Associations of multiple (≥5) chronic conditions among a nationally representative sample of older United States adults with self-reported pain

Introduction

The proportion of older adults in the population is increasing globally and within the United States (US), in part due to increasing life expectancy [1]. This leads to a corresponding increase in the prevalence of multiple chronic conditions among older adults [2], [3], [4]. Data from the National Health Interview Survey (NHIS) indicated that the prevalence of multiple chronic conditions (i.e., ≥1 chronic condition) among US adults was 21.8% in 2001 and had increased to 26.0% in 2010 [5], while another study also using NHIS data indicated the prevalence of multiple chronic conditions was 25.5% in 2012 [5]. Recent estimates indicate that over half (52%) of US adults have at least one chronic condition, and approximately 27% had ≥2 chronic conditions in 2018 [6]. The presence of multiple chronic conditions in an individual, particularly among older adults, is an important public health topic given that chronic conditions can lead to functional impairment in later-life, which is a significant determinant of healthcare resource use such as medical and social services [7, 8], and mortality [9].

Pain, defined by the International Association for the study of pain (IASP) as “an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage” [10], is also prevalent, ranging from 24 to 72% in previous studies [11], [12], [13]. Pain can have detrimental effects on an individual’s quality of life [10] and is often reported as an underlying disability among older adults [14]. Pain can be burdensome to manage, as individuals often have to use several strategies to manage pain [15, 16].

Various demographic characteristics such as age, gender, and race have been shown to be associated to various degrees with pain [17], however the association of a more comprehensive set of personal characteristics on pain among individuals with pain and chronic conditions is not well known. Further research to assess the association between individual characteristics and multimorbidity among older adults is therefore needed, in order to determine appropriate clinical and public health policy needs [18], [19], [20]. The objective of this study was to determine the characteristics associated with the presence of multiple (≥5) chronic conditions among community-dwelling older United States adults with self-reported pain.

Methods

Results

Figure 1 outlines the subject selection process. From the 31,880 subjects available in the 2017 MEPS dataset, 5,076 were eligible for this study (≥5 chronic conditions=1,766, <5 chronic conditions=3,310). This represented a weighted population of 57,074,842, of which 19,372,008 or 33.9% (95% confidence interval [CI]=32.2, 35.6) had ≥5 chronic conditions and 37,702,833 or 66.1% (95% CI=64.4, 67.8%) had <5 chronic conditions.

Figure 1:

Subject selection process.

Table 1 outlines the characteristics of study subjects. Approximately half (51%) the subjects were aged ≥65 years, and the majority were: female (55%), non-Hispanic (90%), white (81%), educated beyond high school (51%), unemployed (61%), had middle/high income (68%), private insurance coverage (61%), married (57%), had no IADLs (91%), no ADLs (95%), no functional limitations (61%), no work limitations (73%), little/moderate severity pain (75%), excellent/very good perceived mental health (52%), excellent/very good/good perceived physical health (73%), did not do regular exercise (58%), and did not smoke (85%). The most common census region was the south (38%). There were differences between those who had ≥5 chronic conditions and those who had <5 chronic conditions for all characteristics (p<0.05) except gender (p=0.5578), race (p=0.5296), and smoking status (p=0.2949).

Table 2 reports the weighted percentage of subjects who had each type of chronic condition. The most common chronic conditions were hypertension (87.4%, 95% CI=85.9%, 88.9%), joint pain (71.4%, 95% CI=69.7%, 73.1%), arthritis (62.1%, 95% CI=60.3%, 63.8%), and hypercholesterolemia (57.0%, 95% CI=55.3%, 58.8%).

Table 3 displays the results of the hierarchical logistic regression analyses and indicates the factors that were predictors of ≥5 chronic conditions. Results are presented for each level of the analysis (i.e., models 1–5). In the fully adjusted analysis (model 5), all of the predisposing factors were predictors of a subject having ≥5 chronic conditions; those aged ≥65 years (vs. 50–64 years), males (vs. females), non-Hispanic (vs. Hispanic ethnicity), and white (vs other race) were associated with greater odds of having ≥5 chronic conditions. Employment status was the only enabling factor associated with ≥5 chronic conditions; those who were unemployed (vs. employed) were associated with greater odds of having ≥5 chronic conditions.

Among need factors, functional and work limitations were associated with higher likelihood of having ≥5 chronic conditions, while little/moderate pain severity (vs. quite a bit/extreme pain severity), and excellent/very good and good (vs. fair/poor) perceived physical health status were associated with lower likelihood of having ≥5 chronic conditions.

None of the personal health practices or external environmental factors were predictors of ≥5 chronic conditions. The fully adjusted logistic regression model had a Wald statistic of <0.0001 and a c-statistic of 0.758.

Discussion

This study first reports the prevalence of multiple chronic conditions and then the associations of multiple chronic conditions among the increasing community-dwelling older adult US population.

Published prevalence estimates of multiple chronic conditions vary widely due to variations in definitions and measurements, population characteristics, study sample sizes, and sampling methods [26]. The current study used nationally representative data of the community-dwelling US older adult population with self-reported pain, and found that the overall prevalence of multiple (≥5) chronic conditions was 33.9% (95% CI=32.2, 35.6). That approximately one-third of the community-dwelling US older adult population with pain also have at least five other chronic conditions indicates the scale of the healthcare challenge to help these individuals manage all of their conditions, including pain, and obtain optimal health outcomes.

This study then found several factors were statistically associated with multiple chronic conditions among older community-dwelling US adults with self-reported pain, which are each addressed below.

All of the predisposing factors (age, gender, ethnicity, race) were statistically associated with multiple chronic conditions in the fully adjusted analyses. These findings add to the literature that assesses the association of such factors among older adults with pain. For instance, community-dwelling US adults aged 50–64 years with pain had lower odds of having multiple chronic conditions compared to adults who were ≥65 years. This finding is consistent with other studies reporting strong associations between age and multiple chronic conditions [27], [28], [29]. However, this phenomenon is not limited to community-dwelling older adults in the US; for example, a study from Canada reported the prevalence of multiple chronic conditions increased with age among middle-aged adults [30]. The current study found males, Whites, and non-Hispanics had greater odds of having multiple chronic conditions than females, other races, and Hispanics respectively. The associations of these variables with chronic conditions differs in the literature, for example, one previous study among US adults aged 20 years and older found that the prevalence of multiple chronic conditions was greater among those aged ≥65 years vs. those aged 45–65 and 20–44 years, greater among women than men, and greater among Whites and Blacks than those who were Hispanic or another race [31]. Furthermore, another study found that the prevalence and mortality of chronic conditions varies between genders and races in the US population, for example, there are more cases of women admitted to hospital with asthma than men, and a higher mortality rate from asthma among non-Hispanic Blacks [32]. While it is not possible to modify these predisposing factors, it can nevertheless be useful to know how they are associated with the presence of multiple chronic conditions among older adults and may assist healthcare professionals when considering the holistic care of these patients and reducing health disparities.

Employment status was the only enabling factor associated with multiple chronic conditions in the current study. Older employed community-dwelling US adults had lower odds of multiple chronic conditions than those who were unemployed. This finding supports previous studies that found a relationship between employment status and multiple chronic conditions, specifically departure from the workforce due to multiple chronic conditions [33], [34], [35]. This finding therefore supports the notion that those with multiple chronic conditions may be unable to work. The association of employment status and multiple chronic conditions may also be explained, at least partially, by resiliency, which is associated with work engagement behaviors among older workers [34].

Several need factors were associated with multiple chronic conditions. The presence of functional and work limitations was associated with greater odds of multiple chronic conditions. A similar study using 2017 MEPS data found that older adults with pain and fewer chronic conditions had a lower likelihood of reporting functional limitations than those with ≥5 chronic conditions [36]. This finding also corresponds with previous work that found individuals with more chronic conditions become functionally impaired earlier than persons with fewer chronic conditions [37]. This finding adds to existing evidence of the association between the prevalence of limitations and multiple chronic conditions among older adults with pain, and given that both limitations and multiple chronic conditions are associated with poorer health outcomes, could be a good factor to target for healthcare interventions in order to improve health outcomes.

In terms of pain severity, the current study found that community-dwelling US older adults who reported little/moderate pain had lower odds of having multiple chronic conditions than those who reported quite a bit/extreme pain. This is an unsurprising finding given that individuals with pain generally suffer from poorer health compared to those who do not have pain [38], [39], [40]. Recent studies from Europe have described how older adults with multimorbidity also have chronic persistent pain [41, 42], while another study from Canada using data from 2006 to 2016 found that each additional comorbid condition was associated with an 8% increase in the odds of reporting pain [43].

Finally, community-dwelling US older adults who reported their health as excellent/very good/good had lower odds of having multiple chronic conditions than those who described their health as fair/poor. This finding concurs with a European study that found increased number of comorbid conditions was associated with a greater likelihood of an individual reporting fair or poor health [44]. Similar findings have been reported in recent studies from Brazil [45], Canada [46], and Denmark [47], thus this study adds data for the US older adult population with pain. These aforementioned need factors may be the easiest modifiable targets for healthcare interventions to improve chronic condition management, pain management, and ultimately health outcomes in this population.

The findings from this study provide additional evidence that greater efforts are needed to help manage and improve the health of older adults with pain from both a clinical and public health perspective. For example, greater interdisciplinary coordination between healthcare professionals may help improve pain management among this population. In addition, perhaps a greater emphasis on disease prevention and health promotion efforts may help prevent or reduce the prevalence of pain among older adults. Finally, from a research perspective, further work is needed that incorporates patient populations with multiple diagnoses that can affect pain, in order to improve our understanding of the factors that influence pain, how to prevent pain, and how to manage pain in these populations.

This study had some limitations. First, multiple chronic conditions were defined as ≥5 conditions, whereas alternative definitions may have led to different results. Second, two of the conditions included are well-known pain diagnoses (arthritis and joint pain), which may have influenced the association between chronic conditions and pain. Third, this study did not account for how well the conditions were controlled (i.e., it only accounted for the presence of the condition). Fourth, data were self-reported, which may have over- or under-estimated the true prevalence of chronic conditions and pain. Finally, this was a cross sectional study that was unable to ascertain causality (i.e., only statistical associations could be identified). Future studies using prospective or longitudinal designs are warranted to assess this relationship over time.

Conclusion

In conclusion, this study provides insight into the prevalence of multiple chronic conditions among a nationally representative sample of older US adults, and identifies individual factors that are significant predictors of multiple chronic conditions in this population. The results from this study re-affirms the need for further attention to older adult’s health care, which is increasingly important given population growth among older adults nationally and globally.