Vital Signs: Prevalence of Doctor-Diagnosed Arthritis and Arthritis-Attributable Activity Limitation — United States, 2013–2015

Introduction

In the United States, doctor-diagnosed arthritis is a common and widespread chronic condition (1,2). Arthritis is a leading cause of disability (3) and is projected to affect 78.4 million adults by 2040 (4). The most common form of arthritis is osteoarthritis; other forms include, but are not limited to, rheumatoid arthritis, gout, lupus, and fibromyalgia. The annual direct medical costs attributable to arthritis are approximately $81 billion in the United States (5). About one million knee and hip joint replacements occur each year; 99% occur because of arthritis-related pain and functional limitations (6). Among adults with arthritis, 27% report severe joint pain (7); one third of adults aged ≥45 years report anxiety or depression (8); three in 10 find stooping, bending, or kneeling very difficult; approximately 20% cannot or find it very difficult to walk three blocks (approximately one quarter mile) or push/pull large objects (9). Adults with arthritis are more than twice as likely to report an injury related to a fall (10), and working-aged adults with arthritis have lower employment rates compared with adults without arthritis (5).

Arthritis is a common comorbid condition among adults with heart disease, diabetes, or obesity, and the combination of arthritis and one of these chronic conditions has been found to be associated with greater levels of physical inactivity (11). Moreover, arthritis may also hinder the ability of adults with prediabetes to engage in the physical activity recommended to prevent diabetes (12).

Many adults with arthritis are prescribed opioids (13), but safer options exist to help manage arthritis-associated pain. The CDC Guideline for Prescribing Opioids for Chronic Pain states that insufficient evidence for and serious risks associated with long-term use of opioid therapy to treat chronic pain exist, and recommends use of exercise therapy, cognitive behavioral therapy, certain interventional procedures, acetaminophen, and nonsteroidal anti-inflammatory drugs for the treatment of arthritis (14). Although medications can help, nonpharmaceutical measures help as well. For example, physical activity decreases pain and improves function, each by almost 40% (15), and self-management education interventions produce improvements in a person’s confidence and skills to manage their condition and can reduce pain, fatigue, and depression by 10% to 20% (16). However, self-management education interventions are underused by adults with arthritis; only about 11% reported ever having taken a course (17). Furthermore, approximately one in three adults with arthritis report no leisure-time physical activity (18). A health care provider’s recommendation to patients with arthritis is important, because adults with arthritis are significantly more likely to attend a self-management education program to learn to manage their condition when recommended by a provider than adults with arthritis who were not recommended (19,20). Physical activity programs can reduce yearly health care costs. For example, an analysis found that participation in EnhanceFitness,* an evidence-based physical activity intervention, reduced total health care costs by $945 per person (21), and produced substantial improvements (up to 18%) in function (e.g., muscle strength and balance) and self-reported health at follow-up at 8 months (22).

To update prevalence estimates of arthritis and arthritis-attributable activity limitations, CDC analyzed 2013–2015 data from the National Health Interview Survey (NHIS).

Methods

NHIS is an annual, nationally representative, in-person interview survey of the health status and behaviors of the noninstitutionalized civilian U.S. adult population. In each household identified, one adult is randomly selected in each family to complete the “sample adult” questionnaire.^† Sampling weights were applied to account for household nonresponse and oversampling of non-Hispanic blacks (blacks), Hispanics, and non-Hispanic Asians (Asians). Poststratification adjustments were applied by the National Center for Health Statistics using 2010 census estimates for the years 2013–2015. NHIS data from 2013 (N = 34,557), 2014 (36,697), and 2015 (33,672) were combined and weighted. Annualized unadjusted and age-adjusted prevalence estimates (standardized to the projected 2000 U.S. standard population) (23) were calculated overall and stratified by selected demographic (sex, age group, race/ethnicity, education level, and employment status) and health (body mass index category,^§ physical activity level,^¶ health status, doctor-diagnosed heart disease,** and doctor-diagnosed diabetes) characteristics. Absolute percent differences for all comparisons to the referent subgroups within each characteristic were considered statistically significant if the 95% confidence intervals of the differences of the age-adjusted estimates did not include zero. Orthogonal linear contrasts were performed to examine trends over time since 2002 in the age-adjusted prevalence of doctor-diagnosed arthritis and arthritis-attributable activity limitations among adults with arthritis.

Having doctor-diagnosed arthritis was defined as answering “yes” to the question “Have you ever been told by a doctor or other health professional that you have some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia?” That standard question, in use since 2002, was designed to incorporate key elements of the 1994 public health definition of arthritis developed by the National Arthritis Data Workgroup, which sought to capture conditions treated by a rheumatologist or considered arthritis or a rheumatic condition by health care providers. Those who responded “yes” to arthritis also were asked “Are you now limited in any way in any of your usual activities because of arthritis or joint symptoms?” Persons responding “yes” to both questions were categorized as having arthritis-attributable activity limitations.

Results

During 2013–2015, an estimated 22.7% (54.4 million; age-adjusted prevalence = 21.0%) of all U.S. adults had doctor-diagnosed arthritis. Almost half (49.6%, 22.2 million) of adults aged ≥65 years had arthritis; 7.1% (8.0 million) of young adults (aged 18–44 years) and 29.3% (24.2 million) of middle-aged adults (aged 45–64 years) had arthritis (Table 1). The majority of adults with arthritis (59.0%, 32.2 million) were aged <65 years. The age-adjusted prevalence of arthritis was significantly higher among women (23.5%) than men (18.1%), and was significantly lower among Hispanics (15.4%) and Asians (11.8%) than non-Hispanic whites (whites) (22.6%), and among adults who completed college or higher (17.9%) than adults who had less than a high school education (21.2%). The age-adjusted prevalence of arthritis was highest among adults who were unable to work (43.9%) (Table 1). The age-adjusted prevalence of arthritis was lower among adults meeting physical activity recommendations (18.1%) than adults who reported insufficient activity (23.1%) or inactivity (23.6%), and higher among adults with fair/poor health (40.5%) than adults with very good/excellent health (15.4%) (Table 1).

The unadjusted prevalences of arthritis among adults with obesity, heart disease, or diabetes were 30.6%, 49.3%, and 47.1%, respectively (Table 1) After adjustment for age, adults who had obesity compared with no obesity, had diabetes compared with no diabetes, or had heart disease compared with no heart disease were approximately 1.5, 1.7, and 1.9 times more likely to have arthritis, respectively (Figure 1).

Among adults with arthritis, an estimated 43.5% (23.7 million; age-adjusted prevalence = 41.7%) had arthritis-attributable activity limitations (Table 2). The age-adjusted prevalence of arthritis-attributable activity limitations was higher among women (43.8%) than men (38.5%); among blacks (48.6%), Hispanics (44.3%), and non-Hispanic multiracial adults (50.5%) than among whites (40.1%); among adults with less than a high school education (52.1%) than among adults with higher education (32.1%–43.6%); and among adults who were unable to work (80%) or unemployed (48.4%) than among adults who were employed (28.3%) (Table 2). The age-adjusted prevalence of arthritis-attributable activity limitations was higher among adults who were physically inactive (54.0%) than adults meeting recommendations (30.1%); among adults with fair/poor health (70.6%) than adults with very good/excellent health (23.2%); and among adults who had obesity (45.2%), heart disease (54.8%), and diabetes (52.5%) (Table 2).

The age-adjusted prevalence of arthritis-attributable activity limitations among adults with arthritis was significantly higher in 2015 (42.8%; 95% CI = 40.5–45.1) compared with 2002 (35.9%; 95% CI = 34.1–37.6), an increase of 19.2% (p-trend <0.001) (Figure 2). The age-adjusted prevalence of doctor-diagnosed arthritis did not change significantly over time (p-trend = 0.71).

Conclusions and Comments

During 2013–2015, an average of more than one in five (54.4 million) adults in the United States had doctor-diagnosed arthritis, with 43.5% (23.7 million) of adults with arthritis reporting arthritis-attributable activity limitations. The prevalence of arthritis-attributable activity limitations among adults with arthritis increased by almost 20% over time (2002–2015) independent of the aging of the U.S. population, resulting in greater pain, disability, costs, and decreased quality of life.

As found in analyses of earlier NHIS surveys (1), women and adults who were unable to work, with fair/poor health, or with obesity, heart disease, or diabetes, not only had a higher prevalence of arthritis, but also had a higher prevalence of arthritis-attributable activity limitations. The prevalence of arthritis among adults who were unemployed was similar to the prevalence among employed adults, but unemployed adults had a much higher prevalence of arthritis-attributable activity limitations, suggesting that arthritis-attributable activity limitations might play a role in their unemployment (5). Similar to past analyses, blacks and whites had a similar prevalence of arthritis, but the prevalence of arthritis-attributable activity limitations was higher among blacks. Hispanics had a much lower prevalence of arthritis, but a proportionately higher prevalence of arthritis-attributable activity limitations.

Our findings suggest that the burden of arthritis is increasing and requires more widespread use of existing, underused evidence-based interventions. Physical activity is a proven strategy for managing arthritis, with known benefits for the management of many other chronic conditions (15). Arthritis is common among adults with heart disease, diabetes, or obesity, and the combination of arthritis and one of these chronic conditions has been associated with higher levels of physical inactivity (11), suggesting that arthritis-specific barriers to physical activity (concerns about worsening pain, damaging joints, and safely exercising) might be important concerns for adults with those conditions. Improving the health of adults with arthritis, including those with these combined chronic conditions, needs to include wider dissemination and implementation of evidence-based interventions. These interventions meet the unique needs of adults with arthritis and have been found to reduce pain and improve function, mood, and confidence to manage health and quality of life (15,16).

A U.S. Department of Health and Human Services initiative addresses the effects of multiple chronic conditions,^†† which now affect one in four adults and are becoming increasingly common as the population ages. A previous study on multiple chronic conditions among adults aged ≥25 years found that arthritis is frequently present among the most common combinations of two and three conditions (24). CDC is promoting greater coordination among chronic disease programs within state health departments to address these chronic disease comorbidity concerns.^§§ Active promotion of evidence-based self-management education and physical activity interventions is appropriate for various chronic conditions. The self-management education^¶¶ and physical activity interventions*** that CDC recommends for adults with arthritis are examples of proven, low-cost, community interventions that can benefit adults with arthritis, physical limitations, and other chronic conditions.

The findings in this report are subject to at least six limitations. First, doctor-diagnosed arthritis was self-reported and not confirmed by a health care professional; however, this case definition was validated for public health surveillance (25). Second, because NHIS is a cross-sectional survey, a causal relationship between risk factors (i.e., obesity or physical activity) and arthritis and arthritis-attributable activity limitations could not be established, although strong evidence exists that indicates obesity is associated with an increased risk for incident knee osteoarthritis (a common form of arthritis) (26). Third, social desirability bias might play a role in some self-reported characteristics, with underreporting of weight and overreporting of height and leisure-time physical activity. Fourth, from 2013 to 2015, the NHIS response rates were 61.2%, 58.9%, 55.2%, respectively, indicating potential nonresponse bias, although survey weights were applied to address this bias and improve external validity (27). Fifth, if multivariate analyses were to be performed, certain observed group differences, such as those related to race/ethnicity and arthritis-attributable activity limitations, might have been explained by differences in prevalence of comorbid conditions or other factors (e.g., health care access). Finally, NHIS does not survey persons in long-term care institutions (e.g., nursing homes for the elderly and hospitals for the chronically ill or physically or intellectually disabled); therefore, this analysis likely underestimates the prevalence of arthritis and arthritis-attributable activity limitations.

Arthritis is a large and growing clinical and public health problem. In 2017, CDC is funding arthritis programs in 12 states to disseminate arthritis-appropriate evidence-based physical activity and self-management education interventions in their communities.^††† Given the high prevalence of arthritis and the increase in arthritis-attributable activity limitations in the United States, health care providers and public health practitioners can address arthritis and other chronic conditions by prioritizing proven, nonpharmaceutical interventions, such as self-management education and appropriate physical activity, as effective ways to improve health outcomes, especially for groups with the highest prevalence of arthritis and arthritis-attributable activity limitations.