The number of adults over the age of 65 years in the United States will reach 70 million by the year 2030 (Centers for Disease Control and Prevention and The Merck Company Foundation. The State of Aging and Health in America 2007, 2007 Accessed March 30, 2016; Mazzeo, [17]). More than 60 % of these older adults will have more than one chronic condition [27]. Physical activity is one way to maintain physical function and decrease morbidity and disability; however, less than 20 % engage in enough physical activity (US Department of Health and Human Services), and less than half receive preventive services [27]. Older adults with arthritis are particularly prone to disability because of chronic pain, co-morbid chronic conditions, and a high rate of obesity. Thus, they represent a high-risk target population for health promotion.

Empowering older adults to engage in physical activity at the recommended levels and to use preventive services for primary and secondary prevention of chronic disease is an important public health goal. Participation in self-management programs may be one venue to increase knowledge and self-efficacy to promote behavior change and prevention of chronic disease. Delivering these programs where people work and live in the community could increase access and participation and ultimately impact outcomes.

In this research we used a collaborative approach to develop, implement, and evaluate a program to address multiple risk factors for disability in older adults with arthritis. We enhanced an existing exercise program targeting older adults with arthritis (Arthritis Foundation Exercise Program, AFEP) by adding a healthy aging behavioral activation program (“10 Keys”™ to Healthy Aging) [19]. We sought to determine if the enhanced program was superior to the AFEP alone in improving arthritis-specific outcomes (function, pain, and stiffness) and increasing preventive service use.

In total, 670 program participants expressed initial interest in the research study, and 462 consented to participate. Of 670 screened, 56 did not meet eligibility criteria and 152 declined to participate. By the 6-month follow-up visit, 416 participants were evaluated, 203 in the enhanced program and 213 in AFEP alone (Fig. 2).

The study population had a mean age of 73 years and 88 % were women. Eighty percent self-identified as white, 18 % black, and the remainder as other race groups. Two-thirds of the population had some education beyond high school and half had an income above $25,000. There were no significant differences between the groups in sociodemographic characteristics, prevalence of common chronic health conditions, body mass index (BMI), or health behaviors (Table 1).

Table 2 shows the results of the primary outcomes of physical performance and arthritis outcomes assessed at baseline, post-program (10 weeks), and at 6 months. Using linear mixed models controlling for baseline values, there were no significant differences in SPPB or WOMAC between the two study arms at 6 months. As shown in Table 3, both groups experienced similar and significant improvements in physical performance and arthritis outcomes. Physical performance improved by 0.3 units in the AFEP/10 Keys group and 0.5 units in AFEP alone; WOMAC scores improved by 3.0 and 3.9 points in the two groups, respectively.

Table 4 shows the proportion of people achieving the goals of the “10 Keys”™ to Healthy Aging over the 6-month follow-up. For all components, improvements in the “10 Keys”™ were not significantly different between the two groups at 6 months controlling for baseline values. At 6 months, more participants had controlled hypertension in both groups, increasing from 60.1 to 76.7 % in the AFEP/“10 Keys”™ group and from 76.5 to 84.9 % in the AFEP alone group. Diabetes control was also improved, from 15.0 to 34.9 and 15.5 to 34.1 %, respectively. When examining subgroups of individuals who had the greatest opportunity to improve, such as those who had uncontrolled hypertension, diabetes, were smoking, had not had cancer screenings or immunizations, etc., improvements were seen in both groups, ranging from about 10 to 50 %. Again, however, there were no differences in these improvements between those who had participated in the combined AFEP “10 Keys”™ program vs. those who participated in the AFEP exercise program alone.

Finally, we explored interactions between sociodemographic factors and the treatment groups in order to determine whether those of lower education or income might have been more likely to benefit and found that the group differences did not vary by education or income status.

These two interventions targeted older adults with arthritis who had limitations in physical function, comorbid conditions, and arthritis-specific symptoms including pain and stiffness. Participants in each intervention group showed similar improvements in these parameters over 6 months. In spite of our efforts to promote better function, reduce symptoms, and promote preventive health behaviors with the enhanced program, improvements were not significantly different. Both groups had improvements in physical performance and arthritis outcomes. This was not unexpected given that both groups had an exercise intervention previously documented to be effective for individuals with arthritis [4]. The magnitude of these improvements was similar to that seen in our pilot study [25] and are considered to be clinically meaningful [15]. In addition to improved physical performance and arthritis outcomes, both groups had improvements in control of hypertension and diabetes. However, the rates of control and improvement in cholesterol, vaccine rates, and bone density screening were less than those previously seen in our report of the benefits of the “10 Keys”™ to Healthy Aging program [19]. Thus, it would appear that the exercise program itself was effective for hypertension and diabetes, as has been reported in other studies of exercise benefits [9, 10].

These benefits were achieved in a community setting with an intervention delivered by community health workers, making the programs more translatable than those requiring health professionals [24]. Community health workers have been shown to improve health outcomes in other groups such as children with asthma and patients with human immunodeficiency virus [20] and for prevention as well as treatment [14]. Given the increase in the aging population, needs for better prevention and limitations in health care resources, community health workers could make important contributions. The benefits achieved here provide a realistic view of the benefits of physical activity in the community, demonstrating a meaningful improvement in physical performance, arthritis outcomes, hypertension, and diabetes. The high level of community engagement in this model can support sustainability [26].

Several potential aspects of this study are important to consider in the planning of prevention programs for older adults. First, our intermediate outcomes did not detect differential improvements in the short 6-month time frame of the evaluation, but further improvements might have occurred with longer follow-up. Potential downstream effects in the prevention of major disabling conditions such as myocardial infarction, stroke, and hip fracture are possible. Second, we may not have had adequate reach to individuals who were less likely to participate in community-based research programs, and these individuals may be those at greatest need for improvement in physical performance, arthritis outcomes, and/or preventive health behaviors. Similar to other programs [19, 21, 23], participants tended to have higher levels of education and have better compliance with preventive behaviors than the general population. The high rates of blood pressure control, glucose regulation, non-smoking and many of the other prevention targets suggest that participation in the research assessments was biased towards healthier persons.

It is possible that our efforts and time dedicated to the added prevention messages were insufficient to achieve behavior change. The time devoted to the “10 Keys”™ was limited compared to the original program [19]. The original “10 Keys”™ program included 3.5 h per person of individual in-person and telephone contacts to actively address problem solving to achieve written goals [19], whereas individual follow-up time in this program was not provided. By design, this aspect of the program was more limited in the AFEP/“10 Keys”™ program, with only group-based follow-up. The group exercise component appears to have been the key ingredient of the program. The community health workers who delivered the programs were not university research staff as in our previous report [19], but we believe that it is the individual follow-up time that would be more critical for behavior change.

Finally, we think it is important to point out that our programs did not address the rather high average BMI in the participants. We are currently testing a weight loss program to determine whether this would result in further improvements in physical performance, and cardiometabolic risk factors [10, 18, 22, 31, 32].

Because older adults are at risk for multiple chronic disabling conditions, it is imperative that programs consider the problem of multimorbidity. Community health workers can be trained to successfully deliver prevention interventions to older adults in community settings. Based on these results, we believe that next steps should include greater efforts to reach individuals in need of intervention with greater emphasis on goal setting and follow-up.