2.

HOURS

Continuing Education

By Todd M. Ruppar, PhD, RN, GCNS-BC, and Vicki S. Conn, PhD, RN, FAAN

OVERVIEW: The importance of promoting physi-

cal activity in adults with chronic illness cannot be overstated. Although numerous studies have evaluated interventions developed to increase physical activity in this population, it can be difficult to interpret the results without the benefit of a statistical analysis across studies. A 2008 meta-analysis synthesized the findings of 163 reports on such studies. This article discusses the implications of those findings, describing the strategies and practices commonly used to promote physical activity in chronic illness and identifying those that are most effective.

Interventions to Promote

in Chronically Ill Adults

Practice implications of clinical studies.
30
AJN July 2010

Physical Activity
Vol. 110, No. 7

ajnonline.com

he health benefits of physical activity in chronic illness are well documented. Phys ical activity has been shown to retard the atherosclerotic process, thereby reducing the risk of subsequent cardiac events in cardiovascu lar disease1, 2; increase physical function in arthritis3, 4; improve glycemic control in type 2 diabetes5, 6; and enhance the quality of life in patients with chronic illness.7 Because nurses perform the vast majority of patient education and health promotion interven tions, they need to know which strategies and prac tices are most effective in increasing physical activity in chronic illness. Many research studies have
ajn@wolterskluwer.com

Photo Dennis MacDonald / Alamy

evaluated attempts to increase physical activity in chronically ill adults, but without the benefit of a sta tistical analysis across studies it can be difficult to detect patterns and interpret results. Meta-analysis is an established means of quanti tatively synthesizing findings across many research studies. The inclusion of moderator analysis in such meta-analyses, when possible, provides statistical evidence of the effectiveness of specific intervention characteristics. (A moderator is a variable that alters the relationship between other variables.) A 2008 meta-analysis, conducted by one of us (VSC) with col leagues, summarized the findings of 163 reports on
AJN July 2010

Vol. 110, No. 7

31

213 independent tests of interventions used to pro mote physical activity among a total of 22,557 adults with a varied assortment of chronic illnesses.8 It took into account variables such as the number of behav iors targeted by the interventions (physical activity only or multiple health behaviors); whether the inter ventions used behavioral strategies (designed to pro duce a direct change in behavior), cognitive strategies (that seek to change knowledge, beliefs, or attitudes), or both; as well as numerous specific practices, in cluding self-monitoring, supervised exercise, tailoring, contracting, and several others (see Table 1).8 Findings showed that several interventions are ef fective in promoting physical activity among diverse patient populations. Sex, age, socioeconomic status, and ethnicity appeared not to influence effectiveness. On average, intervention increased physical activity the equivalent of 945 steps per day, or 48 minutes of physical activity per week, per patientan amount thats likely to be clinically important for sedentary, chronically ill adults. Among the numerous inter vention attributes and practices, however, the effect on physical activity varied considerably. This article addresses the implications of these meta-analysis findings in order to provide nurses with reliable evidence supporting or negating the purported ben efits of common strategies and practices used to pro mote physical activity among chronically ill adults. We report below which strategies and practices were found to be clearly supported by the data, which were modestly supported, and which were not sup ported.

CLEARLY SUPPORTED The 2008 meta-analysis showed interventions to be most effective in promoting physical activity among chronically ill adults when they targeted physical activity exclusively. used behavioral (as opposed to cognitive) strat egies. encouraged self-monitoring. A single target: physical activity. Health promo tion interventions often target multiple health behav iors. For example, clinicians caring for people with diabetes frequently attempt to modify behaviors re lated to diet and medication adherence as well as phys ical activity. Although multiple behavior changes may be required to manage some chronic illnesses, mod ifying physical activity alone may have dramatic ef fects. The meta-analysis found that interventions that were solely focused on improving physical activity levels were more effective in achieving that goal than those that aimed to improve multiple health behav iors simultaneously.8 Whereas a study of patients with type 1 diabetes demonstrated the effectiveness of a consultation inter vention that exclusively targeted exercise behavior,9 interventions seeking to modify multiple behaviors, including physical activity, were less successful. For example, a program developed to promote adherence to diabetes self-management among college students with type 1 diabeteswhich addressed behaviors sur rounding stress, nutrition, medications, and alcohol, as well as exercisehad no significant effect on exer cise behavior.10 Similarly, a three-month study of a

Table 1. Intervention Attributes or Practices that Promote Physical Activity8

Strength of Evidence Strong Attribute or Practice
Single target: physical activity only Behavioral approaches Self-monitoring

Definition
Interventions designed to modify only behavior related to physical activity, rather than multiple health behaviors. Interventions containing at least one behavioral strategy, designed to produce a direct change in behavior related to physical activity. Interventions including practices such as keeping an activity diary, tracking activity in a calendar, or recording activity on a Web site. Exercise overseen by a member of the research team or a health care provider. Adapting the intervention to meet the needs or situation of the patient. An agreement between patient and provider defining the level and duration of physical activity the patient will perform. Participants receive written instruction for the mode, duration, frequency, intensity, and progression of their physical activity. Patients level of physical fitness is evaluated before any physical activity program is initiated. Interventions employ prompts that remind participants to exercise. Recommendations are for moderate- or high-intensity (as opposed to low-intensity) physical activity.
ajnonline.com

Moderate

Supervised exercise Tailoring Contracting Exercise prescription Fitness testing Stimuli and cues Moderate- or high-intensity recommendations

32

AJN July 2010

Vol. 110, No. 7

community-based diabetes education program that addressed multiple self-care behaviors (including ex ercise behavior) found no significant differences in selfcare between participants who attended at least half of the weekly sessions, those who attended fewer than half of the weekly sessions, and a control group.11 We can only speculate about the reasons for these findings, but it may be that focused interventions lead participants to place a greater importance on physical activity and interventions aimed at multiple behaviors dilute the intervention effect. Additionally, by attempt ing only to increase physical activity, interventionists may be able to employ more strategies specific to that aim. It may also be easier for patients to change one behavior at a time, rather than to divide their atten tion among multiple behavior changes, and physical activity may be easier to change than some other be haviors.

have important implications for health care pro viders, who typically attempt to encourage phys ical activity in chronically ill patients through education.8 Its possible that people already know why they should exercise (the typical emphasis of cognitive strategies) but require behavioral strategies to achieve and main tain higher levels of physical activity. Such findings provide evidence for replacing attempts to provide a scientific rationale for the benefits of physical activ ity with behavioral strategies to encourage physical activity. Effective interventions using behavioral strategies to promote physical activity among women after car diac surgery were found to include self-reinforcement through keeping logs, specific goal setting, and nurse feedback.12 Specific goal setting and nurse feedback were also found to be effective in promoting physical

Interventions incorporating any behavioral strategy were more effective than those that included none.
A 2007 meta-analysis of health behavior change interventions tested in adults with type 2 diabetes found that interventions focused exclusively on phys ical activity not only improved metabolic control but achieved an effect that was twice as large as interven tions that attempted to change multiple behaviors.6 Such findings suggest that interventions sequentially targeting different health behaviors may be more ef fective than those that simultaneously target several health behaviors. Behavioral approaches. The 2008 meta-analysis of interventions used to promote physical activity in chronic illness investigated both behavioral and cog nitive approaches. Behavioral practices included in corporating consequences (rewards for performing more physical activity), patient contracts, feedback (in formation from a clinician about the amount of phys ical activity performed), goal setting, self-monitoring, and stimuli and cues. Cognitive approaches included barriers management (teaching patients how to man age obstacles or deterrents to increasing physical ac tivity), decisional balance (working with patients to chart the benefits and costs of increasing physical ac tivity), motivational counseling, problem solving, and ventions incor social cognitive interventions.8 Inter porating any behavioral strategy were found to be more effective than those that included none, and inter ventions employing only behavioral strategies were more effective than those with both behavioral and cognitive components findings that strongly favor behavioral interventions. highlight the importance of using action-oriented approaches to improve physical activity.
ajn@wolterskluwer.com

activity in patients with coronary heart disease,13 and an intervention including individualized graphic (as opposed to verbal) feedback on exercise and goal set ting was effective in promoting physical activity among patients with heart failure.14 In contrast, cognitive interventions focused on dis cussing the advantages of healthful behaviors and situations in which healthful behaviors would be dif ficult to achieve werent found to be effective in pro moting physical activity in the 2008 meta-analysis involving chronically ill adults.8 Likewise, a health ed ucation and counseling intervention in which cli nicians explained the health consequences of inade quate physical activity was ineffective over the long term (at one year) in increasing physical activity among adults in cardiac rehabilitation.15 Self-monitoring is a behavioral strategy that re quires the participant to record her or his daily or weekly physical activity over time, perhaps in an exer cise diary, on a tracking calendar, or on a Web site. For example, in a study of patients with diabetes, subjects used log books to record behavioral goals andphys monitoring ical activity.16 Another approach to self- would be to provide patients with pedometers to re cord their daily steps, a strategy that has been shown to motivate patients with type 2 diabetes to increase their physical activity.17 The 2008 meta-analysis found that strategies in volving self-monitoring significantly improved the ef fectiveness of interventions aimed at promoting phys icalactivity, regardless of other strategies employed.8 Self-monitoring increases participants awareness of their physical activity level and provides them with
AJN July 2010

Vol. 110, No. 7

33

an activity history they can improve. These findings are consistent with a 2002 meta-analysis of interven tions used to increase physical activity among aging adults, which found self-monitoring to be a particu larly important strategy with significant effects.18 MODESTLY SUPPORTED The 2008 meta-analysis of physical activity interven tions modestly supported several specific practices, including supervised exercise, tailoring, contracting, exercise prescription, fitness testing, stimuli and cues, and intensity recommendations.8 Supervised exercise. In some studies, interventions included exercise sessions supervised by a member of the research team or a health care provider (in the case of cardiac rehabilitation) and follow-up to determine whether patients continued with the physical activ ity after the supervised sessions. In one study, adults with osteoporosis who had participated in super vised group exercise sessions had significantly higher physical activity scores compared with control sub jects, even after six months.19 Similarly, patients with multiple sclerosis demonstrated a 17% increase in sport-related activity following a supervised exercise intervention, whereas control subjects with multiple sclerosis remained sedentary.20 Supervised exercise may provide patients an op portunity to receive guidance on exercise techniques, which may improve their exercise experience and thus
34
AJN July 2010

their exercise behavior. Adults with chronic illnesses frequently have safety concerns about exercise, which supervision may alleviate. Supervised exercise also provides a successful exercise experience, which may have a reinforcing effect. The 2002 meta-analysis found supervised exercise to be effective in promoting physical activity in aging adults,18 but the 2008 meta-analysis produced incon clusive findings regarding its effectiveness in chronic illness.8 Because supervised exercise requires consid erable financial resources, it may be prudent to deter mine which populations stand to derive the greatest benefit from its use before employing the practice. Tailoring describes interventions that are system atically adjusted to meet an individuals needs or sit uation. For example, interventions might be tailored to age, sex, or a specific health condition, as well as toa previous physical activity behavior. Some tailored interventions provide messages matched to patients perceived exercise barriers or to their desired benefits. In one study, computerized algorithms were used suc cessfully to tailor home communication interventions to the needs of patients with heart failure, based on in formation the patients provided about themselves.21 Alower-tech approach, using self-report assessments conducted at baseline, three months, and one year, has been used to tailor self-management interventions to adults with diabetes.22 The meta-analysis of physical activity interventions found a trend toward tailored
ajnonline.com

Vol. 110, No. 7

Photo Jeffrey Greenberg / The Image Works

in terventions being more effective than nontailored in terventions, but few interventions met the criteria for categorization as tailoring that the interventions features were systematically modified on the basis of an individual patients characteristics, to make the in tervention more effective for that patient.8 Given the meta-analysis findings and the possibilities for com plex tailoring with computerized algorithms, tailoring is certain to be an important area for future research. Contracting is a strategy in which the participant makes an agreement with the interventionist to per form physical activity at a specified level and for a particular duration. Its been used by patients under going cardiac rehabilitation23 and by chronically ill older adults.24 Contracting defines physical activity expectations for the patient and, in doing so, may provide motivation because patients want to com plete their contractual obligations. Contracting also helps patients take a complex behavior and break it into manageable units. Whereas the idea of exercis ing routinely for many years may be inconceivable to a chronically ill adult, walking at a moderate pace for 30 minutes on Monday, Wednesday, and Friday of thenext week may seem feasible. Success in meeting previous contracts may stimulate future efforts.

testing has been used successfully as a component of interventions in adults with multiple sclerosis20 and older adults with chronic illness.26 Stimuli and cues were components of many phys ical activity interventions that produced modest ef fect sizes in the 2008 meta-analysis involving chron ically ill adults.8 Such stimuli may take the form of calen dars, automated phone calls, beepers, or other prompts that fit the patients circumstances. For in stance, placing walking shoes by the bedroom door may act as a reminder to take an early morning walk. Friends may call each other several times a week to remind each other to exercise. Family involvement has been used as a stimulus in programs aimed at in creasing physical activity in adults with diabetes.27, 28 In one such study, spousal participation in an exer cise and diet program effectively prompted adher encein female (but not male) obese adults with type 2 diabetes.28 Moderate- or high-intensity recommendations. Al though physical activity recommendations must con sider patients underlying chronic illness and health status, interventions that recommended moderateorhigh-intensity (as opposed to low-intensity) physi cal activity trended toward greater effectiveness in the

Physical activity is often as necessary for health maintenance and improvement as medication.
Exercise prescription provides patients with clear (often written) expectations and instructions for per forming physical activity thats appropriate for their level of fitness and health. Exercise prescriptions that detail intensity, duration, and frequency have been shown to be effective in increasing physical activity among women with osteoporosis25 and adults with multiple sclerosis.20 Exercise prescriptions may help chron ically ill patients realize that physical activity is often as necessary for health maintenance and im provement as medication. Additionally, the fact that instructions are individualized and appropriate for the patient increases the chances that the intervention will be successful. Fitness testing describes a structured evaluation of the patients physical fitness performed prior to the ini tiation of a physical activity program. Fitness testing may work in conjunction with other physical activity interventions, such as tailoring, exercise prescription, contracting, or self-monitoring, as a way of providing a baseline from which the patient strives to improve. Fitness testing can prevent injury and improve an ex ercise programs chances of success by helping to ensure that it doesnt exceed the patients ability. In addition, some patients are motivated by receiving objective information about their fitness level. Fitness
ajn@wolterskluwer.com

2008 meta-analysis involving chronically ill adults.8 This finding echoes those of earlier meta- analyses of physical activity interventions in older adults.18 Its possible that patients misinterpret recommendations for low-intensity physical activity, believing that they represent no meaningful change from their current phy sical activity, or that they experience greater real or perceived benefit from more intense physical ac tivity. A study of older adults with knee osteoarthritis found that adherence was greater (84%) among those participating in a high-intensity, progressive strength tr aining program than among those parti cipating in a nutrition education program (65%).29 Similarly, an in tensive, water-based exercise program increased phys ical activity among adults with rheumatoid arthritis.30 NOT SUPPORTED The 2008 meta-analysis indicated that some compo nents of physical activity interventions may have no effect or, possibly, a negative effect on physical activ ity in chronically ill adults.8 Face-to-face vs. mediated intervention. The metaanalysis found that face-to-face interventions held noadvantage over those mediated by telephone, mail, or mass media.8 This finding has important financial implications because face-to-face interventions tend
AJN July 2010

Vol. 110, No. 7

35

to require far greater financial resources than those using mediated delivery. An interventions social context. Physical activity interventions that can be conducted either in groups or in more costly individual settings had no less effect when conducted in group settings. Walking (as compared with other physical activity). Whether interventions recommended walking or other forms of physical activity didnt appear to influ ence effectiveness. Purely cognitive strategies. Interventions consist ing entirely of cognitive strategies were considerably less effective than other interventions. The cognitive strategies of barriers management and problem solv ing appeared to have no effect on an interventions success in promoting physical activity. Purported theoretical foundation. Although many theories have been suggested as the basis for physical activity interventions, most primary studies dont report a theory. Among those interventions claiming a theo retical foundation, the theories are often inadequately implemented. For example, studies claiming to use the transtheoretical model may not match materials to a patients stage of change or appropriately match the processes of change with each stage.31, 32 The 2008 metaanalysis found that physical activity outcomes

were no better in the studies that claimed a basis in social cognitive theory, and some physical activity out comes were better in studies that made no such claim. Given the limitations of theory implementation as described in the literature, it would be premature to draw conclusions from this finding at present. STUDY LIMITATIONS The metaanalysis on which this article is based was limited by the number and range of primary inter vention trials reported in the literature.8 Moderator analyses couldnt be conducted on some intervention characteristics because they were reported too infre quently. This article addresses only the effectiveness of the interventions in promoting physical activity; health outcomes in specific chronic illnesses are re ported elsewhere.1, 3, 6, 7 THE TAKEAWAY The 2008 metaanalysis found that its possible to increase physical activity in adults with chronic ill nesses.8 Findings document that interventions focused on physical activity alone are most effective. Chang ing health behaviors in chronically ill patients is a very complex undertaking, and many conditions require patients to change several health behaviors. When

Looking AHEAD for Health

A successful weight loss program in Rhode Island.
onnie Newman was diagnosed with type 2 diabetes more than 10 years ago, just as a team of medical professionals at Brown University launched the Look AHEAD (Action for Health in Diabetes) study of 5,145 men and women na tionwide to assess the longterm effects of weight loss on people with the chronic illness. Researchers have been monitoring the participants, who were enrolled when they were between the ages of 45 and 74 and were randomly assigned to a program of intensive lifestyle intervention (focusing on weight reduction through moderately intense ac tivity) or to a control group of diabetes educa tion and support. I was one of the first people to respond to the ad looking for people for the study, said New man, 63, of Rhode Island. Shed always had a weight problem, but through the program of in tensive lifestyle change, she lost about 20 pounds and has kept most of it off for a decade by reading every food label, giving up soda, controlling the amounts she eats, and exercising regularly. We all agreed that without this support wed never have been able to do what weve done, she said. My diabetes hasnt progressed as quickly as it could
36
AJN July 2010

have. I feel 100% better than when I started. If it werent for this study, Id either be deador a very sick person. Rena Wing, professor of psychiatry and hu man behavior at Brown and codirector of Look AHEAD, said that during the first year of the study (which is planned to end in 2012), those in the in tensive program lost 8.6% of their starting body weightabout 19 lbs.compared with 1.5 lbs. among those in the less intensive program. Exer cise was the strongest reason for weight loss, fol lowed by program attendance and meal replace ments, according to analysis of the first year. The people who did the best were those who exercised the most and came to meetings, said Wing. The intensive program greatly improved fitness levels, blood pressure, and lipid levels. Newman attends meetings at Miriam Hospi tal in Providence, where Wing is based. Im a mother and a grandmother, Newman said, and this study is for the next generation, who will be prone to type 2 diabetes because of a lack of activity and proper diet. For more information on the Looking AHEAD study, go to www. lookaheadtrial.org.Patrice OShaughnessy
ajnonline.com

Vol. 110, No. 7

boosting physical activity is a desired goal, however, the evidence indicates that outcomes will be better if behavior changes are targeted sequentially. Mounting evidence suggests that behavioral strat egies are more effective than cognitive approaches in improving physical activity.8, 18, 33 Effective interven tions include such behavioral components as self- monitoring, consequences, contracting, feedback, goal setting, and stimuli and cues. Self-monitoring was found to be particularly effective and can be recom mended as part of any physical activity intervention program for adults with chronic illness. Meta-analysis is a powerful means of synthesizing research and determining an interventions effective ness. The results of the meta-analysis on which this article is based indicate that certain intervention strat egies are effective in improving physical activity in adults with chronic illnesses. Further evaluation is needed to determine the effectiveness of other physical activity interventions and how to best implement ef fective interventions in the clinical practice setting. t For 33 additional continuing nursing education articles on research topics, go to nursingcenter. com/ce.
Todd M. Ruppar is an assistant professor and Vicki S. Conn is the Potter-Brinton Professor and associate dean for research at the Sinclair School of Nursing, University of MissouriColumbia. Contact author: Vicki S. Conn, conn@missouri.edu. The re search presented herein was supported by funding from the National In stitutes of Health (R01NR07870). The authors of this article have no significant ties, financial or otherwise, to any company that might have an interest in the publication of this educational activity.

REFERENCES
1. Conn VS, et al. Meta-analysis of interventions to increase physical activity among cardiac subjects. Int J Cardiol 2009; 133(3):307-20. 2. Ornish D, et al. Can lifestyle changes reverse coronary heart disease? The Lifestyle Heart Trial. Lancet 1990;336(8708): 129-33. 3. Conn VS, et al. Physical activity interventions among adults with arthritis: meta-analysis of outcomes. Semin Arthritis Rheum 2008;37(5):307-16. 4. Feinglass J, et al. Effect of physical activity on functional status among older middle-age adults with arthritis. Arthritis Rheum 2005;53(6):879-85. 5. Boule NG, et al. Effects of exercise on glycemic control and body mass in type 2 diabetes mellitus: a meta-analysis of controlled clinical trials. JAMA 2001;286(10):1218-27. 6. Conn VS, et al. Metabolic effects of interventions to increase exercise in adults with type 2 diabetes. Diabetologia 2007; 50(5):913-21. 7. Conn VS, et al. Meta-analysis of quality-of-life outcomes from physical activity interventions. Nurs Res 2009;58(3):175-83. 8. Conn VS, et al. Meta-analysis of patient education inter ventions to increase physical activity among chronically ill adults. Patient Educ Couns 2008;70(2):157-72. 9. Hasler TD, et al. Exercise consultation and physical activity in patients with type 1 diabetes. Practical diabetes international 2000;17(2):44-8. 10. Wdowik MJ, et al. Development and evaluation of an inter vention program: Control on Campus. Diabetes Educ 2000;26(1):95-104.
ajn@wolterskluwer.com

11. Irvine AA, Mitchell CM. Impact of community-based dia betes education on program attenders and nonattenders. Diabetes Educ 1992;18(1):29-33. 12. Allen JK. Coronary risk factor modification in women after coronary artery bypass surgery. Nurs Res 1996;45(5):260-5. 13. Campbell NC, et al. Secondary prevention in coronary heart disease: a randomised trial of nurse led clinics in primary care. Heart 1998;80(5):447-52. 14. Duncan K, Pozehl B. Effects of an exercise adherence inter vention on outcomes in patients with heart failure. Rehabil Nurs 2003;28(4):117-22. 15. van Elderen-van Kemenade T, et al. Effects of a health edu cation programme with telephone follow-up during cardiac rehabilitation. Br J Clin Psychol 1994;33(Pt 3):367-78. 16. Samaras K, et al. Will older sedentary people with non- insulin- dependent diabetes mellitus start exercising? A health promotion model. Diabetes Res Clin Pract 1997;37(2):121-8. 17. Tudor-Locke C, et al. Controlled outcome evaluation of the First Step Program: a daily physical activity intervention for individuals with type II diabetes. Int J Obes Relat Metab Disord 2004;28(1):113-9. 18. Conn VS, et al. Interventions to increase physical activ ity among aging adults: a meta-analysis. Ann Behav Med 2002;24(3):190-200. 19. Dilsen G, et al. The role of physical exercise in prevention and management of osteoporosis. Clin Rheumatol 1989; 8 Suppl 2:70-5. 20. Mostert S, Kesselring J. Effects of a short-term exercise training program on aerobic fitness, fatigue, health percep tion and activity level of subjects with multiple sclerosis. Mult Scler 2002;8(2):161-8. 21. Barnason S, et al. Impact of a home communication inter vention for coronary artery bypass graft patients with ischemic heart failure on self-efficacy, coronary disease risk factor modification, and functioning. Heart Lung 2003; 32(3):147-58. 22. Clark M, et al. Effects of a tailored lifestyle self-management intervention in patients with type 2 diabetes. Br J Health Psychol 2004;9(Pt 3):365-79. 23. Moore SM, Charvat JM. Using the CHANGE intervention to enhance long-term exercise. Nurs Clin North Am 2002; 37(2):273-83, vi-vii. 24. Haber D, Lacy MG. Evaluation of a socio-behavioral intervention for changing health behaviors of older adults. Behavior, health, and aging 1993;3(2):73-85. 25. Iwamoto J, et al. Effect of exercise training and detraining on bone mineral density in postmenopausal women with osteoporosis. J Orthop Sci 2001;6(2):128-32. 26. Rejeski WJ, et al. Older adults with chronic disease: benefits of group-mediated counseling in the promotion of physically active lifestyles. Health Psychol 2003;22(4):414-23. 27. Tapanya S. A biopsychosocial intervention program to improve medical regimen adherence and glycemic control in type 2 diabetic patients [unpublished dissertation]. New Brunswick, Canada: University of New Brunswick; 1996. 28. Wing RR, et al. A family-based approach to the treatment of obese type II diabetic patients. J Consult Clin Psychol 1991;59(1):156-62. 29. Baker KR, et al. The efficacy of home based progressive strength training in older adults with knee osteoarthritis: a randomized controlled trial. J Rheumatol 2001;28(7): 1655-65. 30. Stenstrom CH, et al. Intensive dynamic training in water for rheumatoid arthritis functional class IIa long-term study of effects. Scand J Rheumatol 1991;20(5):358-65. 31. Adams J, White M. Are activity promotion interventions based on the transtheoretical model effective? A critical review. Br J Sports Med 2003;37(2):106-14. 32. Spencer L, et al. Applying the transtheoretical model to exercise: a systematic and comprehensive review of the lit erature. Health Promot Pract 2006;7(4):428-43. 33. Dishman RK, Buckworth J. Increasing physical activity: a quantitative synthesis. Med Sci Sports Exerc 1996;28(6): 706-19.

AJN July 2010

Vol. 110, No. 7

37