consequences of severe pain or fatigue, or psychosocial factors like emotional mood and the ability to socialize and participate in the community.
Only a few studies on SRH have specifically identified or included people with disabilities. Hoeymans, Feskens, Kromhout, and Van den Bos (1999) looked at the association between poor SRH and seven different chronic conditions such as stroke, diabetes, and musculoskeletal conditions in elderly men in Amsterdam. Stroke had the largest impact on poor SRH followed by respiratory symptoms, heart disease, musculoskeletal complaints, and diabetes. Finnegan et al. (2005) looked at profiles of SRH for midlife adults ages 50 to 60 years with chronic illnesses. Data were taken from the Health and Retirement Study, and disability was defined as having one or more chronic health conditions such as arthritis, hypertension, heart disease, or diabetes. Some of the variables associated with different levels of SRH were work limitations, depressed mood, smoking, being overweight, lack of vigorous activity, and having other co-morbid health problems.
The impact of psychological variables, especially depression, on SRH was examined by Han and colleagues (Han, 2002; Han et al., 2001). For a sample of people without disabilities, Han (2002) found that a high degree of depression at baseline was predictive of decline in SRH two years later; and for a sample of older adults with stroke, Han et al. (2001) found that depression explained the SRH variance by an additional 21% after accounting for functional status and number of physical illnesses.
Only one study has compared SRH predictors between people with or without a disability, that of Cott, Gignac, and Badley (1999). Cott et al. (1999) reported results from a national health survey of Canadians 20 years and older. Disability was defined as having a long-term chronic condition such as arthritis, back disorders, heart disease or respiratory disorders. Cott et al. (1999) found that people with disabilities generated a stronger predictive model for SRH than people without disabilities. For persons with disabilities, strong predictor variables for SRH were pain severity, presence of a short-term health condition within the last two weeks, having a lower education, and being unemployed. In contrast, the predictive model for those without a disability was age and the presence of a short-term health condition within the last two weeks.
The purpose of our study was twofold: (a) to investigate what factors are associated with SRH for people with disabilities and (b) to compare the predictor patterns for SRH between people with and without disabilities. According to Finnegan et al. (2005) knowledge of factors associated with SRH among people with a disability can be beneficial, especially if any are amendable to change that can be used to improve SRH. Finnegan et al. (2005) found that midlife adults with chronic conditions who reported positive baseline SRH actually reduced their odds of lower SRH levels in the future by 15% over an 8-year span. In contrast, people with average and poor SRH at baseline increased their odds of poorer future SRH levels by 2% and 4% respectively for each year. This study included adults of all ages, not just older adults, because people with disabilities are likely to experience new health problems at an earlier age than expected which in turn may influence their SRH.
Predictor variables selected for this study were based on several criteria. First we drew from the literature on SRH dimensions and structure to identify a set of predictors. Segovia, Bartlett & Edwards (1989) empirically identified a set of five dimensions of SRH, those of health and disease, emotions, subjective appraisals such as energy or fatigue, and the impact of health on activity restriction and social contacts. The three-factor structure of SRH by Han et al. (2001) identified three broad classes of predictors which appear to combine some of those by Segovia et al., e.g., function, physical disease, and depression. Second we selected variables within those categories that represent some of the more common predictors in other studies as well as variables thought to be relevant to people with a disability. The functional limitation variable selected was functional impairment as measured by instrumental activities of daily living (IADLs) and activities of daily living (ADLs); the physical disease and health condition variables selected were the number of current chronic health conditions, pain, and fatigue; and the psychosocial variables selected were depression and community involvement.
Participants included 241 people with long-term, physical disabilities and 83 people without disabilities. The participants ranged in age from 30 to 93 years, with an average age of 62 (SD = 13.3), with 66% women, 80% non-Hispanic white, and an educational level of 14.8 years (SD = 2.9). People with disabilities included six impairment groups of polio (n = 106), SCI (n = 47), CP (n = 33), rheumatoid arthritis (RA) (n = 12), stroke (n = 12) and other musculoskeletal and neurological disorders (n = 31). Average age at onset was 16.3 (SD = 16.9, range = 0 to 72 years), and average duration since onset was 45.3 years (SD = 19.6, range = 4 to 66 years). Table 1 lists the demographics for the two groups. There were no differences in age, gender, ethnicity or education between those with or without a disability.
The data presented here were part of information collected during a larger interview used in a non-intervention research study on adults aging with a physical disability. This study took place between 2004 and 2006 and was located at Rancho Los Amigos National Rehabilitation Center (RLANRC) in Southern California. RLANRC is a Los Angeles County facility that supports research activities and has an on-site federally approved Institutional Review Board (IRB). The sample with a disability was recruited through the community from local support groups, Independent Living Centers, Abilities Exposition Conferences held in Southern California, and prior RLANRC outpatients. Data were also collected from people with cerebral palsy living in Northern California through the United Cerebral Palsy Foundation in Oakland. The inclusion criteria for people with a disability were having one of six impairments of polio, SCI, CP, RA, stroke or other musculoskeletal and neurological disorders, being at least 5 years post onset, 21 years old or older, and not having a cognitive impairment. Everyone who met the inclusion criteria and was interested in the study could participate. The comparison sample was recruited from family, friends and neighbors of the sample with a disability and from rehabilitation paraprofessional and professional staff. Interested participants were contacted by phone, received an in-depth explanation of the purpose of the study and invited to participate. Final participation involved a face-to-face interview and clinical assessment with each participant. Health data were collected through interviews with a physical therapist and rehabilitation specialist, and psychological data were collected by a clinical psychologist. Transportation was provided for anyone who needed it, thereby ensuring that no participant would be eliminated from the study because of transportation difficulties. The study was reviewed and approved by the RLANRC IRB and each participant signed a con sent form. Data from participants were included in this study if there was no missing data for any of the predictor variables. Out of 331 potential participants, 324 cases had complete data and were used for the current study.
Self-rated health. This is a global one-item measure where participants rated their overall physical health on a 5-point scale as either excellent, very good, good, fair, or poor (excellent = 1 and poor = 5).
Number of current chronic health conditions. The existence of current chronic health conditions was measured by the total number of health conditions a participant had been diagnosed with from a list of 8 common health problems derived from the National Health Interview Survey (U.S. Department of Health and Human Services, 2006). The health conditions used in this study were asthma, diabetes, gastro-intestinal disorders such as GERD or stomach ulcers, hypertension and heart disease, osteo-arthritis, osteoporosis, scoliosis, and thyroid disorder. These conditions represent common types of chronic health problems reported in the general population (U.S. Department of Health and Human Services, 2006). Stroke was not included in the list because it was already addressed as one of the impairments required to participate in this study. These health conditions can be viewed as co-morbidities to the primary impairments in this study. With the exception of osteoporosis, they are not considered a direct consequence of the impairment, although some of the health conditions may exist at higher frequencies for people with a disability, such as diabetes and gastro-intestinal disorders (Campbell, Sheets, & Strong, 1999). For people with SCI, osteoporosis may be a direct consequence of the impairment, as all people with SCI sustain some bone loss, however not all bone loss reaches the criteria for being osteoporotic. Respondents were asked whether they had been diagnosed with or received treatment for any of the chronic health conditions in the 12 months prior to the interview. Conditions were counted only if they were actively treated within the last year. Four of the health conditions were considered permanent, those of diabetes, scoliosis, osteoporosis and osteo-arthritis, and once diagnosed, required lifetime treatment or management. These conditions were counted if the participant had ever been diagnosed with them.
Pain severity. Participants rated their current physical pain on a 4-point pain severity scale of 0 = none, 1 = mild, 2 = moderate and 3 = severe.
Fatigue. An assessment of overall global fatigue was measured by using the Fatigue Severity subscale from the Fatigue Assessment Instrument (Schwartz, Jandorf, & Krupp, 1993). The subscale has 11 items that describe components of fatigue and are scored on a 7-point scale, ranging from 1 = completely disagree to 7 = completely agree. The final score is an average of the items. The subscale has strong internal reliability (alpha = .92). Discriminate validity is also strong, with 81% of neurology and rheumatology outpatients scoring 4 or above, and 89% of healthy controls scoring below 4 (Schwartz et al., 1993). Test re-test reliability was moderate (.69) and may be because fatigue varies over time. However, the high discriminate validity shows that it is a useful measure of discriminating high levels of fatigue experienced by people with impairment from lower levels typically experienced by people with no impairment.
Function. Function was assessed using the self-report ADL and IADL measures from the Older Americans Resources and Services Program (OARS) (Fillenbaum, 1988). The items measured were seven ADLs of eating, grooming, dressing, mobility, bed transfers, showering, and toilet care and seven IADLs of telephone use, transportation, shopping, meal preparation, housework, medication management, and finances. Final scores were the total number of items the participant could do independently without help from either equipment or a person. The scale has satisfactory reliability and validity data (Fillenbaum & Smyer, 1981).
Depression. The Older Adult Health and Mood Questionnaire (OAHMQ) (Kemp & Adams, 1995) was used to measure depression. It is a 22-item, clinically-validated screening instrument that was designed around the DSM III-R criteria for depressive symptoms and developed to assess the presence of depression in adults who have co-existing health problems. Items on the scale cover the major depression components of dysphoric mood, and behavioral, cognitive and physiologic symptoms, and are scored as either true or false (true = 1, false = 0). Final total scores range from 0 to 22. The scores can be used to assign the person to one of three diagnostic groups of not depressed (scores of 0 to 5), significant depressive symptoms (scores of 6 to 10) and probable major depression (scores of 11 to 22). It has been validated on 171 people with SCI (Krause, Kemp, & Coker, 2000), with the scale demonstrating high reliability (retest = .84 and internal reliability = .87), sensitivity (.92), and specificity (.87).
Community activities. The Community Activities Checklist (Kemp & Ettelson, 2001) was designed to measure the number of activities a person engages in during a week in the areas of social, interpersonal, leisure, romantic, pleasurable, and group activities. This measure documents the degree of social interaction and social involvement of an individual. There are 15 items that ask the person to state how many times he or she engaged in that activity during the last seven days. Examples of some items are: "visited in person with a friend or friends," "got out of the house," "went shopping for other than food," "went to a show, theatre, concert or sporting event." Data from a normative study on a preliminary 16-item version indicated that the expected range in persons with a disability was from 3 to 84, with a mean of 34.8. Internal reliability is good, with alpha = .75 (Kemp & Ettelson, 2001).
All data were analyzed using SPSS version 11.5 (SPSS, Inc., 2002). Descriptive statistics were calculated for demographics, SRH, and all predictor variables. Preliminary bivariate correlations were done between the dependent variable of SRH and the health, function, and psychological measures to examine their initial relation to SRH. Stepwise multiple regression analyses were performed separately for the sample with a disability and the non-disability comparison group, with SRH as the dependent variable and the predictor variables of number of chronic health conditions, pain severity, fatigue, IADLs, ADLs, depression, and community activities.
Table 1 presents the t-test and chi-square values for the two samples of those with and without a disability for demographics, SRH, and predictor variables. People with a disability scored significantly lower in SRH and all predictor variables compared to people without a disability. For the disability sample, 36% rated their SRH as excellent or very good, 37% as good, 18% as fair, and 9% as poor. In contrast, 69% of the comparison group rated their health as excellent or very good, 17% as good, 12% as fair, and 2% as poor.
For the psychosocial variables the disability sample in general scored higher in depression and lower in community activities. The percentage of participants who scored within the "no depression" range (score between 0 to 5) was 59% for the disability sample and 77% for the non-disability comparison sample. Participants who scored within the range of major depression (score of 11 or higher) were 15% for the disability group and 6% for the non-disability group.
For the health measures the disability group reported slightly more chronic health conditions (disability group M = 2.2, non-disability group M = 1.6), more pain (disability group M = 1.9; non-disability group M = 1.1) and more fatigue (disability group M = 3.3; non-disability group M = 0.9). Those who reported severe pain were 32% of people with a disability and 15% of people without a disability. For the measures of function, people with a disability had significantly lower function (IADLs M = 2.9 and ADLs M = 3.5) than people without a disability (IADLs M = 6.1, ADLs M = 6.4).
Table 2 reports bivariate correlations between SRH and the predictor variables. At the univariate level, SRH was significantly associated with all the predictors for both groups, with the exception of ADLs for the disability group. The variables with the highest significant correlations with SRH for the disability group were depression (r = .43), community activities (r = -.36), pain severity (r = .31) and fatigue (r = .28). For the non-disability group, variables with the highest SRH correlations were pain severity (r = .53), IADLs (r = -.48), depression (r = .47) and number of chronic health conditions (r = .39).
Separate stepwise multiple regression analyses were performed for each of the groups to identify those variables that had the strongest association with SRH and if the predictor patterns differed between those with and without a disability (see Table 3). For people with a disability, five variables were significantly related to SRH: the two psychosocial variables of depression and community activities, severity of pain, number of chronic health conditions, and fatigue. Functional status as measured by IADLs and ADLs did not enter the equation. Altogether, 30% of the variability in SRH was explained by these measures. For people without a disability the significant predictors were severity of pain, the functional measure of IADLs, and depression, with 41% of the variability of SRH explained.
The purpose of this study was to explore the predictors of self-rated health for adults of all ages experiencing a permanent long-term disability and compare it to people without a disability. Predictors selected were number of current chronic health conditions, functional impairment, pain, fatigue, and psychosocial variables of depression and community involvement. Since functional impairment is very common among people with a disability and a result of the physical impairment, it was hypothesized that it would not influence SRH to the degree that it would for people without a disability. For those with a disability, functional impairment as measured by dependence in ADLs or IADLs represents the severity of the disability and not necessarily illness or poor health, whereas for people with no disability, functional impairment may represent negative health changes, health decline, and frailty.
The major SRH predictors for people with a disability were number of chronic health conditions, pain, and fatigue, and psychosocial factors of depression and community activities. These predictors were very similar to those reported in a qualitative study of 19 focus groups that discussed health barriers among people with a disability (Putnam, Geenen, & Powers, 2003). In contrast the SRH predictors for comparison group without a disability were functional impairment, pain, and depression. For the group without a disability, we expected chronic health conditions to also be a predictor because it measures persistent health problems. However its non-significance in the regression analysis may be due to its high correlation with IADLs (r = .52) and pain (r = .40).
There were two major differences in SRH determinants between the two groups. First, as hypothesized, the functional impairment as measured by ADLs and IADLS did not influence SRH for the disability group, although it was a strong predictor of SRH for those without a disability. Second, community and social activities was a major predictor primarily for people with a disability. For people with a disability, participation in community activities may represent a distinct aspect of subjective health. According to McKinley and Meade (2004), community integration for people with a disability represents their ability to overcome impairment barriers and fully participate in all aspects of life, and has often been considered one of the major goals of successful rehabilitation. As people with disabilities reduce their number of social activities or completely give up an activity, this may suggest to them that they are entering a phase of new health problems and health decline. Also, any decisions to give up or reduce activities may be due to worsening health conditions. For example, in a study on people with SCI, Charlifue and Gerhart (2004) found that a decline in community integration was also associated with decline in mobility, ability to work, and life satisfaction. The focus group study by Putnam, Geenen, and Powers (2003) found that social engagement was identified as one significant indicator of degree of health. It could be that for people with a disability a reduction in community activities rather than functional decline is viewed as representative of worsening health now and in the immediate future.
Chronic health conditions. One finding from our study was that people with a disability reported much lower SRH than did people without a disability. Those with a disability also reported higher levels of pain, fatigue, and more chronic health conditions compared to the non-disability group, thereby having more health problems that would lead to reduced SRH. Our results of more numerous health problems for people with a disability support other research findings about the emergence of new health problems and premature aging found among middle age adults with a disability (Campbell, Sheets & Strong, 1999; Kemp, 2005; Mosqueda, 2004). Also, since this study only collected data on common chronic health conditions, it is possible that people with a disability had other secondary health conditions not identified but directly related to their physical impairment (e.g., pressure sores, urinary track infections) that compromised their SRH. Lastly, it is possible that the addition of each new health problem may have a stronger health impact on people already managing a disability which would lead to lower SRH. Given these findings and possibilities, practitioners seeing people with disabilities may want to place a strong focus on identifying and addressing the prevention and early detection of chronic health conditions because of the likelihood of increased health problems and their more severe impact on reduced SRH.
Depression. Several research studies have shown that depression is strongly and consistently associated with subjective health impressions (Han, 2002; Han & Jylha, 2006; Mulsant, Ganguli, & Seaberg, 1997). Our findings are similar to earlier results, and demonstrate that for people both with and without a disability, depression remains a strong predictor of SRH. One explanation for the relation between SRH and depression is that the somatic symptoms of depression such as low energy and vitality, listlessness, and lack of sleep are used to infer poor health (Mulsant et al., 1997). People simply feel worse when depressed. Also, experiences of low vitality can make it difficult for people to seek medical help or take care of their health. Another explanation by Kemp (2005) who studied people with disabilities is that serious health changes can lead to episodes of depression. New health problems may cause people to worry more about their health and become discouraged about maintaining good health. Longitudinal studies have shown that depression can significantly change SRH over time, for better or worse, with greater depression predictive of later SRH decline, and abatement of depression leading to similar improvements in SRH (Han, 2002; Hart & Jylha, 2006). Given the strong influence of depression on SRH, it would behoove people with a disability and their health care providers to pay particular attention to any signs of depression, as depression can be both an antecedent and consequence of new health problems and health decline.
Practical implications. Health professionals can benefit from knowing how people with disabilities subjectively view their health, and what they consider when deciding if health is good or poor. It is possible that health professionals may erroneously rate the health of people with disabilities lower than is self-perceived, especially if they are not familiar with disabilities. Since SRH is a good predictor of overall health issues for those with disabilities, it would be useful in health examinations, evaluation and future health treatment. Given our findings, health improvement strategies for people with a disability may benefit from the following: (a) diagnosing and treating depression early, especially for practitioners who are not experienced with people with disability and who may overlook depressive symptoms; (b) identifying and addressing specific health conditions that may influence a person's reduction in community activities; and (c) developing strategies for managing pain and fatigue.
Limitations. There are several limitations to this study. First, we do not know if the results of our study are generalizable to people with other physical impairments not included in our sample, for example, people with traumatic brain injury. Our sample was also relatively well educated, with an average of two years of college. Educational level has been associated with better overall health and positive health practices. People with both lower educational levels and a compromised physical system due to disability may have more serious or complicated health problems and needs, leading to different SRH predictors and results than we found. Second, the choice of current chronic health conditions in this study is not exhaustive. The inclusion of more or different health conditions may have produced somewhat different results. This study could have benefited by including more secondary health conditions that directly result from the included impairments, such as urinary track infections or pressure sores. However, since these conditions would differ according to the impairment and would limit comparisons to other studies on this topic, we used measures of pain and fatigue which are common across many physical disabilities (Kemp & Mosqueda, 2004). A third limitation to this study was the need to account for the timing of adjustment to new health conditions. While this study measured current health conditions active within the last year, it is possible that the impact of health conditions on SRH may vary based on level of adjustment. For example, someone recently diagnosed with diabetes within the last 3 months may experience this condition more severely and consequentially rate their SRH lower than someone who has been living with diabetes for a year and has a management program in place.
The latter two issues mentioned above point out a general limitation of research in this area. This limitation is one of consistent use of terms or standard definitions which allow for distinction between disability and chronic health condition or account for health condition severity or acuity. A chronic health condition as defined by the National Health Interview Survey (U.S. Department of Health and Human Services, 2006) is one which produces activity limitation and is either not cured or lasts three or more months; disability is any long- or short-term reduction of a person's activity as a result of an acute or chronic condition. Thus all permanent physical disabilities are chronic health conditions and all chronic health conditions are disabilities regardless of severity. Because the severity of a health condition is not accounted for, definition of health condition cannot be reliably measured by number of chronic health conditions alone. Control for acuity, severity, or temporal adjustment factors are a limitation of this study and others.
Future Research. Evaluation of the results of this study suggest a number of future research topics. There is strong need for research which incorporates measures of severity of chronic health conditions and account for acuity or adjustment to new conditions. Longitudinal research is needed to assess the relationships between changes in SRH and changes in chronic health conditions and other variables. Short-term longitudinal research could be used to examine the relation and predictive strength of SRH on future health outcomes and mortality for people with a disability, given its strong and independent predictive association with mortality in the gerontology research literature. Other research could assess in more detail the relative impact that specific chronic health conditions and secondary health conditions have on SRH, and what combination of health conditions, as well as their severity and acuity, is most strongly associated with SRH. Given the practical implication, it may be important to determine how closely SRH is to the perception of health by health professionals. Also, SRH for people with permanent physical disabilities may be more sensitive to changes in health than would be apparent to health professionals.
The self perception of health status for people with long-term permanent physical disabilities is based on a different set of factors than the self-perception of health status for people without permanent physical disabilities. While SRH for both those with and without disabilities involves pain and depression, for people with disabilities, the number of chronic health conditions, fatigue, and participation in community activities also influences the perception of their health. However, functional limitations and difficulty in performing basic ADL and IADL activities is not predictive of SRH and is not viewed as evidence of poor health by people with a disability. In contrast, for people without disabilities, beyond pain and depression, functional limitation and difficulty in performing ADLs and IADLs strongly and negatively contributes to their self-perceived health status. Although changes in physical abilities to carry-out ADLs and IADLs may influence SRH for both those with and without disabilities, the absolute state of limited functional capacity does not shape the perceptions of health status for people with long-term physical disabilities.
This research was supported in part by grant H133B031002 from the National Institute on Disability and Rehabilitation Research, Office of Special Education and Rehabilitative Services, U.S. Department of Education, Washington, D.C. We would like to thank Penni LaVoot, COTA, MS and Denise Diener, PT, MSG, for their assistance in and contribution to this article.
Benyamini, Y., & Idler, E. (1999). Community studies reporting association between self-rated health and mortality: Additional studies 1995 to 1998. Research on Aging, 21, 392-401.
Campbell, M., Sheets, D., & Strong, P. (1999). Secondary health conditions among middle-aged individuals with chronic physical disabilities: Implications for unmet needs for services. Assistive Technology, 11(2), 3-18.
Charlifue, S., & Gerhart, K. (2004). Community integration in spinal cord injury of long duration. Neuro Rehabilitation, 19(2), 91-102.
Cott, C. A., Gignac, M. A., & Badley, E. M. (1999). Determinants of self rated health for Canadians with chronic disease and disability. Journal of Epidemiology and Community Health, 53, 731-736.
Fillenbaum, G. (1988). Multidimensional functional assessment of older adults. HIllsdale, New Jersey: Lawrence Erlbaum.
Fillenbaum, G., & Smyer, M. (1981). The development, validity and reliability of the OARS Multidimensional Functional Assessment Questionnaire. Journal of Gerontology, 35, 428-434.
Finnegan, L., Marion, L., & Cox, C. (2005). Profiles of self-rated health in midlife adults with chronic illnesses. Nursing Research, 54(3), 167-177.
Han, B. (2002). Depressive symptoms and self-rated health in community-dwelling older adults: A longitudinal study. Journal of American Geriatrics Society, 50, 1549-1556.
Han, B., & Jylha, M. (2006). Improvement in depressive symptoms and changes in self-rated health among community-dwelling disabled older adults. Aging and Mental Health, 19(6), 599-605.
Han, B., Small, B., & Haley, W. (2001). The structure of self-rated health among community-dwelling older adults with stroke. Home Health Care Services Quarterly, 20(4), 1-15.
Hoeymans, N., Feskens, E. J., Kromhout, D., & Van den Bos, (3. A. (1999). The contribution of chronic conditions and dis abilities to poor self-rated health in elderly men. Journal of Gerontology." Medical Sciences, 54(A), M501-M506.
Idler, E., & Benyamini, Y. (1997). Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior, 38, 21-37.
Idler, E., Hudson, S., & Leventhal, H. (1999). The meanings of self-ratings of health. Research on Aging, 21(3), 458-476.
Kemp, B. (2005). What the rehabilitation professional and the consumer need to know. Physical Medicine and Rehabilitation Clinics of North America, 16, 1-18.
Kemp, B., & Adams, B. (1995). The Older Adult Health and Mood Questionnaire: A new measure of geriatric depressive disorder. Journal of Geriatric Psychiatry and Neurology, 8, 162-167.
Kemp, B., & Ettelson, D. (2001). Quality of life while living and aging with a spinal cord injury and other impairments. Topics in Spinal Cord Injury Rehabilitation, 6(3), 116-127.
Kemp, B., & Mosqueda, L. (Eds.) (2004). Aging with a disability: What the clinician needs" to know. Baltimore, MD: John Hopkins University Press.
Krause, J., Kemp, B., & Coker, J. (2000). Depression after spinal cord injury: Relation to gender, ethnicity, aging and socioeconomic indicators. Archives of Physical Medicine and Rehabilitation, 81, 1099-1109.
McKinley, W., & Meade, M. (2004). Community integration following SCI. Neuro Rehabilitation, 19(2), 79-80.
Mosqueda, L. (2004). Physiological changes and secondary conditions. In B. Kemp & L. Mosqueda (eds.) Aging with a disability: What the clinician needs to know (pp. 35-47). Baltimore, MD: John Hopkins University Press.
Mulsant, B., Ganguli, M., & Seaberg, E. (1997). The relationship between self-rated health and depressive symptoms in an epidemiological sample of community-dwelling older adults. Journal of American Geriatrics Society, 45, 954-958.
Putnam, M., Geenen, S., & Powers, L. (2003). Health and wellness: People with disabilities discuss barriers and facilitators to well being. Journal of Rehabilitation, 69, 37-45.
Schwartz, J., Jandorf, L., & Krupp, L. (1993). The measurement of fatigue: A new instrument. Journal of Psychometric Research, 17, 753-762.
Segovia, J., Bartlett, R., & Edwards, A. (1989). An empirical analysis of the dimensions of health status measures. Social Science Medicine, 29, 761-768.
SPSS, Inc. (2002). SPSS base 11.5 user's guide. Chicago: Author.
U.S. Department of Health and Human Services. (2006). Summary health statistics for the U.S. Population: National Health Interview Survey, 2004. DHHS publication No. (PHS) 2006-1561. Hyattsville, Maryland: author.
U.S. Department of Health and Human Services (2000). Disability and secondary conditions. In Healthy People 2010: Understanding and Improving Health (2nd ed.). Washington, DC: U.S. Government Printing Office.
Judith M. Mitchell
Rancho Los Amigos National Rehabilitation Center
Rodney H. Adkins
Rancho Los Amigos National Rehabilitation Center
Judith M. Mitchell, PhD, Research Director, Rehabilitation Research and Training Center on Aging, Rancho Los Amigos National Rehabilitation Center, 7601 E. Imperial Highway, 800 West Annex, Downey, CA 90242. Email: firstname.lastname@example.org
Table 1 Descriptive, SRH and Predictor Variables of Participants with and without a Disability With a Disability Without a Disability Variables n = 241 n = 83 Demographics, (M [SD]) Age 61.6 (13.0) 64.0 (14.1) Education 14.7 (2.9) 15.3 (2.6) % Women 67 63 % non-Hispanic white 82 75 Age at onset 16.3 (16.9) Duration 45.3 (19.6) Self rated health, % Excellent/ very good 35.7 68.7 Good 37.3 16.9 Fair 17.8 12.0 Poor 9.1 2.4 Predictor variables, (M [SD]) Depression 5.5 (4.4) 3.7 (3.6) Community activities 32.6 (12.8) 37.3 (12.0) No. chronic conditions 2.2 (1.5) 1.6 (1.3) Pain severity 1.9 (1.0) 1.1 (1.1) Fatigue 3.3 (2.5) 0.9 (1.9) IADLs 2.9 (2.0) 6.1 (1.7) ADLs 3.5 (2.1) 6.4 (1.2) Total Variables N = 324 t-test / [chi square] Demographics, (M [SD]) Age 62.3 (13.3) 1.4 Education 14.8 (2.9) 1.7 % Women 66 0.58 % non-Hispanic white 80 0.14 Age at onset Duration Self rated health, % 28.32 ** Excellent/ very good 44.1 Good 32.1 Fair 16.4 Poor 7.4 Predictor variables, (M [SD]) Depression 5.1 (4.3) 3.48 ** Community activities 33.8 (12.7) 2.96 * No. chronic conditions 2.0 (1.5) 3.53 ** Pain severity 1.7 (1.1) 6.11 ** Fatigue 2.7 (2.6) 8.00 ** IADLs 3.7 (2.4) 12.77 ** ADLs 4.3 (2.3) 11.73 ** * p<.01. ** p<.001. Table 2 Intercorrelations between SRH and Predictor Variables for People with and without a Disability Variables 1 2 3 4 People with a disability 1. SRH -- 2. Depression .43 *** -- 3. Community activities -.36 *** -.42 *** -- 4. No. chronic conditions .21 *** .11 .02 -- 5. Pain severity .31 *** .30 *** -.08 .21 *** 6. Fatigue .28 *** .33 *** -.07 .15 * 7. IADLs -.13 * -.16 ** .07 -.14 * 8. ADLs -.11 -.11 .09 -.05 People without a disability 1. SRH -- 2. Depression .47 *** -- 3. Community activities -.30 ** -.45 *** -- 4. No. chronic conditions .39 *** .27 * .05 -- 5. Pain severity .53 *** .41 *** -.21 .40 *** 6. Fatigue .29 ** .60 *** -.23 * .37 *** 7. IADLs -.48 *** -.40 *** .36 *** -.52 *** 8. ADLs -.33 ** -.33 ** .31 ** -.48 *** Variables 5 6 7 8 People with a disability 1. SRH 2. Depression 3. Community activities 4. No. chronic conditions 5. Pain severity -- 6. Fatigue .27 *** -- 7. IADLs -.17 ** -.17 ** -- 8. ADLs -0.12 -0.02 .76 *** -- People without a disability 1. SRH 2. Depression 3. Community activities 4. No. chronic conditions 5. Pain severity -- 6. Fatigue .34 ** -- 7. IADLs -.38 *** -.35 *** -- 8. ADLs -.34 *** -.38 *** .79 *** -- * p<.05.** p<.01.*** p<.001. Table 3 Results of the Stepwise Multiple Regression Analyses Variables B Std. Error People with a disability Constant 2.52 .25 Depression .06 .02 3.5 *** Community activities -.02 .01 4.0 *** Pain severity .17 .06 No. chronic conditions .10 .04 Fatigue .06 .03 People without a disability Constant 2.71 .41 Pain severity .30 .09 3.4 *** IADLs -.15 .06 Depression .06 .03 Variables Beta t People with a disability Constant Depression .23 3.5 *** Community activities -.24 - 4.0 *** Pain severity .16 2.7 ** No. chronic conditions .14 2.5 ** Fatigue .13 2.2 * People without a disability Constant Pain severity .34 3.4 *** IADLs -.27 -2.7 ** Depression .22 2.2 * * p<.05.** p<.01.*** p<.001. People with a disability: R = .55, R2 = .30, Adjusted R2 = .29. People without a disability: R = .64, R2 = .41, Adjusted R2 = .39.