Abstract
Objective:
To examine how race and the diagnostic label of Osteoarthritis (OA) affects older adults’ emotions, illness beliefs, and willingness to help a family member.
Methods:
African American and White older adults were randomly assigned to read vignettes describing a sister suffering from chronic pain and disability, either with or without the OA label. Race × diagnostic label ANOVAs were conducted.
Results:
Compared to Whites, African Americans were more optimistic that OA could improve with health care, and showed greater willingness to help their sister. The OA label had little impact on emotions, beliefs, or willingness to help. African Americans rated the sister as having more control of their problem than Whites without the OA label, but providing the diagnosis eliminated this difference.
Discussion:
The diagnostic label of OA had little effect on these older adults, but racial differences indicate that cultural values regarding family caregiving are important in arthritis care.
In the United States, arthritis is the leading cause of disability (Centers for Disease Control and Prevention [CDC], 2010b) and affects more than 70 million Americans (CDC, 2003, 2010b). Arthritis prevalence rates are expected to rise dramatically with the aging of our population (CDC, 2010b; Hootman & Helmick, 2006). Osteoarthritis (OA), the most common form of arthritis, often results in pain and physical limitations (CDC, 2008; National Institute of Arthritis and Musculoskeletal and Skin Disease [NIAMS], 2010). Although the risk of OA increases with age (Arthritis Foundation, 2010; CDC, 2010a), OA is not a normal part of aging and can be improved with behavioral and pharmacological interventions (Anderson & Loeser, 2010; Fitzcharles, Lussier, & Shir, 2010; Hochberg et al., 2012). Previous research shows that young adults, older adults, and individuals diagnosed with OA commonly believe that OA is a part of normal aging, that it cannot be improved with treatment, and attribute pain and joint swelling to age and not to illness (Appelt, Burant, Siminoff, Kwoh, & Ibrahim, 2007; Mingo, McIlvane, & Haley, 2006). Perceptions of illness are potentially tied to an individual’s health-care decisions and ultimately could decrease the likelihood of seeking optimal health care (e.g., seeking a diagnosis for arthritis-related symptoms, utilizing beneficial treatment options). Thus, it has been suggested that helping older adults to view pain and disability as being due to arthritis, and not to old age, would increase appropriate use of health care (Fitzcharles et al., 2010).
In addition to the perceptions and beliefs of the individual with arthritis, family members’ and caregivers’ perceptions about a condition as well as cultural beliefs could impact the health care one seeks or even the support provided to the family member in need of care (Dilworth-Anderson & Gibson, 2002; Hagger & Orbell, 2003; Mingo et al., 2006; Thomas, McIlvane, & Haley, 2012). For example, perceptions that pain and disability are signs of aging and untreatable may result in family members accepting these problems as normal and not pursuing useful health care. When symptoms are attributed to a disease rather than seen as the responsibility of a person, family members are also more sympathetic and willing to provide caregiving assistance (Wadley & Haley, 2001). So disease labels may have effects on both patients, and family members who are in a position to provide care or guidance concerning health-care use.
Research demonstrates that certain illness labels (e.g., Alzheimer’s disease, blindness, cancer, heart disease) can result in positive perceptions and positive responses and other illness labels (e.g., AIDS, obesity, drug addiction) can be stigmatizing (Menec & Perry, 1995; Wadley & Haley, 2001; Weiner, 1993; Weiner, Perry, & Magnusson, 1988). Attribution theory suggests that disease labels can increase the perception that the causes of a problem are outside of the person, and beliefs that an individual deserves sympathy, that their problems are not their responsibility, and increase willingness to provide assistance (Menec & Perry, 1995; Wadley & Haley, 2001; Weiner, 1993). Previous research has shown that family members’ perceptions of an older adult experiencing symptoms can be affected by providing a diagnostic label to these symptoms (Mingo et al., 2006; Thomas et al., 2012; Wadley & Haley, 2001). Wadley and Haley (2001) gave undergraduate students vignettes describing an older parent showing memory problems and disruptive behaviors. Students were either given no diagnostic label, or told that the problems were due to Alzheimer’s disease. Participants provided with a diagnostic label of Alzheimer’s disease reported higher ratings of sympathy, decreased perceptions that the person was responsible for or had control of their problem, and greater willingness to help compared to those who did not receive a diagnostic label. Two previous studies have examined the impact of the disease label of OA on perceptions of family members. Both studies were conducted with undergraduate students, who read about an older parent experiencing pain and disability. Mingo et al. (2006) found that participants who were told that these symptoms were due to OA were more likely to rate their relatives’ problems as due to illness than controls who did not receive the OA label. A follow-up study using similar vignette methodology found that providing the OA label coupled with educational information led the family members to show more sympathy, attribute less responsibility, and less control while increasing helping intentions in comparison to participants not receiving a disease label (Thomas et al., 2012). However, we are not aware of any studies that have examined the effects of the OA disease label on older adults.
Besides diagnostic labels, race can also be very important in determining the health advice a family member may give or the care one is willing to provide. There are marked racial differences in regards to the perceptions of arthritis treatments as well as treatment decisions (Ang, Ibrahim, Burant, Siminoff, & Kwoh, 2002; Ibrahim, Siminoff, Burant, & Kwoh, 2002). For example, in comparison to Whites, African Americans are more likely to use home remedies, use over-the-counter medications, and perceive prayer as a helpful treatment option for arthritis (Ang et al., 2002; Ibrahim, Siminoff, Burant, & Kwoh, 2001). African Americans are also more likely than Whites to report being likely to ask a friend or family for advice concerning arthritis treatment decisions (Ibrahim et al., 2001). These racial differences are thought to be related to cultural factors such as mistrust of the health-care system, and sometimes to culture specific beliefs about illness (Ang et al., 2002; Figaro, Russo, & Allegrante, 2004). Research on family caregiving often shows a greater willingness for African Americans than Whites to feel a cultural obligation to provide care for impaired relatives, including more distant relatives (Dilworth-Anderson et al., 2005). Among young adults, in the vignette study cited above, African American students were more willing than Whites to report that they would provide assistance to a parent with arthritis symptoms, including providing them with daily care and having the parent live with them if necessary (Mingo et al., 2006). While previous research suggests that diagnostic labels of OA and race can have important effects on perceptions of arthritis, the previous experimental studies of disease labels and OA have been conducted only with undergraduate students.
It is important to understand the perceptions of not only younger family members who might help older loved ones with arthritis, but also older adults themselves who may be at a greater risk of having arthritis or being a caregiver of an individual with arthritis, specifically OA. This greater experience with OA and caregiving may have unknown effects on beliefs about OA, emotional responses to impairment, and helping intentions. Therefore, the purpose of the present study was to examine the effects of the OA diagnostic label and race on older adults’ perceptions of a relative with pain and disability. Similar to previous research, we investigated whether this disease label might affect individuals’ emotions, illness beliefs, and willingness to help a family member experiencing pain and disability. Based on previous findings with younger persons, we expected the following:
in older adults the diagnostic label of OA would not affect emotions or willingness to help.
in older adults the diagnostic label of OA would increase perceptions that the problems were due to illness.
in older adults the diagnostic label would decrease perceptions that the person experiencing problems was either responsible for or in control of their difficulties.
older African Americans would be more likely to report positive emotions.
older African Americans would be less likely to believe that the person with arthritis was responsible or in control of their problems.
older African Americans would be more likely to believe that they could benefit from better health care.
older African Americans would be more willing to provide assistance to a relative than their White counterparts.
Method
Participants
The project focused on White and African American older adults. In order to be eligible for the study, participants had to be above the age of 60 and self-identify as White or African American. This was a convenience sample recruited from a number of multipurpose senior centers and senior living complexes in both the Tampa, Florida and Atlanta, Georgia area. Study flyers and the snowball method were used to recruit and inform participants about the study. Participants included 85 Whites and 113 African Americans for a total sample of 198. Descriptive information on the participants is shown in Table 1. White participants were less likely to be female (p = .03) and significantly older (p < .001) than African American participants. Approximately 72% (n = 142) of the sample reported having a sister; there were no significant differences by race or diagnosis. African Americans reported being in contact with their actual sister more frequently (p = .01) and feeling closer to their sister (p < .01) in comparison to the White participants.
Descriptive Information for Study Variables (N = 198).
Procedure
Participants were given packets that included demographic information, a vignette, and questions that asked them to provide their responses to the vignette. These packets were randomly distributed to individuals who had agreed to take part in the study. Participants read one of two vignettes that differed only in whether a diagnostic label was provided for the symptoms that occurred. The vignette described a 75-year-old sister who was experiencing severe pain in her joints, and having trouble maintaining her daily activities. Immediately after reading the vignette, participants completed a series of questions that assessed the participants’ emotions, attributions and illness beliefs, helping intentions, and their relationship with their own sister. The vignette was very similar to ones used in two previous studies (Mingo et al., 2006; Thomas et al., 2012). The questions concerning emotions, attributions and illness beliefs, and willingness to help were developed by Wadley and Haley (2001) and were used in two additional studies (Mingo et al., 2006; Thomas et al., 2012). The only difference in our questions and previously used vignettes and questions was the designation of the relative as a sister versus a parent, and specification of age of the relative as 75. After answering these questions, participants were debriefed by providing a clear explanation of the purpose of the study. The research study was approved by the University of South Florida Institutional Review Board.
Vignette
The vignette described a sister experiencing severe pain in her joints, which was interfering with her ability to fulfill her normal duties around the house. In the second version of the vignette, as indicated by the text in brackets, she was described as being diagnosed with OA. Participants were instructed to read the vignette and imagine the situation was happening to them. The vignette stated: You are on vacation visiting your 75-year-old sister, Mary, for a week after not seeing her for a year. When you arrive, you are surprised to find that your sister has not prepared a home-cooked meal as she always does for your visits. You are also alarmed at the fact that the house is in disarray when you know that your sister has always been picky about a clean house. Although you do not see anything wrong with your sister, during your visit she constantly complains of stiffness in her hips and knees, and of not being able to stand or walk right away.
Measures
Emotions, attributions and beliefs about illness, and willingness to help were assessed with a series of questions that were rated on an 11-point Likert-type scale. Participants were instructed, “Based on the vignette you just read, please read the following questions, and circle the number that best describes your reactions to your sister in this situation.” Participants were also asked if they had a sister, and if so to answer questions about their relationship with that sister. These questions were also modeled after similar questions asked of participants in previous vignette studies (Mingo et al., 2006; Thomas et al., 2012; Wadley & Haley, 2001).
(Emotions)
How angry do you feel toward your sister? (0 = not angry at all, 10 = extremely angry)
How sympathetic do you feel toward your sister? (0 = not at all sympathetic, 10 = extremely sympathetic)
(Attributions and beliefs about illness)
How much do you think your sister’s problem is due to old age? (0 = not at all due to old age, 10 = extremely due to old age)
How much do you think your sister’s problems are due to illness? (0 = not at all due to illness, 10 = extremely due to illness)
How responsible is your sister for her problem? (0 = not at all responsible, 10 = extremely responsible)
How much do you think your sister is able to control her problem? (0 = no control, 10 = extreme control)
How likely is it that your sister’s problems could improve if she received better medical care? (0 = not at all likely, 10 = extremely likely)
(Helping intentions)
If your sister’s problems worsen how willing are you to help her on a daily basis? (0 = not at all willing, 10 = extremely willing)
If your sister’s problems worsen how willing are you to have your sister move in with you? (0 = not at all willing, 10 = extremely willing)
In order to assess whether participants’ actual relationships with sisters affected their ratings, they were asked if they had a sister and if so they were instructed to “Now please think about feelings toward your own sister, and answer the following questions.”
(Feelings toward your own sister)
How close do you feel to your own sister? (0 = not at all close, 10 = extremely close)
How often are you in contact with your sister? (0 = never, 10 = frequently)
How much conflict do you have with your own sister? (0 = no conflict, 10 = extreme conflict)
Statistical Analysis
A series of 2 × 2 (race × diagnostic label) univariate analyses of variance (ANOVA) examined the effects of race (African Americans and Whites) and a diagnostic label (OA label and No label) on one’s emotions (anger and sympathy), attributions of responsibility and control, beliefs about illness (problem due to old age, problem due to illness, sister’s improvement due to better health care), and helping intentions (willingness to help sister daily, willingness to have sister move in). A series of analyses of covariance (ANCOVAs) were conducted to adjust for covariates.
Results
Preliminary Analyses
Preliminary analyses were conducted to examine whether age, gender, and the questions about an actual sister were related to the dependent variables. When these relationships were significant, they were included as covariates. Therefore, an ANCOVA was used to assess the effects of the independent variable after controlling for age, gender, the amount of contact with one’s actual sister, and closeness with one’s actual sister. Henceforth, the latter two variables will be referred to as contact and closeness, respectively.
Primary Analyses
Emotional Responses
2 × 2 ANOVA were used to examine the effects of race and diagnostic label on anger and sympathy. Although there was a trend (p = .059) for African Americans to report less anger toward the sister than Whites, no significant effects for race or label, or their interaction, were found for anger or sympathy. Results indicate after controlling for age, gender, contact, and closeness, there were no significant covariate effects, and there were no significant race (F[1, 114] = .98, p = .32) or diagnostic label (F[1, 114] = .70, p = .40) differences for anger.
Attributions and Illness Beliefs
The ANOVAs for problem due to old age, problem due to illness, and responsibility yielded no significant main effects or interactions. There was a significant main effect of race on the belief that the sister’s problems would improve with better health care (F[1, 194] = 7.98, p = .01). As shown in Table 1, African Americans were more likely to believe that the sister’s problem could improve with better health care compared to Whites. After adjusting for covariates (i.e., age, gender, contact, closeness), the covariate effect for age was significant (p = .03), and the effect of race on the belief that the sister’s problem would improve with better health care was no longer significant (F[1, 114] = .88, p = .35). However, there was a trend for a race × diagnostic label interaction effect (F[1, 114] = 3.34, p = .07).
The ANOVA examining attributions of control revealed a significant main effect for race (F[1, 194] = 4.56, p = .03) and a race ×diagnostic label interaction effect, (F[1,194] = 6.67, p = .01). Post hoc analyses using the Tukey HSD (honestly significant difference) test were conducted to interpret the interaction. Statistical significance was determined by using < .05 as the probability level. As shown in Figure 1, in the absence of a diagnostic label, African Americans attributed more control to the sister than did Whites. When the diagnostic label was available, these racial differences were eliminated. Results indicated that after controlling for age, gender, contact, and closeness, there were no significant covariate effects, there was a trend for the race effect (F[1, 114] = 3.65, p = .06), and the interaction effect remained (F[1, 114] = 7.39, p = .01).
Race and diagnostic differences in control.
Helping Intentions
The ANOVA for helping intentions revealed two main effects for race, including willingness to help the sister on a daily basis, (F[1,194] = 24.22, p < .001), and willingness to have the sister move in, (F[1,194] = 31.30, p < .001). Congruent with hypothesis seven, African American participants were more likely to report a willingness to help the sister on a daily basis and to have the sister move in with them in comparison to their White counterparts. These results are shown in Table 1. No effects were found for either variable for the main effect of diagnostic label or the interaction of race and diagnostic label. After conducting an ANCOVA, the covariate effect for closeness was approaching significance (p = .05), and the effects of race on willingness to help the sister on a daily basis remained (F[1, 114] = 5.17, p = .03). No other significant effects remained after adjusting for covariates.
Discussion
Research focused on rheumatic conditions has given very little attention to the impact of a diagnostic label, particularly an OA diagnosis. The long-held belief that arthritis is a normal part of aging may have led to this lack of attention. The purpose of the present study was to examine the impact of an OA diagnostic label on an older adult’s emotions, beliefs, and willingness to help a family member, and to examine racial differences. With one exception, we did not find evidence to support hypotheses two and three, which state that the diagnostic label of OA had an impact on attributions and illness beliefs. However, an interaction between race and the diagnostic label for control was found; African Americans attributed more control to the family member when not presented with a diagnostic label. The OA label eliminated racial differences in attributions of control, which were higher among African Americans than among Whites when no label was provided. These findings suggest that the diagnostic label of OA has a less potent effect on older adults than among younger persons. Mingo et al. (2006) had found that provision of a diagnostic label of OA affected both attributions that problems were due to illness, and that better health care could improve the problems. Thomas et al. 2012 found that the diagnostic label, when accompanied by additional health-related information, led to less anger and more sympathy, less attribution of control and responsibility, and more willingness to help. There may be a number of reasons for the lack of a labeling effect for older participants. Among older adults, symptoms described in the vignettes such as knee and hip pain, trouble walking, and limitations in activities are much more common than in college student samples used in previous studies. Thus, older people may have more stable attitudes and beliefs about OA, and a simple label may have little effect. This finding supports previous research that suggests in comparison to young adults, older adults are more likely to associate disease and disease related symptoms with aging (Leventhal & Prohaska, 1986; Prohaska, Leventhal, & Keller, 1985) particularly arthritis (Appelt et al., 2007). Such prominent and strongly held beliefs may be difficult to alter. It may be that a diagnostic label by itself is not very informative to older adults, and future studies should examine whether a label plus information, or other more potent manipulations, could alter these attributions.
Consistent with our prediction (i.e., hypothesis seven) and the previous literature, African American participants had higher levels of helping intentions. These findings are in line with the previous study conducted by Mingo et al. (2006) that found when compared with Whites, college-aged African American students were more likely to state that they would help the family member in need of assistance, regardless of their diagnosis. Results indicate that the obligation to care for a family member goes beyond a filial responsibility. Even for older adults caring for other older adults, cultural values may have an effect on the care one is willing to provide. Our findings support research that has found that African Americans feel obligated out of cultural responsibility to provide care for a family member in a time of adversity (Dilworth-Anderson et al., 2005) regardless of age and circumstance. It is important that medical providers are aware of the emphasis on cultural justification within this particular group as it is likely to impact the care of an older adult, particularly those with arthritis. There was also a trend (p = .059) for African Americans to report less anger toward the sister than White participants. While this should be interpreted cautiously since it did not reach conventional levels of significance, this finding is consistent with the idea that African Americans are more accepting of impaired relatives and place high cultural value on providing care to disabled relatives (Dilworth-Anderson et al., 2005).
The finding that African Americans were more likely than Whites to think that the condition of the family member could be improved with better health care was consistent with previous research done with college-aged individuals (Mingo et al., 2006) and supported hypothesis six. While findings need to be further investigated, it may be that there is a connection between this belief and access to health care. According to the literature, minorities disproportionately lack access (i.e., the right and opportunity) to adequate health care (Institute of Medicine [IOM], 2003). Therefore, this finding may not be unique to OA, but relevant to the idea that if one had generally better access to health care, health conditions would be improved. In concordance with general health disparities, access poses a problem for minorities with arthritis (Song, Chang, Tirodkar, Chang, Manheim, & Dunlop, 2007). In comparison to Whites, minorities with arthritis are less likely to have adequate medical access resources, with access factors explaining 60% of excess risk of developing disability for African Americans (Song et al., 2007). In order to continue to make efforts to eradicate health disparities, future research should examine the impact of access on perceptions of health conditions. In support of our hypothesis (i.e., hypothesis 2), the diagnostic label did not affect perceptions on helping intentions. Thomas et al. (2012) found that coupling a diagnostic label with additional health-related information about aging or arthritis increased helping intentions. Future work with older adults could assess whether additional information about health and OA yielded similar effects.
While covariance analyses did not substantially alter the results, adjustment for covariates eliminated several significant racial effects, including the effects of race on the belief that the sister’s problem would improve with better health care, and willingness to move in with the sister. This suggests that racial differences may be partially explained by racial differences in significant covariates such as age and closeness to the family member.
There are several limitations in the study. The most important limitation is that the project used convenience sampling of older adults. Examination of these issues with population studies would improve the external validity of the results. Another important limitation to consider is the vignette methodology. Although vignettes have been well accepted for testing models of social behavior (Alexander & Becker, 1978; Weiner, 1980) in a variety of health-related decision making issues (Weisse, Sorum, Sanders, & Syat, 2001), vignettes only posit hypothetical situations that are not assumed to be identical to responses in actual situations. Use of vignettes that vary additional specific details of the hypothetical situation has also been recommended, as a way to assess whether these specific elements of vignettes (e.g., one week visit time, tone of expression of pain) might have effects, and to allow for more generalizable and varied vignette stimuli (Taylor, 2006). Future studies with larger samples could include such methodological innovations. Despite the methodological limitations, the study designed allowed us to control for an important variable (i.e., diagnostic label) and was modeled after other vignettes that have shown to produce valid findings. In addition, prior laboratory studies using this methodology have yielded findings that are consistent with those from field studies (Sacco & Dunn, 1990) giving us confidence in the validity of our vignette findings.
In conclusion, our findings reinforce extensive findings that family members play important roles in caring for older adults with conditions such as arthritis, and that older African Americans report especially strong intentions to care for their older relatives. Health-care providers should be aware of the role that African Americans play in caring for someone with arthritis symptoms and provide information to both the patient and their family members. Our results also suggest that, in contrast to younger persons, simple interventions such as communicating a diagnosis may have minimal impact on altering attributions of older adult caregivers. Arthritis is a growing public health concern and our findings along with previous research supports the need for educational and self-management arthritis interventions that target older adults with arthritis and older adults caring for someone with arthritis as a way to minimize the negative individual and societal impact of the condition. Health-care providers could also play a role by discussing perceptions of arthritis and arthritis-related symptoms with older adult patients and caregivers, and using this conversation to dispel myths that arthritis is simply due to old age and is untreatable. Considering that arthritis is the leading cause of disability in the United States (CDC, 2010b), efforts of this type could potentially minimize disability and enhance the public health of our older-adult population. It is important that we continue our efforts to understand the influence of race/culture and such issues as diagnostic labels and health information on health and health-care disparities and use this knowledge to improve care for diverse older adults.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Arthritis Foundation, Florida Chapter Student Fellowship Award and in part by the National Institute on Aging Training Grant T32-AG000272. The research study was approved by the University of South Florida Institutional Review Board (Protocol# - 103837).