Arthritis includes more than 100 diseases and conditions that affect joints and connective tissue (Centers for Disease Control and Prevention [CDC], 2017). The prevalence of arthritis increases with age and is higher among women and adults with obesity. Arthritis is a leading cause of disability, with approximately 43% (22.7 million) of Americans with diagnosed arthritis reporting activity limitations.
From 1997 to 2005, the number of Americans with arthritis comorbidities (e.g., diabetes, heart disease) increased from 36.8 to 44.9 million (CDC, 2017). Compared to people without arthritis as one comorbidity of multiple chronic conditions (MCCs), those with arthritis and comorbidities had higher prevalence of social participation restriction, serious psychological distress, and work limitations (Qin et al., 2015). However, the characteristics of people aging with multimorbidity have not been well described in terms of symptom experience or health resources (Tripp-Reimer et al., 2020). The complexities involved in managing MCCs, especially the associated chronic disabling symptoms, could explain poor outcomes in self-management, health care use, and quality of life (QoL) (Arthritis Foundation, 2019; CDC, 2017; Han & Han, 2016; Radner, 2016; Tripp-Reimer et al., 2020; van Onna & Boonen, 2016; Xu et al., 2017). Self-management optimization for MCCs thus requires more adaptive strategies and behaviors to align with individuals' health priorities as the disease-specific evidence-based guidelines may not be applicable (Boyd et al., 2019; Tripp-Reimer et al., 2020).
Self-regulation, defined as health promotion through self-monitoring of health behaviors and social support, is the main adaptive component of self-management (Moore et al., 2016). The self-regulation strategies of selection, optimization, and compensation (SR-SOC) are goal-directed coping strategies to adapt to stresses with deliberate resource seeking and use by the individual (Freund & Baltes, 2002). Selection refers to goal setting by individuals and includes elective selection of goals that are important for individuals' health and well-being, and loss-based selection of goals adapted to aging-related loss (Freund & Baltes, 2002). Optimization is the means to manage and optimize relevant skills to achieve the selected goals based on individuals' capabilities (Freund & Baltes, 2002). Compensation is the use of alternative means to compensate for aging-related losses and deficits and maintain performance using accumulative wisdom and external or technical aids (Freund & Baltes, 2002). A literature review found that SR-SOC strategies guided middle-aged and older adults with chronic diseases, such as arthritis or multimorbidity, to adapt to the associated chronic disabling symptoms (e.g., pain, fatigue, cognitive impairment, sleep disturbance, depression, anxiety) and manage complex health issues to live well and meaningfully (Zhang & Radhakrishnan, 2018). Older adults with osteoarthritis applied SR-SOC strategies to adapt to disabilities, where selection included “giving up, restricting, or limiting activities”; optimization consisted of “increased planning and effort to engage in activities, using movement, and periods of rest to avoid pain or to augment energy reserves”; and compensation comprised the “use of assistive devices, gadgets, and furniture as well as modifying or substituting behaviors or activities such as choosing appropriate shoes” (Gignac et al., 2002, p. 522). A qualitative study also concluded the themes of self-management in the SOC domains and emphasized the value of leisure activities and leisure-incorporated education for arthritis self-management, such as “commitment to leisure for health purposes” (Janke et al., 2012, p. 364). The strategy uses differed by available resources, such as social support from medical professionals (Janke et al., 2012). SOC training significantly (p < 0.05) improved the effects of acceptance and commitment therapy for older adults with chronic musculoskeletal pain from arthritis in pain acceptance, pain-related anxiety, and pain interference in walking ability (Alonso-Fernandez et al., 2016).
Degree of disability has been found to serve as the mediator between the number of chronic conditions and SR-SOC strategies (Yuen & Vogtle, 2016). In a longitudinal study among older adults with MCCs, negative self-perceived aging moderated the effect of SR-SOC strategies on self-rated health and life satisfaction after the occurrence of a serious health event (Wurm et al., 2013). Figure 1 shows the synthesized theoretical framework of Optimizing Aging with MCCs to understand self-regulation strategy use in managing arthritis and MCCs (Zhang et al., 2020). Based on the theoretical framework, previous study on the outcomes has found that physical symptoms; SR-SOC strategies, especially optimization; and income adequacy significantly predicted arthritis self-efficacy (Zhang et al., 2020). Arthritis self-efficacy and mental and physical symptoms significantly predicted QoL (Zhang et al., 2020).
Theoretical framework model: Optimizing aging with multiple chronic conditions (arthritis and comorbidities).
The current study aimed to describe the health resources and chronic disabling symptom characteristics of community-dwelling people aging with arthritis and MCCs and examine how these factors influence and predict SR-SOC strategy use after controlling for demographics and comorbidities. Thus, we proposed to answer two questions: (1) What are the health resources and chronic disabling symptom characteristics of community-dwelling people aging with arthritis and MCCs; and (2) How do the health resources and chronic disabling symptoms influence and predict SR-SOC strategy use after controlling for demographics and comorbidities? Our hypotheses were: (a) community-dwelling people aging with arthritis and MCCs on average report less health resources with more frequent chronic disabling symptoms than those with arthritis in general in prior studies; and (b) demographics, comorbidities, health resources, and chronic disabling symptoms are associated with and predict SR-SOC strategy use among community-dwelling people aging with arthritis and MCCs.
Data were collected from February to November 2018. The university Institutional Review Board (IRB) approved the study. Participants were recruited from local aging communities, mainly senior centers, and apartments. Gatekeepers of the settings were contacted about the study and provided with flyers for distribution. Flyers about the study and the contact information of the investigators were posted at the settings and online, with permission. People interested in the study contacted the principal investigator (PI) for eligibility. If eligible, participants chose one-time on-site or online surveys. In addition, with permission by the gatekeepers, the PI set up tables and gave presentations on the study, with arthritis self-management health information hand-outs to appeal to the community residents for one-time on-site data collection. On-site, purposeful, and snowball sampling helped recruit more minorities, such as female, low-income, disabled, and racial or ethnic minorities to avoid selection bias. Each participant received 5 US dollars as compensation.
Inclusion criteria were: (a) age >50 years; (b) community-dwelling; (c) diagnosed with arthritis and at least one other comorbidity; and (d) able to read, write, and understand English. Written materials used larger fonts for those with mild vision problems.
Gignac et al.'s (2002) study used 15 independent variables to predict the frequencies of SOC (N = 209, R2 = 0.31 to 0.59). Effect sizes of regressions ranged from 0.28 to 0.57 in Gpower 3.1 (Erdfelder et al., 1996). Based on these effect sizes, 17 independent variables in this study required a sample size of 67 to 119 (Erdfelder et al., 1996). One hundred forty individuals completed surveys, of which 114 to 120 participants had complete data to predict SOC frequency subscale scores and total scores. The rates of missing data thus ranged from 14.3% to 18.5%.
Functional Comorbidity. The Functional Comorbidity Index (FCI) measured participants' 18 diagnosed comorbidities. The scale was developed with data from 9,423 Canadian adults and 28,349 American adults treated for spine ailments (Groll et al., 2005). The FCI associated with physical function more than the Charlson and Kaplan-Feinstein Comorbidity Indexes. The FCI was scored as a simple count (yes/no) of the diagnoses. Arthritis types and a fill-in space of unlisted chronic conditions were added to the FCI items.
Health Literacy. The 4-item Brief Health Literacy Screening Tool (BRIEF) evaluated self-reported health literacy (Haun et al., 2012). Participants were asked about their reading, learning, and understanding of health information, with scores ranging from 1 (always/not at all confident) to 5 (never/extremely confident), and total scores ranging from 4 to 20 (4 to 12, inadequate; 13 to 16, marginally adequate; 17 to 20, adequate), with higher scores indicating better health literacy. An example of a question is: “How often do you have a problem understanding what is told to you about your medical condition?” This scale significantly associated with the Short Assessment of Health Literacy (r = 0.42) and Rapid Estimate of Adult Literacy in Medicine (r = 0.4). In the current study, Cronbach's alpha for the scale was 0.84.
Social Network. The Lubben Social Network Scale (LSNS-6) measured social network with number of relatives and friends available for help and support (Lubben et al., 2006). Participants rated their experiences in the past 6 months regarding items such as, “How many relatives do you see or hear from at least once a month?” Responses to the six items were scaled from 0 (none) to 5 (nine or more), with the total score ranging from 0 to 30. Higher total scores indicate better social network and support. The LSNS-6 associated with mortality, hospitalization, health behaviors, depressive symptoms, and overall physical health and was found valid among older adults (Lubben et al., 2006). In the current study, Cronbach's alpha was 0.89 for the total scale and 0.88 for the two subscales.
Patient–Health Care Provider Communication. The Patient–Health Care Provider Communication Scale assesses patients' perceived communication quality with health care providers (Salt et al., 2013). Participants rated the behaviors of health care providers and themselves through items such as “[health care provider] takes my health concerns seriously” or “[patient] feels comfortable telling health care provider about health concerns.” Responses to the 21 items are scaled from 1 (not at all like) to 4 (very much like). The scale comprises two sub-scales: quality communication and negative patient–health care provider communication. Higher total scores indicate better support from health care providers, and total scores range from 21 to 84. In the current study, Cronbach's alpha was 0.96, with 0.98 for the quality communication sub-scale and 0.73 for the negative communication subscale.
Health Insurance. Participants chose their health insurance coverage type from “employment-based, direct-purchase, Medicare, Medicaid, Military Health Care, uninsured, other.” This measure was used by the U.S. Census Bureau (Vornovitsky, 2015). The total number of insurance coverage types was computed because some people have multiple types of coverage to supplement their primary insurance or as a switch (Berchick et al., 2018).
Self-Reported Health Measures. The Patient-Reported Outcomes Measurement Information System (PROMIS) Adult Self-Reported Health Measures (PROMIS Item Bank v1.0–2.0, Short Form 4a, 4 items) were used to evaluate the symptom severity of pain interference, fatigue, emotional distress–depression, emotional distress–anxiety, sleep disturbance, and cognitive abilities (HealthMeasures, 2017). Participants rated how they experienced symptoms through questions such as, “How much did pain interfere with your day-to-day activities?” in the past 7 days. Responses are scaled from 1 (not at all/never) to 5 (very much/always/very often) on how frequently patients have experienced the symptoms. Total scores range from 4 to 20. Higher scores indicate higher symptom influences on pain interference, fatigue, depression, anxiety, and higher cognitive abilities. In the current study, Cronbach's alpha of the PROMIS scales ranged from 0.77 (sleep disturbance) to 0.95 (pain interference).
Selection, Optimization, and Compensation. The SOC Questionnaire evaluates the use of SR-SOC strategies in daily life. It includes four parts: elective selection, loss-based selection, optimization, and compensation. Participants rated the items considering their healthy lifestyles in the past 6 months through items such as, “When I think about what I want in life, I commit myself to one or two important goals” (Freund & Baltes, 2002). Examples such as commitment to leisure or health purposes from prior arthritis and MCC studies were provided (Gignac et al., 2002; Janke et al., 2012; Yuen & Vogtle, 2016). When responding to the questionnaire items, participants rated their frequencies of using such strategies from 1 (never) to 5 (always). For the current study, the short 12-item version with three items for each of the four subscales separately measuring elective selection, loss-based selection, optimization, and compensation strategies, and 5-point Likert scale was used as it contained the highest total and sub-scale internal consistencies (Zhang & Radhakrishnan, 2018). Total scores range from 1 to 60. In the current study, Cronbach's alpha was 0.91 for the total scale and ranged from 0.7 (compensation) to 0.86 (selection) for the subscales.
Descriptive analyses were used to describe the demographic characteristics, MCCs, health resources, symptoms, and SR-SOC strategies. Pearson correlations were used to demonstrate the relationships among demographic variables, MCCs, health resources, symptoms, and SR-SOC strategies. Principal component factor analysis resulted in the symptom clusters. This yielded two symptom clusters, mental and physical. The mental symptom cluster included depressive, anxiety, and sleep symptoms, whereas the physical symptom cluster included pain, fatigue, and cognitive symptoms. Using them instead in the regression analysis avoided the multicollinearity assumption violation. This method was similar to a prior study on multiple sclerosis (Zhang et al., 2017b).
Variables correlated with the dependent variables with p < 0.25 were included in the following regression analyses as predictors. Hierarchical multivariate stepwise regressions were used to demonstrate how demographics, MCCs, health resources, and chronic disabling symptoms predicted SR-SOC strategies. In the hierarchical multivariate stepwise regression analyses, correlated factors were entered hierarchically based on the theoretical framework; within each level's block, the stepwise method was used to obtain the best precise regression models that predicted outcome variables. Alpha was set at p < 0.05 (two-tailed) for all analyses. Assumptions were tested and not violated.
Mean substitution was used for the scales with less than 15% missing items. For those missing more than 15%, multiple imputation was used to generate five imputed datasets. Multiple values were used to estimate for each item of missing data using statistical models and then pooled for the final analysis. Sensitivity analyses compared whether the conclusions were sensitive to assumptions about the pattern of missing data (Zhang et al., 2017a).
Sample Description and Characteristics
Table 1 summarizes the background characteristics of the sample. The sample comprised mostly female participants (70%), with a mean age of 71 years, less than a Bachelor's degree (55.4%), White (33.8%) or African American (33.1%), and living with others (64%). Forty percent of participants were married. Approximately 36.7% reported fairly adequate income, whereas 25.9% reported poorly adequate income. Most of the sample (97%) had health insurance (number of health insurance types: mean = 1.4, SD = 0.54, range = 0 to 2). Medicare (70%), employment-based insurance (24%), and Medicaid (15%) were the three most common sources and types of insurance. As multiple types of insurance coverage were common (42.1%), the number of health insurance types was chosen to represent the health insurance as one health resource in the following correlation and regression analyses. Osteoarthritis (51%) or rheumatoid arthritis (28%) were the most common arthritis types. The FCI count ranged from 2 to 14 with a mean of 3.8. The top four comorbidities were obesity (44%), diabetes (34%), visual impairment (29%), and degenerative disc disease (27%).
Sample Demographics (N = 140)
Table 2 presents the health resource and chronic disabling symptom characteristics. On average, participants reported a marginally adequate level of health literacy. Participants reported fewer people with whom to speak to at ease about private matters. Lower scores indicated smaller social networks for support; participants reported a medium level of social networks in the current study. Participants reported moderate quality of communication with their health care providers with relatively lower communication quality in items such as, “Is concerned about my understanding of my health” and “Presents me with all of the treatment options.” Symptoms with frequencies from the most to least were: sleep disturbance, pain interference, fatigue, cognitive ability, anxiety, and depression.
Sample Characteristics of Health Resources and Chronic Disabling Symptoms
Table 3 summarizes the SR-SOC strategy use frequencies of the sample. On average, items such as “Optimization: I make every effort to achieve a given goal,” “Optimization: If something matters to me, I devote myself fully and completely to it,” and “Compensation: When things don't go as well as they used to, I keep trying other ways until I can achieve the same result I used to,” were used most frequently, whereas items such as “Loss-Based Selection: When I can't do something important the way I did before, I look for a new goal,” and “Elective Selection: I concentrate all my energy to few things,” were used least frequently.
Sample Uses of Selection, Optimization, and Compensation Strategies
Correlations Between Sample Characteristics and SR-SOC Strategies
Bivariate correlations between sample characteristics and SR-SOC strategies were first analyzed and significant findings were synthesized. SR-SOC strategies associated significantly and positively with age but not with FCI. SR-SOC strategies associated significantly and positively with health care provider communication quality and marginally with number of health insurance types. Both symptom clusters associated significantly as well as negatively with SR-SOC strategies, total scores, and subscale scores. Nonetheless, correlations between SR-SOC strategies and mental symptom cluster (depressive, anxiety, and sleep symptoms) were lower than those correlations with depressive symptom scores alone.
Predicting SR-SOC Strategies
Table 4 summarizes the regression results of predictors for the total and subscale scores of SR-SOC strategies. Age, physical symptom cluster (pain, fatigue, and cognitive symptoms), and health care provider communication quality significantly predicted SR-SOC strategies (p < 0.05, adjusted R2 = 0.24). Age and physical symptom cluster (pain, fatigue, and cognitive symptoms) significantly predicted elective selection (adjusted R2 = 0.11). Physical symptom cluster and number of health insurance types significantly predicted loss-based selection after controlling for age (adjusted R2 = 0.12). Physical symptom cluster significantly predicted selection after controlling for age (adjusted R2 = 0.15). Physical symptom cluster, health care provider communication quality, and mental symptom cluster (depressive, anxiety, and sleep) significantly predicted optimization after controlling for age (adjusted R2 = 0.32). Physical and mental symptom clusters significantly predicted compensation (adjusted R2 = 0.14).
Multivariate Hierarchical (Stepwise) Regressions for Self-Regulation Strategies of Selection, Optimization, and Compensation
To the best of our knowledge, the current study is the first to describe the health resources and chronic disabling symptom experience of community-dwelling people aging with arthritis and MCCs, and the first to explore how these characteristics predicted SR-SOC strategy use after controlling for demographics and comorbidities based on the theoretical framework. The sample demographic characteristics were generally consistent with the updated 2017 Older Americans' Profile (Administration for Community Living, 2018) and national arthritis and MCCs disparities (Arthritis Foundation, 2017; Barbour et al., 2017; CDC, 2017; Centers for Medicare & Medicaid Services, 2012). Although more Hispanic participants than African American participants were expected from Texas, some Hispanic participants were excluded because of language barriers. Lack of recall of arthritis type (20%) could be explained by reasons such as lack of engagement or quality communication between patients and health care providers (Moore et al., 2016). The comorbidity frequencies were close to national statistics (Arthritis Foundation, 2017). Differences in the comorbidity statistics could be explained by the fact that the FCI could not capture all comorbidities (e.g., hypertension), but instead captured only those that most influenced functional status (Arthritis Foundation, 2019).
On average, participants reported a marginally adequate level of health literacy. This is lower than the health literacy of patients with rheumatoid arthritis in a national study (Caplan et al., 2014). A medium-sized social network was mostly reported but was still less than another study on older adults with osteoarthritis (Gignac et al., 2002). The quality of participants' communication with their health care providers was moderate. Compared with the prior study (Salt et al., 2013), worse quality was also reported in the current study with more variances. The comorbidity-associated complexities in physical function and lower information processing could explain the relatively lower health literacy levels, and social isolation. Better communication between health care providers and patients on the additional comorbidities could improve communication quality. Most of the sample (97%) had health insurance. However, existing insurance coverage did not address all patient preferences for treatment, as one participant reported that the insurance coverage limited the consideration for use of acupuncture.
Participants had more problems with pain and fatigue compared to other symptoms. Interestingly, however, sleep disturbance appeared as the most frequent symptom. This could be explained by pain-related sleep disorders or other comorbid conditions, such as diabetes. Sleep interventions are thus needed. After comparing with a prior study's baseline descriptive statistics on older adults with chronic musculoskeletal pain (Deyo et al., 2016), similar symptom frequencies but larger variances were found in the current study. Compared with the general population (HealthMeasures, 2019), participants in this study reported slightly higher depressive and anxiety symptoms but similar variances in fatigue and anxiety.
The use frequencies of SR-SOC strategies were similar between sub-scales, and from highest to lowest were optimization, compensation, loss-based selection, and elective selection. A prior study on the adaptation to disability showed that older adults with osteoarthritis used more compensation and optimization than selection strategies (Gignac et al., 2002). The decreased difference between the frequencies of strategy use could be explained by the measurement difference (qualitative descriptive counts versus subscale totals).
Physical symptom cluster, health care provider communication quality, and age were significant predictors of SR-SOC strategy use. These results show the importance of physical functioning, sufficient communication with health care providers, and age in promoting self-regulation among people living with arthritis and MCCs. Similar to a prior study on older adults aged >75 years with multimorbidity, degree of disability was found as a mediator between the number of chronic conditions and SR-SOC strategies (Yuen & Vogtle, 2016). In the current study, physical chronic disabling symptoms, including pain, fatigue, and cognitive abilities as a cluster, significantly predicted SR-SOC strategies and acted as a mediator of FCI similarly. Age was also a predictor among community-dwelling older adults with osteoarthritis, and selection and compensation were significantly predicted by the changed capacity or goals and disability in personal care (Gignac et al., 2002). Although age was not a predictor for SR-SOC strategies among those with MCCs at Time 1 baseline in a longitudinal study (Wurm et al., 2013), this could be explained by the scale difference (5-point Likert 12-item in the current study versus 4-point Likert 4-item scale). Accordingly, to initiate and accelerate coping strategies and processes, age or the accumulative self-management experience, better symptom management, and communication with health care providers serve as three key factors.
Age and physical functioning significantly predicted elective selection. The number of health insurance types, following physical symptom cluster, significantly predicted loss-based selection after controlling for age. Physical symptom cluster, health care provider communication quality, and mental symptom cluster significantly predicted optimization after controlling for age. Physical and mental functioning significantly predicted compensation. In regard to the regression results on age, physical and mental disabling symptom clusters were consistent with a prior study in which age and mobility significantly predicted optimization (Gignac et al., 2002). The current study further discovered the important role of health resources, including the number of health insurance types and health care provider communication quality in coping with loss-based selection and optimization strategies. Considering this finding, more or better individualized health insurance policies, such as including acupuncture, could provide more options for patients' self-regulation with arthritis and MCCs. Health care provider communication quality could be improved with better health care delivery design. A personal mobile health record application is a good example mentioned as a tool to improve communication with health care providers.
Implications emerged for clinical practice for people aging with arthritis and MCCs. Nurses could refer to the American Geriatrics Society (AGS) stepwise approach and decision making framework (AGS Expert Panel on the Care of Older Adults with Multimorbidity, 2012; Boyd et al., 2019) and thus provide better patient-centered and integrative self-management interventions, such as diet modifications for arthritis and MCCs. First, nurses should be aware of arthritis comorbidities and the associated social and health disparities from public reports or via FCI screening. Second, nurses should build good patient communication and be aware of the available and needed health resources. Third, with better communication, nurses could evaluate patients' symptoms, symptom trajectory, and their co-occurrence and true causes for better management. Finally, nurses may consider promoting the use of SR-SOC strategies, especially optimization and compensation, with personalized examples to help initiate and accelerate self-management.
The current study has limitations, including: (a) selection bias that limits the generalizability of the findings; (b) self-reported measures; (c) lack of diversity in language options; (d) exclusion of people with severe vision and hearing problems; (e) impact of depressive symptoms could have been underestimated as depressive symptom correlated higher to SR-SOC strategies than the mental symptom cluster; (f) missing data can influence the analysis results with different correlations and predictors; and (g) the cross-sectional correlational study design could not determine causality.
Future studies should focus on single-comorbidity subgroups, years since diagnosis or start of symptoms as predictors, personalized health insurance measures, integrative care or patient-centered communication measures, the detailed coping process and roles of SR-SOC in qualitative or mixed-method studies, and integrative and case-wise self-management interventions designed for arthritis and MCCs.
SR-SOC strategies, especially optimization, have been found to play an important role in building arthritis self-efficacy for better self-management and QoL for people aging with arthritis and MCCs. For SR-SOC strategy use, especially optimization, facilitators included age, adequate income and health resources, and most notably, quality communication with health care providers, and barriers included chronic disabling symptoms. As such, nurse researchers need to further understand the MCC–associated complexities, especially focusing on chronic disabling symptoms and communication quality with health care providers for more targeted interventions.
- Administration for Community Living. (2018). 2017 profile of older Americans. https://acl.gov/sites/default/files/Aging%20and%20Disability%20in%20America/2017OlderAmericansProfile.pdf
- Alonso-Fernández, M., López-López, A., Losada, A., González, J. L. & Wetherell, J. L. (2016). Acceptance and commitment therapy and selective optimization with compensation for institutionalized older people with chronic pain. Pain Medicine, 17(2), 264–277 doi:10.1111/pme.12885 [CrossRef] PMID:26304771
- American Geriatrics Society Expert Panel on the Care of Older Adults with Multimorbidity. (2012). Patient-centered care for older adults with multiple chronic conditions: A stepwise approach from the American Geriatrics Society. Journal of the American Geriatrics Society, 60(10), 1957–1968 doi:10.1111/j.1532-5415.2012.04187.x [CrossRef] PMID:22994844
- Arthritis Foundation. (2017). Arthritis by the numbers: Book of trusted facts & figures. http://www.arthritis.org/getmedia/e1256607-fa87-4593-aa8a-8db4f291072a/2019-abtn-final-march-2019.pdf
- Arthritis Foundation. (2019). Arthritis comorbidities. http://web.archive.org/web/20190326081852/https://www.arthritis.org/living-with-arthritis/comorbidities/
- Barbour, K. E., Helmick, C. G., Boring, M. & Brady, T. J. (2017). Vital signs: Prevalence of doctor-diagnosed arthritis and arthritis-attributable activity limitation—United States, 2013–2015. MMWR. Morbidity and Mortality Weekly Report, 66(9), 246–253 doi:10.15585/mmwr.mm6609e1 [CrossRef] PMID:28278145
- Berchick, E. R., Hood, E. & Barnett, J. C. (2018). Health insurance coverage in the United States: 2017. https://www.census.gov/content/dam/Census/library/publications/2018/demo/p60-264.pdf
- Boyd, C., Smith, C. D., Masoudi, F. A., Blaum, C. S., Dodson, J. A., Green, A. R., Kelley, A., Matlock, D., Ouellet, J., Rich, M. W., Schoenborn, N. L. & Tinetti, M. E. (2019). Decision making for older adults with multiple chronic conditions: Executive summary for the American Geriatrics Society Guiding Principles on the Care of Older Adults With Multimorbidity. Journal of the American Geriatrics Society, 67(4), 665–673 doi:10.1111/jgs.15809 [CrossRef] PMID:30663782
- Caplan, L., Wolfe, F., Michaud, K., Quinzanos, I. & Hirsh, J. M. (2014). Strong association of health literacy with functional status among rheumatoid arthritis patients: A cross-sectional study. Arthritis Care and Research, 66(4), 508–514 doi:10.1002/acr.22165 [CrossRef] PMID:24023051
- Centers for Disease Control and Prevention. (2017). Arthritis. https://www.cdc.gov/arthritis/index.htm
- Centers for Medicare & Medicaid Services. (2012). Chronic conditions among Medicare beneficiaries. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Chronic-Conditions/Downloads/2012Chartbook.pdf
- Deyo, R. A., Ramsey, Katrina, Buckley, D. I., Michaels, L., Kobus, A., Eckstrom, E., Forro, V. & Morris, C. (2016). Performance of a Patient Reported Outcomes Measurement Information System (PROMIS) short form in older adults with chronic musculoskeletal pain. Pain Medicine, 17, 314–324 doi:10.1093/pm/pnv046 [CrossRef] PMID:26814279
- Erdfelder, E., Faul, F. & Buchner, A. (1996). GPOWER: A general power analysis program. Behavior Research Methods, Instruments, & Computers, 28(1), 1–11 doi:10.3758/BF03203630 [CrossRef]
- Freund, A. M. & Baltes, P. B. (2002). Life-management strategies of selection, optimization, and compensation: Measurement by self-report and construct validity. Journal of Personality and Social Psychology, 82(4), 642–662 doi:10.1037/0022-3522.214.171.1242 [CrossRef] PMID:11999929
- Gignac, M. A. M., Cott, C. & Badley, E. M. (2002). Adaptation to disability: Applying selective optimization with compensation to the behaviors of older adults with osteoarthritis. Psychology and Aging, 17(3), 520–524 doi:10.1037/0882-79126.96.36.1990 [CrossRef] PMID:12243392
- Groll, D. L., To, T., Bombardier, C. & Wright, J. G. (2005). The development of a comorbidity index with physical function as the outcome. Journal of Clinical Epidemiology, 58(6), 595–602 doi:10.1016/j.jclinepi.2004.10.018 [CrossRef] PMID:15878473
- Han, G.-M. & Han, X.-F. (2016). Comorbid conditions are associated with healthcare utilization, medical charges and mortality of patients with rheumatoid arthritis. Clinical Rheumatology, 35(6), 1483–1492 doi:10.1007/s10067-016-3277-y [CrossRef] PMID:27106546
- Haun, J., Luther, S., Dodd, V. & Donaldson, P. (2012). Measurement variation across health literacy assessments: Implications for assessment selection in research and practice. Journal of Health Communication, 17(Suppl. 3), 141–159 doi:10.1080/10810730.2012.712615 [CrossRef] PMID:23030567
- HealthMeasures.(2017).Patient-Reported Outcomes Measurement Information System. http://www.healthmeasures.net/explore-measurement-systems/promis
- HealthMeasures. (2019). Scoring instructions. http://www.healthmeasures.net/score-and-interpret/calculate-scores/scoring-instructions
- Janke, M. C., Jones, J. J., Payne, L. L. & Son, J. S. (2012). Living with arthritis: Using self-management of valued activities to promote health. Qualitative Health Research, 22(3), 360–372 doi:10.1177/1049732311421179 [CrossRef] PMID:21890714
- Lubben, J., Blozik, E., Gillmann, G., Iliffe, S., von Renteln Kruse, W., Beck, J. C. & Stuck, A. E. (2006). Performance of an abbreviated version of the Lubben Social Network Scale among three European community-dwelling older adult populations. The Gerontologist, 46(4), 503–513 doi:10.1093/geront/46.4.503 [CrossRef] PMID:16921004
- Moore, S. M., Schiffman, R., Waldrop-Valverde, D., Redeker, N. S., McCloskey, D. J., Kim, M. T., Heitkemper, M. M., Guthrie, B. J., Dorsey, S. G., Docherty, S. L., Barton, D., Bailey, D. E. J. Jr.. , Austin, J. K. & Grady, P. (2016). Recommendations of common data elements to advance the science of self-management of chronic conditions. Journal of Nursing Scholarship, 48(5), 437–447 doi:10.1111/jnu.12233 [CrossRef] PMID:27486851
- Qin, J., Theis, K. A., Barbour, K. E., Helmick, C. G., Baker, N. A. & Brady, T. J.Centers for Disease Control and Prevention.(2015). Impact of arthritis and multiple chronic conditions on selected life domains – United States, 2013. Morbidity and Mortality Weekly Report, 64(21), 578–582 PMID:26042649
- Radner, H. (2016). Multimorbidity in rheumatic conditions. Wiener Klinische Wochenschrift, 128(21–22), 786–790 doi:10.1007/s00508-016-1090-x [CrossRef] PMID:27738754
- Salt, E., Crofford, L. J., Studts, J. L., Lightfoot, R. & Hall, L. A. (2013). Development of a quality of patient-health care provider communication scale from the perspective of patients with rheumatoid arthritis. Chronic Illness, 9(2), 103–115 doi:10.1177/1742395312455440 [CrossRef] PMID:22822176
- Tripp-Reimer, T., Williams, J. K., Gardner, S. E., Rakel, B., Herr, K., McCarthy, A. M., Hand, L. L., Gilbertson-White, S. & Cherwin, C. (2020). Integrated model of multimorbidity and symptom science. Nursing Outlook, 68(4), 430–439 doi:10.1016/j.outlook.2020.03.003 [CrossRef]
- van Onna, M. & Boonen, A. (2016). The challenging interplay between rheumatoid arthritis, ageing and comorbidities. BMC Musculoskeletal Disorders, 17, 184 doi:10.1186/s12891-016-1038-3 [CrossRef] PMID:27118031
- Vornovitsky, M. (2015). Measuring health insurance coverage with the Current Population Survey and the American Community Survey. https://www.census.gov/content/dam/Census/newsroom/press-kits/2015/20150828_healthinsur_vornovitsky.pdf
- Wurm, S., Warner, L. M., Ziegelmann, J. P., Wolff, J. K. & Schuz, B. (2013). How do negative self-perceptions of aging become a self-fulfilling prophecy?Psychology and Aging, 28(4), 1088–1097 doi:10.1037/a0032845 [CrossRef] PMID:24128074
- Xu, X., Mishra, G. D. & Jones, M. (2017). Evidence on multimorbidity from definition to intervention: An overview of systematic reviews. Ageing Research Reviews, 37, 53–68 doi:10.1016/j.arr.2017.05.003 [CrossRef] PMID:28511964
- Yuen, H. K. & Vogtle, L. K. (2016). Multi-morbidity, disability and adaptation strategies among community-dwelling adults aged 75 years and older. Disability and Health Journal, 9(4), 593–599 doi:10.1016/j.dhjo.2016.03.004 [CrossRef] PMID:27093881
- Zhang, W., Becker, H., Stuifbergen, A. K. & Brown, A. ( 2017a, February 22–25. ). Does repeated measures ANOVA with multiple imputation, or intention-to-treat yield different results? Southern Nursing Research Society 31st Annual Meeting. , Dallas, TX. .
- Zhang, W., Becker, H., Stuifbergen, A. K. & Brown, A. (2017b). Predicting health promotion and quality of life with symptom clusters and social supports among older adults with multiple sclerosis. Journal of Gerontological Nursing, 43(10), 27–36 doi:10.3928/00989134-20170406-04 [CrossRef]
- Zhang, W. & Radhakrishnan, K. (2018). Evidence on selection, optimization, and compensation strategies to optimize aging with multiple chronic conditions: A literature review. Geriatric Nursing, 39(5), 534–542 doi:10.1016/j.gerinurse.2018.02.013 [CrossRef]
- Zhang, W., Radhakrishnan, K., Becker, H., Acton, G. J. & Holahan, C. K. (2020). Self-regulation strategies as predictors: Managing multimorbidities among community-dwelling people aging with arthritis. Western Journal of Nursing Research, 42(9), 698–707 doi:10.1177/0193945919893679 [CrossRef]
Sample Demographics (N = 140)
| Female||98 (70)|
| Male||42 (30)|
|Educational level (n = 139)|
| Less than high school||22 (15.8)|
| High school graduate||39 (28.1)|
| Associate's degree||16 (11.5)|
| Bachelor's or graduate degree||62 (44.6)|
|Race/ethnicity (n = 139)|
| White||47 (33.8)|
| African American||46 (33.1)|
| Hispanic, Latino, or Spanish||18 (12.9)|
| Asian||15 (10.8)|
| Other||13 (9.4)|
| Married/significant other||56 (40)|
| Other||84 (60)|
|Income adequacya (n = 139)|
| Good||52 (37.4)|
| Fair||51 (36.7)|
| Poor||36 (25.9)|
|Mean (SD) (Range)|
|Age (years)||70.5 (9.3) (52 to 91)|
Sample Characteristics of Health Resources and Chronic Disabling Symptoms
|Characteristic||Mean (SD) (Range)||Median||n|
|Health literacy||16.2 (6.7) (6 to 20)||17||135|
|Family social network subscale||8.1 (3.6) (0 to 15)||8||139|
|Friend social network subscale||8.2 (3.6) (0 to 15)||9||136|
|Social network total||16.3 (6.7) (0 to 30)||17||136|
|Quality communication subscale||55 (14.1) (17 to 68)||61||138|
|Negative communication subscale||13.7 (2.7) (4 to 16)||14||138|
|Patient–health care provider communication quality scale total||68.6 (14.9) (33 to 84)||73.5||138|
| Pain interference total||10.9 (4.2) (4 to 20)||12||137|
| Fatigue total||10.8 (4.2) (4 to 20)||11||138|
| Distress–depression total||6.9 (3.3) (4 to 20)||6||135|
| Distress–anxiety total||7.1 (3.4) (4 to 19)||6||133|
| Sleep disturbance total||12 (3.6) (4 to 20)||12||134|
| Cognitive ability total||13.2 (4) (4 to 20)||13||137|
Sample Uses of Selection, Optimization, and Compensation Strategies
|Item||Mean (SD) (Range)||Median||n|
|ES: I concentrate all my energy to few things (e.g., health-promoting aspects of leisure).||3.2 (1) (1 to 5)||3||137|
|ES: I always focus on the one most important goal at a given time (e.g., restrict or limit activity).||3.3 (0.9) (1 to 5)||3||137|
|ES: When I think about what I want in life, I commit myself to one or two important goals (e.g., commitment to leisure or health purposes).||3.5 (1) (1 to 5)||3||137|
|LBS: When things do not go as well as before, I choose one or two important goals (e.g., focusing on essential activities).||3.5 (0.9) (1 to 5)||3||137|
|LBS: When I cannot do something important the way I did before, I look for a new goal (e.g., give up or avoid activity).||3.1 (1) (1 to 5)||3||137|
|LBS: When I cannot do something as well as I used to, I think about what exactly is important to me (e.g., restructuring participation in leisure).||3.5 (0.9) (1 to 5)||3||137|
|O: I keep working on what I have planned until I succeed (e.g., plan ahead to avoid problem).||3.7 (0.9) (2 to 5)||4||135|
|O: I make every effort to achieve a given goal (e.g., engage in exercise to optimize function).||3.8 (1) (1 to 5)||4||135|
|O: If something matters to me, I devote myself fully and completely to it (e.g., personal beliefs as motivators).||3.8 (0.9) (1 to 5)||4||135|
|C: When things do not go as well as they used to, I keep trying other ways until I can achieve the same result I used to (e.g., modify or change the way activities are performed).||3.8 (1) (1 to 5)||4||135|
|C: When something in my life is not working as well as it used to, I ask others for advice or help (e.g., seek for external aids).||3.3 (0.9) (1 to 5)||3||135|
|C: When it becomes harder for me to get the same results, I keep trying harder until I can do it as well as before (e.g., use assistive devices).||3.5 (0.9) (1 to 5)||3||135|
|Elective selection subscale||10 (2.5) (5 to 15)||10||137|
|Loss-based selection subscale||10.1 (2.3) (5 to 15)||10||137|
|Selection subscale total||20.1 (4.4) (10 to 30)||20||137|
|Optimization subscale total||11.3 (2.5) (5 to 15)||12||135|
|Compensation subscale total||10.6 (2.2) (5 to 15)||11||135|
|Selection, optimization, and compensation total||42 (8) (23 to 60)||42||135|
Multivariate Hierarchical (Stepwise) Regressions for Self-Regulation Strategies of Selection, Optimization, and Compensation
|Final Model Estimate||B||β||t||p Value|
|Selection, optimization, and compensation total (n = 119)a|
| Physical symptom cluster 2||−2.89||−0.37||−4.39||<0.001***|
| Health care provider communication quality||0.09||0.18||2.15||0.03*|
|Elective selection (n = 120)b|
| Physical symptom cluster 2||−0.64||−0.26||−2.99||<0.01**|
|Loss-based selection (n = 125)c|
| Physical symptom cluster 2||−0.76||−0.33||−3.78||<0.001***|
| Number of health insurance types||0.93||0.21||2.41||0.02*|
|Selection (n = 120)d|
| Physical symptom cluster 2||−1.43||−0.33||−3.82||<0.001***|
|Optimization (n = 118)e|
| Physical symptom cluster 2||−1.03||−0.45||−5.5||<0.001***|
| Health care provider communication quality||0.03||0.17||2.12||0.04*|
| Mental symptom cluster 1||−0.38||−0.16||−1.98||0.05*|
|Compensation (n = 114)f|
| Physical symptom cluster 2||−0.68||−0.35||−3.97||<0.001***|
| Mental symptom cluster 1||−0.38||−0.18||−2.1||0.04*|