Exercise and Osteoarthritis: Are We Stopping Too Early? Findings From the Clearwater Exercise Study

Transcription

1 Journal of Aging and Physical Activity, 2006, 14, , Human Kinetics, Inc. Exercise and Osteoarthritis: Are We Stopping Too Early? Findings From the Clearwater Exercise Study Frances V. Wilder, John P. Barrett, and Edward J. Farina The value of exercise for people with knee osteoarthritis (OA) receives continuing consideration. The optimal length of study follow-up time remains unclear. A group of individuals with knee OA participating in an exercise intervention was followed for 2 years. The authors quantified the change in knee-pain scores during Months 1 12 and during Months Eleven individuals with radiographic knee OA and knee-pain scores of 2+ were evaluated. Pain scores were collected weekly from participants who exercised three times a week. Participants demonstrated pain reduction during both time periods. Pain reduction during Months 13 24, 10.7%, was slightly higher than pain reduction during Months 1 12, 7.8%. Among people with knee OA who exercise, these findings suggest that knee-pain amelioration continues beyond 12 months. Clinicians should consider encouraging long-term exercise programs for knee-oa patients. To best characterize the effect of exercise on knee pain, researchers designing clinical trials might want to lengthen the studies duration. Key Words: study duration, trial design, pain outcome, knee osteoarthritis Arthritis and chronic joint symptoms affect nearly 70 million Americans, or about one of every three adults, making it one of the most prevalent diseases in the United States (Centers for Disease Control and Prevention, 2002). Specifically, osteoarthritis (OA) is one of the most frequent causes of physical disability among adults. More than 20 million people in the United States have the disease (National Institute of Arthritis and Musculoskeletal and Skin Diseases, 2002). Pain and loss of function are the two most prevalent symptoms among those with knee OA. In an endeavor to enhance quality of life for those with knee OA, numerous approaches to pain management have been investigated. These include weight reduction (Christensen, Astrup, & Bliddal, 2005), medicines (Biggee & McAlindon, 2004), alternative therapies (Morelli, Naquin, & Weaver, 2003), exercise (Borjesson, Robertson, Weidenhielm, Mattsson, & Olsson, 1996; Callaghan, Oldham, & Hunt, 1995; Ettinger et al., 1997; van Baar et al., 1998), and surgery (March et al., 1999). Wilder and Barrett are with the Arthritis Research Institute of America, 300 S. Duncan Avenue, Ste 240, Clearwater, FL Farina is with the Sarasota Therapy Center, 1945 Versailles St., Sarasota, FL

2 170 Wilder, Barrett, and Farina Among patients with knee OA, the value of exercise as an effective approach to pain management continues to receive considerable attention (Castaneda, Bigatti, & Cronan, 1998; Damush, Perkins, Mikesky, Roberts, & O Dea, 2005; Hurley, Mitchell, & Walsh, 2003; McCarthy et al., 2004). Because exercise is the adoption of a long-term health behavior, commensurately long prospective investigations are warranted to summarize the relationship between knee pain and exercise. Three reviews of exercise and OA have recently been published (Baker & McAlindon, 2000; McCarthy & Oldham, 1999; van Baar, Assendelft, Decker, Oostenforp, & Bijlsma, 1999). A 2004 summary of the reviews notes, Longer trials tended to be more effective than those of shorter duration and, the higher the dose of exercise, the more effective (Minor, 2004, p. 81). The length of follow-up time needed for these epidemiological studies remains unclear. In a group of participants with radiologically confirmed knee OA participating in an exercise study, we recorded self-reported knee-pain scores. All participants were followed for 2 years. Our case-series investigation was a pilot study conducted to provide supporting data for a future clinical trial addressing exercise and OA. This article summarizes our findings pertaining to 11 participants OA-pain scores over the 2-year study period. We partitioned our 2-year study into two phases: Months 1 12 and Months Our primary objective was to quantify and compare the changes in knee-pain scores for each of the two time periods. Methods Data collected from the Clearwater Exercise Study, a pilot investigation conducted to provide supporting data for a future trial about exercise and OA, were analyzed. Initiated by the Arthritis Research Institute of America, Inc., the study was a community-based study conducted in Clearwater, FL. The ambulatory, community-dwelling study population was composed of volunteer participants who were recruited by various outreach methods. Recruitment approaches included newspaper notifications, presentations at local civic meetings (e.g., Rotary Club), community organization bulletins, and friend referrals. In addition, Pinellas County employees paycheck stubs included a notice about our study recruitment efforts in their community, inviting participation. An institutional review board approved the study, and all participants signed an informed consent. Participants were subsequently screened for radiographic knee OA. Before entering the study, participants had a sedentary lifestyle, with the exception of 1 participant who did occasional walking. Procedures Study participants followed a structured routine, executed three times per week. The goal of the program was to increase joint flexibility, muscle strength, and muscle endurance. A fitness trainer supervised the institute-based exercise sessions. The 25-min routine incorporated three components. A 5- to 7-min aerobic warm-up period opened the session with participants self-selecting either a treadmill (0% incline) or a stationary vertical bike (speed mph). The aerobic warm-up was followed immediately by a weight-resistance routine. The routine was individualized by adjusting the amount of weight for each participant. Participants responses to the resistance-training stress were carefully monitored. Once the participants

3 Osteoarthritis and Exercise Duration 171 exhibited the ability to reach the maximum number of repetitions of a targeted weight for three consecutive visits, their weight was increased. In consideration of our participants age and fitness level, resistance started at a low level and progressed slowly. Strength-training endpoints were limited to submaximal efforts to volitional fatigue. The number of sets and repetitions were consistent for each study participant and progressed over time. The number of sets and repetitions for Weeks 1 4 were 2 10, 2 15, 3 10, and 3 15 (e.g., two sets of 10 repetitions each = 2 10). Subsequently, this sequence of sets and repetitions was repeated every 4 weeks. Participants weight adjustments were initiated only at the beginning of a sequence (i.e., after the 3 15 series). The machines used were a leg press, a hip-extension and -flexion instrument, a freestanding handgrip device, and a universal gym. We chose a noninclined leg press to increase quadriceps, hamstrings, gluteus maximus and minimus, anterior tibialis, and gastrocnemius/soleus strength. The primary gains were anticipated in the quadriceps and hamstrings, thus increasing the likelihood of providing support to the knee joint. In addition, because this is an exercise that engages many of the stabilizers in the upper and lower leg and the hip, we anticipated gains in overall balance and proprioception from this exercise. Hip-extension and -flexion exercise was performed in the standing position. The purpose was to increase iliopsoas and gluteus maximus strength. A handheld dynamometer was used with the participants arm at the side, elbow flexed to 90. A 3-min aerobic cool-down on either the treadmill or the stationary vertical bike concluded the exercise session. Participants The Clearwater Exercise Study s inclusion criteria were being of either sex, age 40 years and older, with radiological evidence of OA at the hands, feet, knees, shoulders, or cervical spine. For the purpose of the current investigation, however, only study participants presenting with knee OA were included. Some of our participants also had OA at other sites. Eligibility for the current study investigating knee pain included a baseline knee-pain score of 2 or higher (on a scale of 0 to 10). Exclusion criteria included expecting to relocate from the study area within 24 months of entry, an inability to obtain a release from their primary health-care provider, being mentally incapable to provide informed consent, or having a severe mobility limitation (e.g., confined to a wheelchair) that rendered them unable to partake in the exercise regimen. Of the 109 Clearwater Exercise Study participants, 54 had knee OA at baseline. Among this group, 25 had an average knee-pain score of 2 or higher. Sixteen of these completed 2 years of follow-up. Five of the 16 participants were lacking sufficient data to be included in the analyses. The resulting sample size for our case series was 11. Baseline characteristics of our study group are shown in Table 1. The average age of our participants was 74 years. Our group presented with an average kneepain score of 4.9 and an average body-mass index (BMI) of 29 (wt/ht 2 ). Most of our participants reported medication use the month before study entry (n = 8). All of our 11 participants had knee OA, and 10 also had OA of the hand, foot, or cervical spine. Among those who selected the stationary bike for their warm-up, the average baseline pain score was slightly lower (four tenths of a point) than the baseline score among those who selected the treadmill (p =.77). We conducted

4 172 Wilder, Barrett, and Farina Table 1 Baseline Characteristics of Participants (N = 11) M Min Max SD n % Knee-pain score a K&L knee-oa grade b K&L knee-oa grade change c Age (years) Body-mass index (wt/ht 2 ) Arthritis medication use d OA at any other site e Female Married Note. OA = osteoarthritis. a Likert scale of b Kellgren and Lawrence osteoarthritis (OA) grading scale: 0 4 (0 = no OA, 4 = severe OA). Left-knee (0 4) and right-knee (0 4) scores were combined (0 8). c Radiographic OA-score change during the 24-month period. Left-knee scores (0 4) and right-knee scores (0 4) were combined (0 8). Mean change reflects the combined scale of 0 to 8. p <.14. d During the month before study entry, participants self-reported taking no arthritis medication. e Kellgren and Lawrence OA Grade 2+ at any of the following joints: second distal interphalangeal joint, third proximal interphalangeal joint, first carpometacarpal joint of the thumb, first metatarsal phalangeal joint, or cervical spine. Table 2 Evaluation of Baseline Differences by Study Retention Status (Continuous Variables) Retained (n = 11) Dropout (n = 14) M Min Max M Min Max Knee-pain score a Kellgren and Lawrence knee-oa grade b Age (years) Body-mass index Note. All retention-status comparisons were statistically nonsignificant at p <.05. a Likert scale, b Osteoarthritis (OA) grading scale: 0 4 (0 = no OA, 4 = severe OA). Left-knee scores (0 4) and rightknee scores (0 4) were combined for a scale of 0 to 8. an evaluation of potential study bias because of follow-up losses (Tables 2 and 3). To better describe our study population, we evaluated demographic and clinical characteristics for differences between those who were retained and those who dropped out. The retained group reported a baseline knee-pain score five tenths of a point lower than the dropout group (p =.62). Of the 16 factors considered, only one was statistically significant. Nonknee sites affected by osteoarthritis could have

5 Osteoarthritis and Exercise Duration 173 Table 3 Evaluation of Baseline Differences by Study Retention Status (Binary Variables) Retained (n = 11) Dropout (n = 14) n % n % Arthritis-medication use a Radiographic hand OA b Radiographic foot OA c Radiographic cervical-spine OA d OA at any other site e OA at exactly 1 other site OA at exactly 2 other sites OA at exactly 3 other sites Warm-up choice: stationary bicycle Warm-up choice: treadmill Female Married Note. Unless otherwise noted, all comparisons were statistically nonsignificant. OA = osteoarthritis. a During the month before study entry, participants self-reported taking no arthritis medication. b Kellgren and Lawrence OA Grade 2+ at any of the following joints: distal interphalangeal joint, third proximal interphalangeal joint, or first carpometacarpal joint of the thumb. c Kellgren and Lawrence OA Grade 2+ at the first metatarsal phalangeal joint. d p =.05 e Kellgren and Lawrence OA Grade 2+ at any of the following joints: second distal interphalangeal joint, third proximal interphalangeal joint, first carpometacarpal joint of the thumb, first metatarsal phalangeal joint, or cervical spine. influenced study retention. The number of other (nonknee) sites affected with OA by retention status was evaluated (Table 3). The percentage of participants who presented with OA at any other site was similar for those who were retained (91%) and those who were dropouts (93%). Further investigation of the number of other sites affected showed differences by retention status. Participants with foot OA (another weight-bearing joint) appeared more likely to drop out. Those who stayed in the study had a higher percentage of cervical-spine OA (82%) than did those who dropped out (43%). Baseline left- and right-knee OA scores were combined for a scale of 0 8. The dropout group demonstrated a lower average radiographic knee- OA score (4.6) than did the retained group (5.4). Although participants were able to self-select either a weight-bearing (treadmill) or a non-weight-bearing (stationary bicycle) warm-up, retention status was comparable between warm-up methods (p =.94). The mean pain score among those who selected the stationary bike was slightly lower (four tenths of a point) than that of those who selected the treadmill (p =.77). Among the 14 participants who dropped out, none reported withdrawing because of OA or study-related issues. Participants were lost to follow-up for the following reasons: moved out of state (1), scheduling problems (2), unrelated health issues (2), family considerations (1), and general study withdrawal (8).

6 174 Wilder, Barrett, and Farina Data Collection and Measures Participants knee-pain scores were collected weekly through a self-administered questionnaire. We used a standard Likert scale to record self-reported knee-pain level (Bolognese, Schnitzer, & Ehrich, 2003). The scale s values spanned from 0 (no pain) to 10 (severe pain). Although visual analog scales (VAS) are frequently used to assess self-reported pain, a recent study demonstrated that VAS and Likert responses are highly correlated for discriminating treatments in people with OA (Bolognese et al.). Because Likert responses can be easier to interpret, this scale was selected for the current study. Participants recorded their pain scores before engaging in their exercises. Because of the variability and subjective nature of self-reported pain scores, we pooled participants baseline scores and follow-up scores. Specifically, for the first 12-month period we generated a baseline knee-pain score by averaging each participant s scores during Months 1 and 2. For this same 12-month period, we generated a follow-up knee-pain score by averaging each participant s scores during Months 11 and 12 (see Figure 1). For the time period spanning Months 13 24, the average knee-pain scores for baseline (Months 13 and 14) and follow-up (Months 23 and 24) were calculated in the same manner. To determine OA status, each radiograph was graded 0 to 4 by the ordinal criteria of Kellgren and Lawrence: 0 = absent, 1 = questionable osteophytes and no joint-space narrowing, 2 = definite osteophytes with possible joint-space narrowing, 3 = definite joint-space narrowing with moderate multiple osteophytes and some sclerosis, and 4 = severe joint-space narrowing with cysts, osteophytes, and sclerosis present (Kellgren & Lawrence, 1957). A board-certified radiologist interpreted the radiographs. Participants whose radiographs were interpreted as Grade 0 or 1 were considered free of OA. Those whose radiographs were interpreted as Grade 2, 3, Months 1 12 Baseline Follow-up Months Baseline Follow-up Figure 1 Method to quantify participants baseline and follow-up knee-pain scores for the two time periods. Black cells represent months of baseline pain scores. The gray cells represent months of follow-up pain scores.

7 Osteoarthritis and Exercise Duration 175 or 4 were classified as having OA. Left-knee scores (0 4) and right-knee scores (0 4) were combined for a scale of 0 8. Methods of Analysis Frequencies and means were used to describe baseline characteristics of the study sample. Data used were from the Likert pain scale, the intake survey form, and the radiographic-assessment sheet. Demographic information and clinical factors such as OA at other sites were summarized to provide general information about the study participants. Our study objective was to quantify the changes in knee-pain scores for each of the two time periods, Months 1 12 and Months Accordingly, for each time period we tested whether the mean difference in pain scores (baseline minus follow-up) was significantly different from zero. To accomplish this, we conducted a one-sample, paired-difference t test. Values for p less than or equal to.05 were considered statistically significant. SAS version 9.1 software was used for all statistical analyses. Results Our primary objective was to quantify the change in knee-pain scores for each of the two time periods. Our results demonstrated a reduction in knee-pain score for both of the time periods examined. During the first 12 months, the participants experienced an average score reduction of 0.38 (SD = 3.1) points (Table 4), but wide variability in pain-score change was noted. The largest reduction in score was 5.3 points, and the greatest increase in pain score was 5.9 points. Overall, this reflected a reduction in knee-pain score of 7.8% from baseline (Table 4). During the subsequent time period (Months 13 24), the participants demonstrated a 0.47-point (SD = 0.59) drop in pain score. Score-change variability was markedly lower the largest score reduction was 1.4 and the largest score increase was During this latter time period, these changes reflected an overall 10.7% reduction in knee-pain scores (p <.02; Table 4). Although this was not directly related to our primary study hypothesis, ancillary analyses indicated a statistically nonsignificant drop of 0.8 points in the knee X-ray scores (p <.14) over the 2-year study period (Table 1). Discussion Among our case series of knee-oa participants, we quantified change in knee-pain scores for each of the two time periods. Participants experienced a reduction in knee pain during each of the two periods, Months 1 12 and Months The painscore reduction reported during Months 13 24, 10.7%, was slightly greater than the pain-score reduction reported during Months 1 12, 7.8%. Although only a clinical trial can infer causal relationships, our findings suggest that individuals with knee OA can benefit from a program of regular exercise. Just as important, our data imply that knee-pain amelioration extends beyond 12 months of exercise participation. Because our case-series group lacked a comparable cohort of nonexercisers with knee OA, we cannot explicitly state the impact that the exercise regimen had on

8 176 Wilder, Barrett, and Farina Table 4 Pain Score and Effect Size by Study Time Period (N = 11) Baseline a Follow-Up b Effect Size Time period M Min Max SD M Min Max SD M Min Max SD % Change a For time period 1 12, average scores during Months 1 2; for time period 13 24, average scores during Months b For time period 1 12, average scores during Months 11 12; for time period 13 24, average scores during Months

9 Osteoarthritis and Exercise Duration 177 OA-related knee pain. Nonetheless, OA is widely characterized as a progressive disease (Raynauld et al., 2004), and many would contend that with the passage of time, participants knee pain would have either remained stable or increased. Although most studies investigating the relationship between exercise and knee OA have been carried out for periods of less than 2 years, researchers are beginning to conduct longer studies (Ettinger et al., 1997; Rejeski, Ettinger, Martin, & Morgan, 1998). Our case series offers evidence that invites future research on the benefits of longer term participation in exercise. We suggest that these findings are relevant to clinicians, researchers, and patients in three ways. In tandem with previously published research on exercise and knee-oa pain (Baker & McAlindon, 2000; Fransen, McConnell, & Bell, 2002, 2003; Petrella, 2000), our results should encourage clinicians to recommend exercise to patients with OA as an approach to the management of knee pain. Second, our pilot data supply a framework to guide future researchers selection of study duration when conducting trials in this field. To best characterize the effect of exercise on knee pain, researchers designing clinical trials might want to extend the length of study follow-up time. Without question, a suitably conducted clinical trial will demand considerable resources. An unnecessarily lengthy trial can misuse such resources, whereas a shorter trial might fail to capture the full effect of the exercise intervention. Clearly, there must be a balance as these, and other important design issues, are determined. Finally, if these data suggest that exercise participation for more than 12 months results in continued relief from knee pain (if not an enhanced level of relief), then people with knee OA should be encouraged to hold fast to their exercise routines for an extended time period. Knee-joint instability, a common symptom of the degeneration process secondary to OA, increases patients pain and often their fear of falling. Exercises that strengthen the surrounding joint muscles and facilitate improved lower extremity balance and proprioception play a key role in joint protection and maintaining an older person s overall functional capacity. Physical activity through flexibility and joint movement enhances the diffusion of joint fluid through the cartilage matrix nourishing the joint. These range-of-motion-type exercises greatly reduce stiffness and have been shown to be effective in significantly reducing joint swelling. Aerobic exercise for patients with OA improves cardiovascular fitness and general overall health. Benefits might also include weight control and the prevention of hypertension, thus enhancing quality of life. Limitations Our modest sample size limited our ability to fully summarize the relationship between exercise duration and OA-related knee pain. Findings from a case series of 11 participants, albeit informative with regard to the study s 2-year duration, have a restricted range of implications. Future similar studies with larger sample sizes might have the statistical power to test hypotheses related to exercise and OA-related pain. Our findings, though, should be interpreted appropriately in light of the sample size and the case-series design. A larger sample would have also enabled us to comment on pain differences and study duration by sex, age group, and BMI category. The absence of a control group also limits the extent to which we can infer from our findings.

10 178 Wilder, Barrett, and Farina Our study lacked detailed data pertaining to participants use of pain-relief medication. Such data would have provided the ability to identify any significant modifications in participants medication use that might have influenced our study outcome, pain. Although we were able to broadly address arthritis medications by examining participants medication status during the month before study entry, the data were inadequate to properly address this important area. Specifically, knowledge of medication status and dosage (both disease modifying and pain relief) would have enabled us to comment on the extent to which our findings might have been attributable to such an influence. We were, however, able to ascertain that the 8 participants who reported using arthritis medication during the month before study entry continued this pattern of use during the study. Of the 3 participants who reported no arthritis-medication use at baseline, 2 subsequently reported such use. The net increase in participants medication use might have influenced the percentage of pain reduction reported in our findings. As with many prospective investigations, losses to follow-up could have influenced our findings. Among the 25 people eligible for our pain study (had knee OA and a pain score of 2+), only 11 were included in the final analyses. Nine participants did not remain in the study for 2 years, and 5 had missing data. Notwithstanding, t-test results indicated that the two groups (those retained and those who dropped out) did not differ significantly by baseline pain score (p =.62). The mean baseline knee-pain score for those who dropped out (5.4) was elevated relative to the mean baseline pain score for those who were retained (4.9). Our records do not indicate that participants withdrew because of pain considerations. If this difference in baseline pain level did, in fact, contribute to the dropouts decisions to withdraw, such might have influenced our findings. Among the 11 retained, those with higher baseline pain appeared to experience a greater drop in pain scores than did those with lower baseline pain levels. If study participants dropped out because of relatively high baseline pain levels, our findings of pain amelioration among a group of exercisers could be somewhat underestimated. Retained participants demonstrated a higher mean X-ray score for baseline knee OA than did those who subsequently dropped out. As suggested by these data in our modest case series, the radiographic presentation and symptomatic presentation of knee OA are frequently inconsistent. Our pilot protocol could have been strengthened with more rigorous efforts to enhance the participant-retention rate. Future researchers investigating the efficacy of exercise in people with OA might wish to incorporate numerous strategies aimed at minimizing dropout. Continual reinforcement of participants valuable contributions might serve this purpose. Awarding personalized appreciation certificates at regular intervals might encourage attendance (e.g., member of the Six Month Club ). Study participants might feel a stronger sense of belonging if the exercise area assumed a social atmosphere. This can be achieved with easy rapport with the exercise leader, a bulletin board for family photos and postcards, a sitting area with magazines and a water cooler, and perhaps a lending library of paperback books. Finally, preliminary study findings that are appropriate to be shared can be displayed via large, laminated posters in the exercise area. Although future exercise and OA studies will need to address losses to follow-up, these are examples of strategies that might increase the likelihood of minimizing dropout.

11 Osteoarthritis and Exercise Duration 179 Conclusion Our findings complement existing literature espousing the benefits of regular exercise on OA-related knee pain. These data afford researchers examining exercise and knee OA additional insight for the design of future trials. As a sizable percentage of baby boomers enter their senior years, the field of OA is experiencing an escalating level of interest. Healthy People 2010 (U.S. Department of Health and Human Services, 2000) is a comprehensive national agenda for promoting health and preventing disease. One of the agenda s goals reflects this research focus. Goal #2-1 states, Increase the mean number of days without severe pain among adults who have chronic joint symptoms (p. 11). Effective approaches to managing pain from knee OA will affect the quality of life for millions of people. Epidemiological studies that further characterize the relationship between duration of exercise and knee pain will move us closer to this goal. Use it or lose it and motion is lotion are simple yet powerful clichés according to which OA sufferers must live their lives. The initiation, but even more important the maintenance, of regular exercise for OA has exponential benefits over time. Increased strength and stability heighten confidence during activities of daily living and functional performance. Exercise helps prevent other associated chronic diseases such as Type 2 diabetes mellitus and coronary artery disease. Moreover, patients who permanently adopt exercise as a regular part of their lifestyle will improve quality of life and decrease disabilities associated with pain caused by OA. References Baker, K., & McAlindon, T. (2000). Exercise for knee osteoarthritis. Current Opinions in Rheumatology, 12, Biggee, B.A., & McAlindon, T. (2004). Glucosamine for osteoarthritis: Part I, review of the clinical evidence. Medicine & Health Rhode Island, 87(6), Bolognese, J.A., Schnitzer, T.J., & Ehrich, E.W. (2003). Response relationship of VAS and Likert scales in osteoarthritis efficacy measurement. Osteoarthritis and Cartilage, 11(7), Borjesson, M., Robertson, E., Weidenhielm, L., Mattsson, E., & Olsson, E. (1996). Physiotherapy in knee osteoarthrosis: Effect on pain and walking. Physiotherapy Research International, 1, Callaghan, M.J., Oldham, J.A., & Hunt, J. (1995). An evaluation of exercise regimes for patients with osteoarthritis of the knee: A single blind randomized controlled trial. Clinical Rehabilitation, 9, Castaneda, D.M., Bigatti, S., & Cronan, T.A. (1998). Gender and exercise behavior among women and men with osteoarthritis. Women s Health, 27(4), Centers for Disease Control and Prevention. (2002). Prevalence of self-reported arthritis or chronic joint symptoms among adults United States, Morbidity and Mortality Weekly Report, 51(42), Christensen, R., Astrup, A., & Bliddal, H. (2005). Weight loss: The treatment of choice for knee osteoarthritis? A randomized trial. Osteoarthritis and Cartilage, 13(1), Damush, T.M., Perkins, S.M., Mikesky, A.E., Roberts, M., & O Dea, J. (2005). Motivational factors influencing older adults diagnosed with knee osteoarthritis to join and maintain an exercise program. Journal of Aging and Physical Activity, 13(1), Ettinger, W.H., Jr., Burns, R., Messier, S.P., Applegate, W., Rejeski, W.J., Morgan, T., et al. (1997). A randomized trial comparing aerobic exercise and resistance exercise

12 180 Wilder, Barrett, and Farina with a health education program in older adults with knee osteoarthritis. The Fitness Arthritis and Seniors Trial (FAST). Journal of the American Medical Association, 277, Fransen, M., McConnell, S., & Bell, M. (2002). Therapeutic exercise for people with osteoarthritis of the hip or knee. A systematic review. Journal of Rheumatology, 29(8), Fransen, M., McConnell, S., & Bell, M. (2003). Exercise for osteoarthritis of the hip or knee. Cochrane Database System Review, 3, CD Hurley, M.V., Mitchell, H.L., & Walsh, N. (2003). In osteoarthritis, the psychosocial benefits of exercise are as important as physiological improvements. Exercise and Sport Sciences Reviews, 31(3), Kellgren, J.H., & Lawrence, J.S. (1957). Radiological assessment of osteoarthrosis. Annals of Rheumatic Disease, 16, March, L.M., Cross, M.J., Lapsley, H., Brnabic, A.J., Tribe, K.L., Bachmeier, C.J., et al. (1999). Outcomes after hip or knee replacement surgery for osteoarthritis. A prospective cohort study comparing patients quality of life before and after surgery with age-related population norms. Medical Journal of Australia, 171(5), McCarthy, C.J., Mills, P.M., Pullen, R., Roberts, C., Silman, A., & Oldham, J.A. (2004). Supplementing a home exercise programme with a class-based exercise programme is more effective than home exercise alone in the treatment of knee osteoarthritis. Rheumatology, 43(7) McCarthy, C.J., & Oldham, J.A. (1999). The effectiveness of exercise in the treatment of osteoarthritic knees: A critical review. Physical Therapy Review, 4, Minor, M.A. (2004). Impact of exercise on osteoarthritis outcomes. Journal of Rheumatology, 31(Suppl.), 70, Morelli, V., Naquin, C., & Weaver, V. (2003). Alternative therapies for traditional disease states: Osteoarthritis. American Family Physician, 67(2), National Institute of Arthritis and Musculoskeletal and Skin Diseases. (2002). Handout on health: Osteoarthritis. NIH Pub. No Washington, DC: National Institutes of Health. Petrella, R.J. (2000). Is exercise effective treatment for osteoarthritis of the knee? British Journal of Sports Medicine, 34(5), Raynauld, J.P., Martel-Pelletier, J., Berthiaume, M.J., Labonte, F., Beaudoin, G., de Guise, J.A., et al. (2004). Quantitative magnetic resonance imaging evaluation of knee osteoarthritis progression over two years and correlation with clinical symptoms and radiologic changes. Arthritis and Rheumatism, 50(2), Rejeski, W.J., Ettinger, W.H., Jr., Martin, K., & Morgan, T. (1998). Treating disability in knee osteoarthritis with exercise therapy: A central role for self-efficacy and pain. Arthritis Care and Research, 11(2), Thomas, K.S., Muir, K.R., Doherty, M., Jones, A.C., O Reilly, S.C., & Bassey, E.J. (2002). Home based exercise programme for knee pain and knee osteoarthritis: randomised controlled trial. British Medical Journal, 325(7367), 752. U.S. Department of Health and Human Services. (2000). Healthy People 2010 (2nd ed.). Washington, DC: U.S. Government Printing Office. van Baar, M.E., Assendelft, W.J.J., Decker, J., Oostenforp, R.A.B., & Bijlsma, J.W.J. (1999). Effectiveness of exercise therapy in patients with osteoarthritis of the hip or knee: A systematic review of randomized clinical trials. Arthritis and Rheumatism, 42, van Baar, M.E., Dekker, J., Oostendorp, R.A., Bijl, D., Voorn, T.B., Lemmens, J.A., et al. (1998). The effectiveness of exercise therapy in patients with osteoarthritis of the hip or knee: A randomized clinical trial. Journal of Rheumatology, 25(12),