Ottawa panel evidence-based clinical practice guidelines for therapeutic exercises and manual therapy in the management of osteoarthritis.

Introduction

Osteoarthritis (OA) affects a large proportion of the population. Its prevalence is increasing dramatically as the populations of industrialized countries age and the baby boomers enter older adulthood. (1) It has been estimated that the prevalence will increase in the United States from 43 million in 1997 to 60 million in 2020. (2) Similarly, in Canada, an increase from 2.9 in 1991 to 6.5 million in 2033 (a 124% increase) is expected. (3) Osteoarthritis is recognized as a substantial source of disability with significant social and financial costs due to surgical and medical interventions and frequent absenteeism from work. (1,4) In 1994, the total cost for arthritis and rheumatism in Canada was estimated to be between Can $4.3 billion and $7.3 billion, (5) and the estimated medical expenses (excluding cost of time lost from paid or unpaid work) were estimated to be between Can $1.7 billion and $2.5 billion. (6)

Efficiency and efficacy of rehabilitation interventions in OA management have an obvious bearing on the direct and indirect costs of the disease. The development of evidence-based clinical practice guidelines (EBCPGs) will assist patients and clinicians in maximizing their rehabilitative efforts. Evidence-based clinical practice guidelines are systematically developed statements to help practitioners and clients choose proper health care for specific clinical circumstances (7) and can improve both a patient's health outcomes and the process of care. (8) A rapid and exponential growth in evidence-based clinical practice guideline (EBCPG) development has been observed in the last decade and may have resulted in several occasions of conflicting guidelines on the same topic. (9,10) These inconsistencies are attributed to variations in EBCPG development processes and quality. (9,11,12) Several authors (10,13,14) have recommended that all EBCPGs be assessed in a systematic manner using a standardized appraisal tool.

The Ottawa Panel was convened to evaluate the strength of the scientific evidence on the efficacy of therapeutic exercises (TE) for patients with OA. A previous and similar article, using the rigorous methodology, (15) was written by the Ottawa Panel on rheumatoid arthritis (RA). (16) The Ottawa Panel also is preparing EBCPGs on the use of TE, electrotherapy, and thermotherapy modalities and on patient education and splinting and orthosis for patients with OA. In this article, the Ottawa Panel considers various types of TE: specific strengthening exercises, general physical activity, and manual therapy combined with exercises.

Several systematic reviews and meta-analyses on the effectiveness of TE for patients with OA have been published in the scientific literature, demonstrating a strong interest in this intervention. Two meta-analyses using Cochrane Collaboration methods have been conducted for the management of patients with OA: the effectiveness of exercise for managing patients with hip and knee OA (17) and the ideal intensity of exercise for OA management. (18) Of 3 systematic reviews on the effectiveness of exercise for managing patients with OA, one was published a few years ago in a scientific journal (19) and 2 were completed more recently and focused on the efficacy of strengthening exercises (20) and fitness exercises. (21) Eight other reviews (22-29) exist on TE for arthritis. Several of these reviews need updating, were not systematic, or were not specific to OA. Nevertheless, all of these reviews unanimously agreed that TE are beneficial for patients with OA, depending on the type and application of exercises (eg, strengthening, fitness, or combination of manual therapy and exercise), the outcomes, the specific joint affected, and the stage of the disease. To our knowledge, no reviews are available on manual therapy (alone or combined with exercises); only one randomized controlled trial (RCT) has been published on this topic. (30)

Several EBCPGs are available for the management of patients with OA using TE. (31-36) These EBCPGs have been developed mainly for medical and surgical interventions and are often not precise regarding rehabilitation interventions. Only British Medical Journal (34) has published recommendations on exercise, but they were based on existing systematic reviews that had not been recently updated (Appendix 1). All of the aforementioned EBCPGs are generally flawed. The authors did not use a systematic literature search to find the studies that ultimately formed the basis of the EBCPGs, and although the authors reviewed the scientific results of each study, they did not synthesize the studies. The guidelines were developed for limited clinical practice areas. Although the EBCPGs were based on the current scientific literature, their authors used a nonstandardized approach to combine the scientific results; thus, the evidence of intervention efficacy is confusing, particularly in the presence of contradictory results. The authors also did not use a rigorous grading system to assess the evidence. Finally, with one exception, none of the guidelines have been updated recently. The Ottawa Panel is proposing more precise EBCPGs (involving specific joints, outcomes, periods of intervention, and disease stages) based on a rigorous quantitative method. (15) We believe that various people could benefit from using our guidelines, including patients, physical therapists, rheumatologists, physiatrists, orthopedic surgeons, occupational therapists, and family physicians. Our aim in developing the guidelines was to advance the proper use of the interventions studied (in this article, TE and manual therapy).

Methods

For this project, we used the same methods (15) as those of a previous study conducted by the Ottawa Panel on TE for patients with RA. (16) Evidence from RCTs and observational studies were identified and synthesized using methods defined by the Cochrane Collaboration that minimize bias by using a systematic approach to literature search, study selection, data extraction, and data synthesis. At the start of our OA project, we defined an a priori protocol that was used for separate systematic reviews of trials relating to each intervention. The strength of evidence was graded as level I for RCTs or level II for nonrandomized studies. An expert panel developed a set of criteria for grading the strength of both the evidence and the recommendation. The Ottawa Panel decided that evidence of clinically important benefit (defined as a difference of more than 15% relative to a control based on panel expertise and empiric results) in patient-important outcomes was required for a recommendation. Statistical significance also was required but was insufficient alone. Patient-important outcomes were decided by consensus as being pain, functional status, patient global assessment (defined as "patient's assessment of overall disease activity or improvement" (37)), quality of life, and return to work, providing that these outcomes were assessed with a validated scale that yields reliable data. (15)

Target Population

Studies of adult patients (> 18 years of age) with classical or definite OA as defined by Klippel et al (38) were included in our literature search. Patients with OA that affected peripheral joints were eligible to participate. Patients at different stages of the disease participated in the included clinical trials; some trials involved patients with both chronic and acute conditions. All stages of the disease were included in our analysis. The recommendations state the disease stage for which the intervention is most appropriate. If, however, the trial on which the recommendation was based did not mention disease stage, neither does our recommendation (Appendix 2). Most trials involved patients with chronic OA (>12 years' duration).

Various exclusion criteria were established:

* Studies of patients with OA involving spinal problems (excluded due to the numerous associated signs and symptoms and because the Philadelphia Panel guidelines for low back pain (39) and neck pain (40) were recently developed by the same methodologists);

* Studies of patients who recently had surgery;

* Patients with other rheumatologic or musculoskeletal problems (eg, fractures, tendinitis, or bursitis), clinically important medical problems, or psychiatric conditions that could hamper rehabilitation or reduce functional status;

* Studies of subjects without known pathology or impairments; and

* Studies of subjects with mixed arthritic conditions such as the sample in a study by D'Lima et al. (41)

Table 1 lists the complete inclusion and exclusion criteria.

Study Inclusion/Exclusion Criteria

Generally, comparisons of 2 active interventions (head-to-head studies) were excluded for the same reasons explained in the previous publication on the Ottawa Panel EBCPGs on RA. (16) Examples of head-to-head studies include dynamic exercises versus isometric exercises, (42) individual versus group exercises, (43) home exercises versus aquatics, (44) walking versus patient education, (45) sham electrical stimulation versus patient education combined with TE, (46) aerobics (walking) versus strengthening exercises, (45,47) and walking versus jogging in water. (48) Some studies had several comparative groups, and only some of the group comparisons were eligible to be included.

Other excluded interventions comprised surgery, drug, or psychosocial (nonphysical) interventions. For instance, the RCTs on exercises after a total hip replacement for severe hip OA were excluded; RCTs with frequent use of continuous passive motion (CPM) fop lowing a total knee arthroplasty for severe knee OA (49-61) also were excluded. However, practitioners can refer to a recent meta-analysis on the efficacy of CPM combined with physical therapy versus physical therapy alone (n=799) following a total knee arthroplasty for knee OA (62) to find further recommendations on these postsurgery interventions (grade A for flexion deformity and time to achieve 90 degrees of flexion and grade C+ for active knee flexion range of motion [ROM], pain related to analgesic use, and number of patients needing postoperative manual therapy). Postsurgery intervention studies usually allowed samples with varying proportions of patients with OA and RA. Most of the RCTs on efficacy of postsurgery interventions such as CPM recruited subjects with mixed arthritic conditions, which is the reason they are excluded in this article.

Subjects who received placebo, were untreated, or received routine conventional therapeutic approaches were acceptable control groups. If concurrent interventions (eg, electroanalgesia and medication) were provided to the experimental and control groups, these interventions were included. However, interventions where the patient acts as his or her own control were not included. A priori, we did not include or exclude studies based on the quality of their methods. However, we did consider quality when grading our recommendations.

The categories of interventions selected were approved by the Ottawa Panel according to the study's description of the intervention. Category selection also was influenced by previous work performed by the Ottawa Methods Group (15) and by the Ottawa Panel on TE for patients with RA. (16)

Results of Literature Search

Through a literature search (Appendix 3), 609 potential articles on TE and manual therapy for OA were identified. Based on the selection criteria checklist (Tab. 1), 113 studies were potentially relevant; 26 of these studies were ultimately included (30,42,45,47,48,63-83) (Appendix 2). One of the 26 studies had a follow-up study, so we have counted these 2 studies as one, using the number of patients in the original study when calculating patient numbers (Appendix 2). The other trials were excluded for various reasons (Tab. 2). (19,32,41,43,44,46,49-61,84-153) The search identified 31 articles on manual therapy, 3 of which were initially seen as relevant. (30,122,132) Only one article (30) was included (Appendix 2).

It was not possible to pool data to develop the following EBCPGs. Each statement of recommendation represents one trial for a specific intervention (in terms of session/ treatment duration and frequency) for a specific clinical outcome and a specific period of time. The included studies were gathered into general (ie, strengthening, general physical activity, combination of exercises) and more specific (eg, isometric, isotonic, isokinetic, eccentric, concentric, aerobic) types of TE according to the description by the trial investigators. The reader needs to refer to the tables of included studies to find more details about the characteristics of the therapeutic application of a specific TE included in the following EBCPGs.

Therapeutic Exercises

EBCPGs Related to Strengthening Exercises

Lower-extremity (LE) strengthening versus control, level 1 (3 RCTs, n= 103): grade A for pain getting up and clown from floor and functional status (clinically important benefit); grade C+ for pain during walking, pain while climbing and descending stairs, arthritis activity, functional tasks, and quadriceps femoris muscle peak torque (clinically important benefit); grade C for stiffness, mobility, quadriceps femoris muscle force, muscle activation, and qualify of life (no benefit). Patients with a diagnosis of OA of the knee.

Lower-extremity isometric strengthening versus control, level 1 (1 RCT, n=67): grade A for pain getting down to and up from floor (clinically important benefit); grade C+ for pain getting down and up stairs and timed functional tasks (clinical benefit); grade C for stiffness and functional status (no benefit). Patients with a diagnosis of OA of the knee.

Isotonic resistance training versus isotonic combined with isokinetic (Kinetron *) resistance training for quadriceps femoris and hamstring muscles, level 1 (1 RCT, n= 15): grade C for quadriceps femoris muscle peak torque (no benefit). Patients with a primary diagnosis of OA of the knee.

Isotonic combined with isokinetic (Kinetron) resistance training for quadriceps femoris and hamstring muscles versus control, level 1 (1 RCT, n= 18): grade C for muscle force (no benefit). Patients with primary diagnosis of OA of the knee.

Eccentric resistance training (Cybex *) for quadriceps femoris and hamstring muscles versus control, level 1 (1 RCT, n= 17): grade C for muscle force (no benefit). Patients with primary diagnosis of OA of the knee.

Concentric resistance training for quadriceps femoris and hamstring muscles versus control, level 1 (1 RCT, n= 15): grade A for pain at rest and during activities (clinically important benefit); grade C for global functional status (no benefit). Patients with knee OA bilaterally and grade II or III OA.

Concentric-eccentric resistance training for quadriceps femoris and hamstring muscles versus control, level 1 (1 RCT, n=14): grade A for pain at rest and during specific functional activities: 15-m walk and stair climbing/descending time (clinically important benefit). Patients with knee OA bilaterally and grade II or III OA.

Home strengthening program for quadriceps femoris muscle versus control, level 1 (I controlled clinical trial [CCT], n=53): grade A for pain, functional status, energy level, and ROM in flexion (clinically important benefit); grade C for physical mobility, muscle force, swelling, and exercise (no benefit). Patients with OA of the knee.

General LE exercise program (including muscle force resistance, flexibility, and mobility/coordination) versus control, level 1 (8 RCTs, n=876): grade A for pain at night and ability on stairs (clinically important benefit); grade C for knee flexion ROM, muscle force, knee joint position sense, kinesthesia, stance, gait, functional status, qualify of life, muscle activation, stiffness, and physical activity (no benefit). Patients with a diagnosis of OA.

Progression versus no-progression LE strengthening exercises, level 1 (1 RCT, n=179): grade A for pain at rest and ROM (clinically important benefit); grade C for stiffness and functional status (no benefit). Patients with radiographic evidence of OA in the tibiofemoral compartment.

Hand strengthening versus control, level 1 (1 RCT, n=40): grade A for pain and grip force (clinically important benefit). Patients who met the American College of Rheumatology criteria for hand OA. (154)

EBCPGs Related to General Physical Activity, Including Fitness and Aerobic Exercises

Whole-body functional exercise versus control, level 1 (4 RCTs, n=864): grade A for pain and functional status (mobility, walking, work, disability in activities of daily living [ADL]) (clinically important benefit); grade C for knee flexion ROM, quadriceps femoris muscle force, hamstring muscle force, gait, stair climbing time, climbing self-efficacy, and qualify of life (no benefit). Patients with OA of the knee.

Walking program versus control, level 1 (6 RCTs, n=711): grade A for pain, functional status, stride length, disability transferring from bed, disability bathing, aerobic capacity, exercise endurance, energy level, physical activity, and sleep (clinically important benefit); grade C+ for disability in ADL (clinical benefit); grade C for walking speed, disability toileting, disability dressing and stairs, morning stiffness, and quality of life (no benefit). Patients with OA.

Jogging in water versus control, level 1 (1 RCT, n-79): grade A for physical activity (clinically important benefit); grade C for walking time, morning stiffness, pain, grip force, trunk ROM, functional status, and exercise endurance (no benefit). Patients with current symptoms of chronic pain and stiffness in involved weight-bearing joints.

Water exercises versus control, level 1 (1 RCT, n=30): grade C for hip and shoulder abduction torque and ROM (no benefit). Patients with OA or RA diagnosed by a rheumatologist or an orthopedic physician.

Yoga versus control, level 1 (1 RCT, n=30): grade A for pain during activity and ROM (clinically important benefit); grade C for tenderness, grip force, swelling, and hand function (no benefit). Patients with OA of the distal interphalangeal or proximal interphalangeal joints of the fingers.

EBCPGs Related to the Combination of Exercises

Manual therapy combined with exercise versus control, level 1 (1 RCT, n=83): grade A for pain (clinically important benefit); grade C for 6-minute walk distance (no benefit). Patients with a diagnosis of OA.

Summary of Trials

Twenty-nine trials (n=2,486 patients) evaluated different types of TE for managing OA-affected joints of the upper extremities and LEs. Most of the trials compared these exercises with a control, but the trials examined different kinds of TE. The strengthening exercise trials were as follows: LE strengthening (n=345), (42,70,71,79) LE isometric strengthening (n=102), (42) isotonic resistance training versus isotonic combined with isokinetic (Kinetron) resistance training for the knee (n=32), (70) isotonic combined with isokinetic (Kinetron) resistance training for the knee (n=32), (70) eccentric resistance training (Cybex) for the knee (n=32), (70) concentric resistance training for the knee (n=23), (67) concentric-eccentric resistance training for the knee (n=23),67 home program strengthening for the knee (n=81), (47) general LE exercise program (including muscle force, flexibility, and mobility/ coordination) (n=490), (64,65,68,77,78,82,83) progression in LE strengthening exercises versus no progression (n= 179), (75) and home program hand strengthening (n=40). (80)

Several RCTs examined general physical activities, including fitness and aerobic exercises, such as whole-body functional exercises (n=864), (45,63,72,73,76) walking program (n=1,089), (45,47,48,69,73,74,76) jogging in water (n= 115), (48) water exercise (n=30), (81) and yoga (n=30). (66) One trial was related to the combination of manual therapy and exercises (n=83). (30)

Twenty-three included trials were RCTs (42,45,47,48,63-67,69-76,78,80-83) and 3 trials were CCTs (47,68,77) (Appendix 2). We used the Jadad scale to decide whether a study was an RCT or a CCT. (15)

The trials examined 2 basic types of exercises. The first type involved strengthening exercises, such as resistance isometric, stretching, eccentric, and concentric exercises; these exercises were specific to different muscles. The other type focused on whole-body functional strengthening programs and included aerobic conditioning and general fitness. Program duration, treatment schedule for exercise intervention, and length of exercise session varied from 4 weeks (64) to 18 months (45,73,76) for program duration, from once a week (66) to 10 times a day (80) (depending on the phase of the program) for treatment schedule, and from 5 minutes to longer per exercise session (length of exercise session increased depending on tolerance) (74) (Appendix 2).

Strengthening Exercises

Lower-extremity strengthening versus control (4 RCTs, n = 345), (42,70,71,79) showed clinical benefits for pain during walking, pain ascending and descending the stairs, quadriceps femoris muscle peak torque, and timed functional tasks (Tab. 3). Statistically significant differences were found for pain (Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC]; Fig. 1a), pain while getting up from the floor (weighted mean difference [WMD]=-2.36, 95% confidence interval [CI]= -4.22 to -0.50; Fig. 1a), and functional status (Tab. 3). However, other outcomes were not statistically significant. Outcomes were measured at the end of intervention (4 months for Topp et al (42) and 8 weeks for Kreindler et al (70) and Schilke et al (79)) or at follow-up (6 weeks for Kreindler et al (70)) (Figs. 1a-f).

[FIGURE 1 OMITTED]

For LE isometric strengthening versus control (1 RCT, n=102), (42) clinical benefits were found for pain getting down and up from the floor, pain while going up and down stairs, and timed functional tasks but not for stiffness and functional status (Tab. 4). Statistically significant differences were found for pain while getting down to the floor (WMD=-2.05, 95% CI=-3.62 to -0.48) and up from the floor (WMD=-2.14, 95% CI= -4.01 to -0.27). Stiffness, functional limitation, pain, time to get down to the floor and to get up, and time to go up and down the stairs were not considered to be clinically important benefits at 4 months (Figs. 2a-d).

[FIGURE 2 OMITTED]

One trial (RCT, n=32) (70) showed no statistically significant difference or clinically important benefit for quadriceps femoris muscle peak torque in patients with OA either at the end of a 6-week intervention or at a 6-week follow-up. This trial compared isokinetic resistance training versus isotonic and isokinetic (Kinetron) resistance training for the knee (Fig. 3), resistance training and Kinetron versus control (Fig. 4), and eccentric resistance training (Cybex) for the knee versus control (Fig. 5).

[FIGURES 3-5 OMITTED]

Statistically significant differences favored concentric exercises over control (1 RCT, n=23)67 for pain (WMD=-17.7, 95% CI=-22.79 to -12.61) and functional status (WMD=-10.85, 95% CI=-21.34 to -0.36) at 8 weeks (Figs. 6 and 7). A clinically important benefit was observed for pain (Tab. 5) but not for functional status.

[FIGURE 6 OMITTED]

For concentric-eccentric versus control (1 RCT, n=23), (67) clinically important benefits and statistically significant differences were observed for pain (WMD=-11.40, 95% CI=-17.95 to -4.85) (Tab. 6, Fig. 8) and functional status (WMD=-12.15, 95% CI=-22.67 to -1.63) (Tab. 6) at 8 weeks. Results for 15-m walk, stair-climbing time, and stair-descending time also were significant, as were results for pain at night, pain sitting, pain rising from a chair, pain standing, and pain climbing stairs (Tab. 6).

[FIGURE 8 OMITTED]

One CCT examined home program strengthening for knee versus control (n=81) (47) and showed clinically important benefits for pain, functional status, energy level (Tab. 7), and ROM (results not shown) but not for physical mobility (Tab. 7). Statistically significant data were found for WOMAC-pain (WMD=3.00, 95% CI=1.58 to 4.42), visual analog scale (VAS)-pain (WMD=3.30, 95% CI=2.62 to 3.98), WOMAC physical function index (WMD=9.90, 95% CI=8.08 to 11.72), Nottingham Health Profile (NHP)-pain (WMD=10.60, 95% CI=8.90 to 12.30), NHP-energy (WMD=15.90, 95% CI=14.87 to 16.93), NHP-physical mobility (WMD=7.10, 95% CI=4.14 to 10.06), NHP-sleep (WMD=3.40, 95% CI=0.89 to 5.91) (Figs. 9a-c, all at follow-up of 6 months), swelling (WMD=12.5, 95% CI=5.51 to 19.49), and ROM (WMD=19.5[degrees], 95% CI=5.69[degrees] to 33.31[degrees]) (results not shown). However, no statistically significant difference was observed for muscle force or exercise tolerance (results not shown). No clinically important effects were found for muscle force, swelling, or exercise tolerance.

Several trials examined general LE exercise programs (including muscle force, flexibility, and mobility/coordination) versus control (7 RCTs, n = 690). (64,65,68,77,78,82,83) Important benefits were demonstrated for pain intensity and ability to step down but not for ROM, muscle force, gait, functional status, quality of life (Tabs. 8 and 9), knee joint position at 6 weeks (Fig. 10f), or muscle activation at 6 weeks (Fig. 10b). Statistically significant differences were found for the following outcomes:

[FIGURE 10 OMITTED]

* ROM in knee flexion, most affected knee at 10-12 weeks (WMD=10.00[degrees], 95% CI=5.91[degrees] to 14.09[degrees]) (Fig. 10a);

* ROM in knee flexion, least affected knee at 10-12 weeks (WMD=10.00[degrees], 95% CI=7.75[degrees] to 12.25[degrees]) (Fig. 10a);

* ROM in knee flexion, least affected knee at 12-month follow-up (WMD=12.00[degrees], 95% CI=7.06[degrees] to 16.94[degrees]) (results not shown);

* isometric quadriceps femoris muscle force at 6 weeks (WMD=73 N, 95% CI=25.75 to 120.25 N) (Fig. 10b);

* quadriceps femoris muscle voluntary activation at 6 weeks (WMD=14.0%, 95% CI=5.87% to 22.13%) (Fig. 10b);

* aggregate functional performance time at 6 weeks (WMD=-8.47, 95% CI=-16.79 to -0.15) (Fig. 10d);

* functional status at 6 weeks (WMD=-3.50, 95% CI=-4.94 to -2.06) (Fig. 10d);

* mean change in physical function score at 3-month follow-up (WMD=-3.54, 95% CI=-6.04 to -1.04) (Fig. 10d);

* mean change in WOMAC-pain at 8 weeks (WMD= -12.10, 95% CI=-14.24 to -9.96) (Fig. 10e);

* mean change in pain at 10 to 12 weeks (WMD=-17.10, 95% CI=-29.99 to -4.21) (Fig. 10e);

* mean change in global pain score at 6-month follow-up (WMD=-1.87, 95% CI=-2.76 to -0.98) (Fig. 10e);

* mean change in pain (VAS), walking at 10 to 12 weeks (WMD=-7.07, 95% CI=-13.90 to -0.24) (Fig. 10e);

* mean change in pain (VAS), stairs (WMD=-10.42, 95% CI=-18.58 to -2.26) (results not shown);

* pain at night at 12-month follow-up (WMD=-4.00, 95% CI=-5.94 to -2.06) (Fig. 10e);

* pain at rest at 10 to 12 weeks (WMD=-1.50, 95% CI=-2.80 to -0.20) (Fig. 10e);

* pain on weight bearing at 10 to 12 weeks (WMD= -2.00, 95% CI=-3.02 to -0.98) (Fig. 10e);

* mean change in isometric knee extensor muscle force at 8 weeks (WMD=13.20 N, 95% CI=11.96 to 14.44 N) (Fig. 10g);

* mean change in isometric knee flexor muscle force at 8 weeks (WMD=9.00 N, 95% CI=8.04 to 9.96 N) (Fig. 10g);

* mean change in fast speed at 8 weeks (WMD=6.70 cm/s, 95% CI=6.34 to 7.06 cm/s) (Fig. 10h);

* mean change in fast cadence at 8 weeks (WMD = 1.60 steps/min, 95% CI=I.40 to 1.80 steps/min) (Fig. 10h);

* mean change in fast stride length at 8 weeks (WMD=4.30 cm, 95% CI=3.99 to 4.61 cm) (Fig. 10h);

* quality of life measured with Medical Outcomes Study 36-Item Short-Form Health Survey questionnaire (SF-36) at 8 weeks (WMD=3.10, 95% CI=2.76 to 3.44) (Fig. 10i);

* mean change in WOMAC-function at 8 weeks (WMD=-7.80, 95% CI=-8.48 to -7.12) (Fig. 10k);

* improvement in self-reported disability at 24-week follow-up (WMD=-1.10, 95% CI=-1.91 to -0.29) (Fig. 10k);

* mean change in left quadriceps femoris muscle force at 6-week follow-up (WMD=10.86%, 95% CI=3.15% to 18.57%) (results not shown); and

* mean change in SF-36-physical function at 8 weeks (results not shown).

No statistically significant data were found for the remaining outcomes: ROM in knee extension and flexion (Fig. 10a); improvement in hip and knee ROM at 24-week follow-up (Fig. 10a); improvement in knee or hip muscle force at 24-week follow-up (Fig. 10c); Health Status Survey (HSS) score (Fig. 10d); pain, pain during walking, and pain at night at 10 to 12 weeks (Fig. 10e); knee joint position sense at 6 weeks (Fig. 10f); peak torque of the knee extensors and flexors at 10 to 12 weeks (Fig. 10g); step frequency and stride length at 10 to 12 weeks (Fig. 10h); stance at 10 to 12 weeks or at 12-month follow-up for most affected and least affected LEs (Fig. 10j); walking speed and stair-climbing time at 10 to 12 weeks or at 12-month follow-up (Fig. 101); Algofunctional Index-pain at 10 to 12 weeks or at 12-month follow-up (Fig. 10m); improvement in physical activity at 10 to 12 weeks (Fig. 10n); and ability to step down (Fig. 10o).

For progression versus no-progression LE strengthening exercises (one RCT, n=179), (75) clinical benefits (Tab. 10) and statistically significant differences were found for ROM in knee flexion (WMD=13[degrees], 95% CI=11.55[degrees] to 14.45[degrees]) and pain at rest (WMD=-23 mm, 95% CI=-24.03 to -21.97). No important differences were found for WOMAC-stiffness, WOMAC-pain, or pain after walk test. Outcomes were measured after 8 weeks (Figs. 11a-c).

[FIGURE 11 OMITTED]

Hand strengthening versus control (one RCT, n=40) (80) showed clinically important benefits for pain and grip force at 3 months (Tabs. 11 and 12, Figs. 12a-b). Statistically significant differences were found for pain (WMD=7.43, 95% CI= 1.78 to 31.04) and change in grip force in the right hand (WMD=0.11, 95% CI=0.09 to 0.13) and the left hand (WMD=0.10, 95% CI=0.09 to 0.11) (Fig. 12b).

[FIGURE 12 OMITTED]

General Physical Activities, Including Fitness and Aerobic Exercises

For whole-body functional exercise versus control (5 RCTs, n=864), (45,63,72,73,76) clinically important benefits were found for pain and functional status (mobility, walking, work, and disability on ADL). Statistically significant differences were found for numerous outcomes:

* pain frequency in transfer at 9 months (WMD= 0.88, 95% CI=0.49 to 1.27) (Fig. 13a);

* pain intensity in transfer at 3 months (WMD= -0.94, 95% CI=-1.33 to -0.55), at 9 months (WMD=-0.46, 95% CI=-0.84 to -0.08), and at 18 months (WMD=-0.37, 95% CI=-0.70 to -0.04) (Fig. 13a);

* pain (WMD=-0.80, 95% CI=-1.29 to -0.31) (Fig. 13b);

* functional status measured with the Arthritis Impact Measurement Scales (AIMS) (WMD=5.49, 95% CI=3.92 to 7.06) (results not shown);

* functional status measured with the Arthritis Impact Measurement Scales 2 (AIMS2): arthritis pain (WMD=-0.85, 95% CI=-1.52 to -0.18) (Fig. 13b);

* functional status measured with AIMS2: mobility level at 12 weeks (WMD=-0.50, 95% CI=-0.93 to -0.07) favoring the control group (Fig. 13c);

* functional status measured with AIMS2: walking and bending at 12 weeks (WMD=-1.25, 95% CI= -2.08 to -0.42) (Fig. 13c);

* functional status measured with AIMS2: level of tension at 12 weeks (WMD=2.58, 95% CI=1.88 to 3.28) (Fig. 13c);

* hamstring muscle and low back flexibility at 12 weeks (WMD=3.63 in, 95% CI=2.04 to 5.22 in) (Fig. 13d);

* 5-minute walk test at 12 weeks (WMD=42.19 m, 95% CI=14.19 to 70.19 m) (Fig. 13e);

* hamstring muscle isometric torque at 30 degrees: most affected LE at 12 weeks (WMD=8.85 N*m, 95% CI=I.91 to 15.79 N*m) (Fig. 13f);

* hamstring muscle isometric torque at 30 degrees: least affected LE at 12 weeks (WMD=10.51 N*m, 95% CI=3.24 to 17.78 N*m) (Fig. 13f);

* quadriceps femoris muscle isometric torque at 60 degrees: most affected LE at 12 weeks (WMD=19.80 N*m, 95% CI=4.75 to 34.85 N*m) (Fig. 13f);

* quadriceps femoris muscle isometric torque at 60 degrees: least affected LE at 12 weeks (WMD = 17.03 N*m, 95% CI=I.08 to 32.98 N*m) (Fig. 13f);

* hamstring muscle isometric torque at 60 degrees: most affected LE at 12 weeks (WMD=8.02 N*m, 95% CI=0.88 to 15.16 N*m) (Fig. 13f);

* hamstring muscle isometric torque at 60 degrees: least affected LE at 12 weeks (WMD=10.99 N*m, 95% CI=3.68 to 18.30 N*m) (Fig. 13f);

* hamstring muscle isokinetic torque at 30[degrees]/s: most affected LE at 12 weeks (WMD=10.98 N*m, 95% CI=0.38 to 21.58 N*m) (Fig. 13g);

* hamstring muscle isokinetic torque at 30[degrees]/s: least affected LE at 12 weeks (WMD=10.98 N*m, 95% CI=3.24 to 17.78 N*m) (Fig. 13g);

* hamstring muscle isokinetic torque at 90[degrees]/s: most affected LE at 12 weeks (WMD=9.73 N*m, 95% CI=-1.40 to 20.86 N*m) (Fig. 13g);

* cadence at 3 months (WMD=0.87 steps/min, 95% CI=0.33 to 1.41 steps/min) (Fig. 13h), at 9 months (WMD=0.94 steps/min, 95% CI=0.37 to 1.51 steps/min), and at 18 months (WMD=2.08 steps/ min, 95% CI=1.56 to 2.60 steps/min) (results not shown);

* stride length at 3 months (WMD=2.17 cm, 95% CI=I.18 to 3.16 cm), at 9 months (WMD=2.84 cm, 95% CI=1.77 to 3.91 cm), and at 18 months (WMD=6.49 cm, 95% CI=5.49 to 7.49 cm) (Fig. 13i);

* walking speed at 3 months (WMD=3.77 cm/s, 95% CI=2.60 to 4.94 cm/s), at 9 months (WMD=4.37 cm/s, 95% CI=3.12 to 5.62 cm/s), and at 18 months (WMD=7.79 cm/s, 95% CI=6.60 to 8.98 cm/s) (Fig. 13j);

* stance time at 3 months (WMD=-0.01 s, 95% CI=-0.01 to -0.01 s) and at 18 months (WMD=-0.02 s, 95% CI=-0.02 to -0.02 s) (Fig. 13k);

* percentage of swing at 3 months (WMD=0.39, 95% CI=0.18 to 0.60), at 9 months (WMD=-0.36, 95% CI=-0.59 to -0.13), and 18 months (WMD=0.54, 95% CI=0.32 to 0.76) (Fig. 13k);

* stair-climbing time at 18 months (WMD=-1.92 s, 95% CI=-2.01 to -1.83 s) (Fig. 13l);

* climbing self-efficacy score at 18 months (WMD=9.32, 95% CI=8.86 to 9.78) (Fig. 13m);

* quality of life (WMD=3.10, 95% CI=2.97 to 3.23) (results not shown); and

* disability in bathing (WMD=0.41, 95% CI=0.18 to 0.91) (results not shown).

[FIGURE 13 OMITTED]

No clinically important benefits were found for quadriceps femoris and hamstring muscle force at 12 weeks (Figs. 13f-g), knee flexor ROM, gait, or quality of life (results not shown). No statistical data were found for pain intensity and frequency in ambulation at 3, 9, and 18 months (Fig. 13a); pain frequency in transfer at 9 and 18 months (Fig. 13a); AIMS2 hand and finger functional status, arm functional status, self-care tasks, household tasks, social activity, support from friends, work, or mood at 12 weeks (Fig. 13c); quadriceps femoris muscle isometric and isokinetic torque at 30 degrees (Figs. 13f-g), quadriceps femoris muscle isokinefic torque at 90 degrees (Fig. 13g), or hamstring muscle isokinetic torque at 90 degrees (Fig. 13g), all at 12 weeks; stance time at 9 months (Fig. 13k) ; or incidence of disability in ADL, disability in transferring from a bed to a chair, disability in toileting, disability in dressing and eating, or quality of life measured by HSS score (results not shown).

Six RCTs and 1 CCT examined walking versus control (n=1,089), (45,47,48,69,73,74,76) and trials discovered clinical benefits for pain, functional status, stride length, disability transferring from bed, disability bathing, disability in ADL, energy level, medication use, aerobic capacity, and quality of life (Tabs. 13 and 14). No clinical benefits were found for walking speed (Tab. 13), pain in ambulation (results not shown), disability toileting, or disability dressing (Fig. 13e, both at 18-month follow-up).

Statistically significant results were shown for the following outcomes:

* pain frequency in ambulation at 3 months (WMD= -0.56, 95% CI=-1.07 to -0.05) and in transfer (WMD=-0.42, 95% CI=-0.77 to -0.07) (results not shown);

* pain intensity in transfer at 3 months (WMD=-0.55, 95% CI=-1.02 to -0.08), at 9 months (WMD= -0.46, 95% CI=-0.84 to -0.08), and at 18 months (WMD=-0.41, 95% CI=-0.76 to -0.06) (results not shown);

* NHP-physical mobility, -pain, -energy, and -sleep (Tab. 13);

* WOMAC-physical function and -pain (Tab. 13);

* VAS-pain (Tab. 13);

* walking speed at 3 months (WMD=3.69 cm/s, 95% CI=2.47 to 4.91 cm/s), at 9 months (WMD=10.29 cm/s, 95% CI=9.05 to 11.53 cm/s), and at 18 months (WMD=10.29 cm/s, 95% CI=9.07 to 11.51 cm/s) (Tab. 13);

* cadence at 3 months (WMD=3.56 steps/min, 95% CI=3.00 to 4.12 steps/min), at 9 months (WMD=3.69 steps/min, 95% CI=3.12 to 4.26 steps/min), and at 18 months (WMD=3.77 steps/ min, 95% CI=3.21 to 4.33 steps/min) (Tab. 13);

* stance time at 3 months (WMD=-0.04 s, 95% CI=0.04 to -0.04 s), at 9 months (WMD=-0.03 s, 95% CI=-0.03 to -0.03 s), and at 18 months (WMD=-0.03 s, 95% CI=-0.03 to -0.03 s) (results not shown);

* percentage of swing at 3 months (WMD=0.86, 95% CI=0.64 to 1.08) and at 18 months (WMD=0.54, 95% CI=0.32 to 0.76) (Tab. 13);

* stride length at 9 months (WMD=7.53 cm, 95% CI=6.48 to 8.58 cm) and at 18 months (WMD=7.54 cm, 95% CI=6.52 to 8.56 cm) (Tab. 13);

* climbing self-efficacy score at 18-month follow-up (WMD=8.00, 95% CI=7.55 to 8.45) (Fig. 14a);

* 6-minute walk test at 8 weeks (WMD=111.50 m, 95% CI=76.24 to 146.77 m) (Fig. 14b);

* stair-climbing time at 18-month follow-up (WMD= -1.41, 95% CI=-1.51 to -1.31) (Fig. 14c);

* general health status at 18-month follow-up (WMD=3.59, 95% CI=3.46 to 3.72) (Fig. 14d);

* incidence of disability at 18-month follow-up (WMD=0.52, 95% CI=0.28 to 0.96) (Fig. 14e);

* disability in transferring from a bed to a chair at 18-month follow-up (WMD=0.42, 95% CI=0.22 to 0.79) (Fig. 14e);

* disability in bathing (WMD=0.38, 95% CI=0.17 to 0.84) (Tab. 14) and in dressing at 18-month follow-up (WMD=0.28, 95% CI=0.10 to 0.83) (Fig. 14e);

* fast speed at 8 weeks (WMD=15.00 m/min, 95% CI=6.53 to 23.47 m/min) (Fig. 14f);

* fast stride at 8 weeks (WMD=0.20 m, 95% CI=0.03 to 0.37 m) (Fig. 14f);

* free speed at 8 weeks (WMD=7.00 m/min, 95% CI=0.34 to 13.66 m/min) (Fig. 14g);

* free stride at 8 weeks (WMD=0.20 m, 95% CI=0.08 to 0.32 m) (Fig. 14g);

* AIMS--pain at 8 weeks (WMD=-1.00, 95% CI=-1.79 to -0.21) (Fig. 14i);

* AIMS--physical activity at 8 weeks (WMD=-2.22, 95% CI=-3.25 to -1.19) (Tab. 13) and at 12 weeks (WMD=-I.30, 95% C1=-2.48 to -0.12) (Tab. 14, Fig. 14j);

* AIMS-medication use (WMD=2.74, 95% CI=1.93 to 3.55) (Tab. 13);

* 15.2-m (50-ft) walking time at 8 weeks (WMD= -2.00 s, 95% CI=-2.97 to -1.03 s) and at 12 weeks (WMD=-0.90 s, 95% CI=-1.71 to -0.09 s) (Fig. 14p for 12 weeks only);

* exercise heart rate at 12 weeks (WMD=16.00 bpm, 95% CI=3.94 to 28.06 bpm) (Fig. 14q);

* aerobic capacity at 12 weeks (WMD=5.10 mL/kg [min.sup.-1], 95% CI=2.88 to 7.32 mL/kg [min.sup.-1]) (Fig. 14s); and

* exercise endurance at 12 weeks (WMD=3.30 min, 95% CI=1.12 to 5.48 min) (Fig. 14s).

No statistically significant data were found for the following: pain intensity in ambulation (results not shown), pain frequency in ambulation at 9 and 18 months (results not shown), pain frequency in transfer at 9 and 18 months (results not shown), stride length (Tab. 13), disability in toileting and in eating at 18-month follow-up (Fig. 14e), fast cadence at 8 weeks (Fig. 14f), free cadence at 8 weeks (Fig. 14g), NHP--physical mobility and WOMAC--physical function at 6-month follow-up (Fig. 14h), AIMS-pain at 12 weeks and 9-month follow-up (Tab. 13, Fig. 14i), AIMS--physical activity at 9-month follow-up (Fig. 14j), AIMS--arthritis impact at 8 weeks (Fig. 14k), grip force at 12 weeks and 9-month follow-up (Fig. 14m), NHP--pain at 6-week follow-up (Fig. 14n), morning stiffness at 12 weeks and 9-month follow-up (Fig. 14o), 15.2-m (50-ft) walking time at 9-month follow-up (Fig. 14p), resting systolic blood pressure and resting diastolic blood pressure at 12 weeks and 9-month follow-up and exercise heart rate at 9-month follow-up (Fig. 14q), trunk flexibility at 12 weeks (Fig. 14r), and aerobic capacity and exercise endurance at 9-month follow-up (Fig. 14s).

Physical activity and aerobic capacity yielded clinically important benefits favoring jogging in water versus control (one RCT, n=l15)48 (Tab. 15). However, no clinical benefits were shown for functional status (AIMS--physical activity) at 12 weeks (Tab. 15), pain at 12 weeks and 9-month follow-up (results not shown), morning stiffness at 12 weeks and 9-month follow-up (results not shown), trunk ROM at 12 weeks and 9-month follow-up (results not shown), exercise heart rate (Tab. 15), or exercise endurance at 12 weeks (Tab. 15). Statistically significant differences were found for AIMS--physical activity at 12 weeks (WMD=-1.20, 95% CI=-2.29 to -0.11) (Fig. 15a), 15.2-m (50-ft) walking time at 12 weeks (WMD=-l.10 s, 95% CI=-2.12 to -0.08 s) (Fig. 15e), exercise heart rate at 12 weeks (WMD=13.00 bpm, 95% CI=1.32 to 24.68 bpm) (Fig. 15f), exercise endurance at 12 weeks (WMD=2.80 min, 95% CI=0.23 to 5.37 min) (Fig. 15h), and aerobic capacity (WMD=5.90 mL/kg [min.sup.-1], 95% CI=3.30 to 8.50 mL/kg [min.sup.-1]) (results not shown). AIMS-pain, morning stiffness, grip force, trunk flexibility, and resting blood pressure offered no statistically significant differences (results not shown for last). One RCT that compared water exercises with control (n=30) (81) yielded no statistically significant differences and no clinical benefits for torque or ROM at 6 weeks (Figs. 16a-b).

[FIGURES 15-16 OMITTED]

For yoga versus control (one RCT, n=30), (66) clinically important benefits were found for ROM and pain during activity at 6 weeks (Figs. 17b-c) but not for tenderness, swelling, hand functional status, or grip force at 6 weeks (Figs. 17a, d, e, and f). Statistically significant data were found for mean change in tenderness of right hand (WMD=1.80, 95% CI=0.99 to 2.61) and left hand (WMD=1.73, 95% CI=0.63 to 2.83), mean change in pain during activity (WMD= -3.29, 95% CI=-5.30 to -1.28), and mean change in ROM of right hand (WMD=10.02, 95% CI=6.50 to 13.54), all at 6 weeks (Figs. 17a-c). No statistical data were found for mean change in hand pain at rest, mean change in ROM of left hand, mean change in circumference of the hands, mean change in hand functional status, or mean change in grip force of both hands (Figs. 17b-f).

[FIGURE 17 OMITTED]

Manual Therapy Combined With Therapeutic Exercises One RCT (n=83) (30) was found on manual therapy combined with exercises, and this RCT compared the intervention with a control. Important clinical benefits were demonstrated for pain but not functional status (Tab. 16). Statistically significant data were found for all the outcomes: 6-minute walk test at 4 weeks (WMD=81.90 m, 95% CI=22.85 to 140.95 m) and at 8 weeks (WMD=77.70 m, 95% CI=18.59 to 136.81 m) (Fig. 18a) and pain at 4 weeks (WMD=-416.00, 95% CI=-618.15 to -213.85) and at 8 weeks (WMD=-471.90, 95% CI=-732.81 to -210.99) (Fig. 18b).

[FIGURE 18 OMITTED]

Strength of Published Evidence Compared With Other Guidelines

Good evidence (level I, RCT) shows that various kinds of exercises and manual therapy are useful for patients with OA, with different outcomes occurring depending on the intervention: strengthening exercises relieve pain at rest and during functional activities and improve knee ROM, quadriceps femoris muscle peak torque, grip force, level of energy, and functional stares; general physical activities, including fitness and aerobic exercises, relieve pain during functional activities and improve stride length, functional status, and aerobic capacity; and manual therapy combined with TE relieves pain.

Three sources have considered the strength of evidence: The Philadelphia Panel, (36) American Pain Society, (33) and Ontario Program for Optimal Therapeutics. (35) All 3 sources reported good-quality evidence for TE, including strengthening exercises and general physical activities (Appendix 1). To our knowledge, the scientific literature offers no guidelines on manual therapy for patients with OA.

Clinical Recommendations Compared With Other Guidelines

The Ottawa Panel concluded that good evidence exists supporting the inclusion of all of the following main categories of interventions in the management of patients with OA: strengthening exercises (grade A for pain at rest and during functional activities, ROM, grip force, level of energy, and functional stares; grade C+ for quadriceps femoris muscle peak torque, specific functional activities, and timed functional activities); general physical activities, including fitness and aerobic exercises (grade A for pain during functional activities, stride length, functional stares, energy level, aerobic capacity, and medication use; grade C+ for disability in ADL); and manual therapy combined with exercises (grade A for pain). The recommendations related to strengthening exercises and general physical activities generally concur with all other existing guidelines. (31-36)

Practitioners' Response to Ottawa Panel Guidelines

The 5 practitioners who reviewed our guidelines agreed with the recommendations. Four practitioners found the recommendations to be clear; 1 practitioner was uncertain which type of exercise was effective. The Ottawa Panel explained that, depending on the specific outcome, interventions with grade A, B, or C+ are beneficial. Guideline summaries (see evidence-based clinical practice guidelines in Appendix 4) were rewritten for better comprehension. A decision aid was created to clarify the application of the guidelines. This aid can be found on the University of Ottawa Web site (www.health.uottawa.ca/rehabguidelines).

Discussion

The Ottawa Panel EBCPGs (grouped together in Appendix 4) were rigorously developed (15,16) using an extensive systematic review of TE (Figs. 1-18). Numerous grade A (n=13) and C+ (n=3) recommendations have been developed for TE for patients with OA. Various outcomes useful for rehabilitation practitioners and patients with OA were considered, such as pain, functional status, and quality of life. However, more evidence is needed to determine the efficacy of TE in the management of patients with OA. Evidence on the effectiveness of the specific type of muscle contraction to be used during resistance training, on water exercises, and on hip strengthening is lacking, as is indicated by the grade C recommendations for these interventions. No harmful side effects were reported.

Limitations

Even though the Ottawa Panel EBCPGs on OA were developed using a rigorous methodology, (15) similar methodological weaknesses were identified compared with the Ottawa Panel EBCPGs on RA. (16) More precise characteristics of the therapeutic application (eg, dosage, type of exercise used, intensity, frequency) need to be reported by investigators to reproduce the exercise programs (eg, quadriceps femoris and hamstring muscle strengthening), especially when they were proven effective. (76,78)

The Ottawa Panel EBCPGs for the management of patients with OA, however, were in concordance with Appraisal of Guidelines Research and Evaluation (AGREE) criteria (155) and yielded results identical to those of the previous Ottawa Panel EBCPGs developed for RA (16) (see University of Ottawa Web site [www. health.uottawa.ca/rehabguidelines]). Furthermore, the Ottawa Panel EBCPGs generally concur with previous and relatively recent EBCPGs (31-36) and systematic reviews for OA (19-29,156) and fit entirely with the recommendations from the Work Group on physical activity. (157,158)

Therapeutic Exercises

The Ottawa Panel concluded that TE is beneficial for patients with OA. Benefits are recognized for pain at rest and during functional activities, knee ROM, quadriceps femoris muscle peak torque, grip force, stride length, level of energy, functional status, and aerobic capacity. Quality of life also was enhanced (statistical significance only) after an 8-week LE strengthening exercise program (65) and 18 months after a walking program. (66)

Progressive exercises (75) are promising prospects for OA patient management, but results were inconclusive regarding the ideal intensity of the exercise program. (18) Results for RA, however, were conclusive, with low-intensity exercises being recommended for patients with RA. (16)

All 3 main categories of exercises and physical activity are widely used and are effective for the management of patients with OA. The efficacy of exercises is mainly based on the results of short-term RCTs, RCTs that are relatively good quality, considering that exercise is a physical intervention and thus blinding is an issue. (159) Exercises and physical activity are promising interventions for reducing pain and improving functional status, aerobic capacity, and quality of life. (47,48,67,79) They also offer the potential to reduce body weight (160) and to prevent biomechanical problems (161,162) and further joint damage in patients with OA. (163)

Information on the long-term effect of the exercise program and specifications of the therapeutic application (intensity and dosage) are lacking. Researchers believe that the long-term efficacy of exercises for patients with OA (with or without other interventions) is influenced by a variety of factors, including physiological, biomechanical, psychosocial, and environmental factors. (161,163,164) Thus, researchers (157,158,162,165,166) have suggested multidimensional clinical management of patients with OA due to the multivariate nature of the disease.

Patients with OA tend to adopt sedentary lifestyles. (167) The main challenge is to find effective strategies to help these patients adopt and sustain regular physical activity habits so that they can benefit from the positive effects and avoid the negative consequences and vicious cycle of inactivity. Inactivity can lead to chronic comorbidity problems (eg, obesity, cardiovascular conditions, diabetes) that affect joint health, functional status, and quality of life in patients with OA. (126,168,169) Change in lifestyle among patients with OA is necessary to promote sustained physical activity. (164,170) Fortunately, the level of participation in regular aerobic physical activity can be modified through behavioral interventions. (171-173)

The identification of predisposing, enabling, and reinforcing factors (174) for increasing the level of participation in regular physical activity and exercise is essential. Addressing these factors collectively may increase the likelihood of intrapersonal, interpersonal, and environmental changes that are desirable for sustaining a new behavior regarding TE and physical activity. Predisposing intrapersonal and interpersonal factors include psychological factors (eg, attitude, perceived behavioral control, self-efficacy, motivation, perceived health, expected benefits, depressive symptoms, fear of exercise and of experiencing pain, perceived stress and effort), biological factors (eg, comorbidities, body mass index, smoking status, functional capacity), and demographic factors (eg, age, sex, education). These characteristics may interact with the format of the program and ultimately will determine the success of the physical activity intervention. Enabling factors are factors that affect behavior directly or indirectly through an environmental indicator (163,175) and include the structure of the intervention program, the necessary physical activity skills and equipment, the format of the program (community-based versus clinical setting), the type and frequency of expert supervision and guidance provided, accessibility, time, weather, and the costs incurred by the participant. Reinforcing factors appear subsequent to the change in behavior and provide continuing reward or incentive for the new behavior to be maintained by the individual. Primary reinforcing factors include social support, health practitioner influence, peer influence, feedback from significant others, vicarious reinforcement, incentives, mastery, self-monitoring activity, goal attainment, and enjoyment of the activity. Reinforcing factors will ultimately determine whether the patient continues with the physical activity program and thus will have an impact on the long-term quality of life. Future studies examining the benefits of TE programs in the management of patients with OA will need to identify an effective physical activity program, enhance a sustained physical activity program integrating behavioral interventions, be patient specific, develop a patient education program, and facilitate regular physical activity in the community. (163) The Work Group recommends increasing the awareness of EBCPGs on exercise and physical activity programs among patients with arthritis, practitioners, health care administrators, educators, and policy makers. (157)

Manual Therapy

One study (55) with an acceptable research design was identified. Yet, although the combination of manual therapy and exercise reduces pain in patients with OA, the specific effect of manual therapy could not be determined in that study. Indeed, the reduction of pain after exercise is observed in patients with arthritis in general. (176-178) In the study by Deyle et al, (30) exercise may have contributed to the reduction of pain, but the magnitude was not measured. A recent head-to-head RCT (110) that compared the relative efficacy of manual therapy compared with exercise therapy alone for hip OA showed that manual therapy was significantly more effective than TE for patient global assessment, pain, stiffness, functional status, and ROM after 5 weeks (9 consecutive treatment sessions). Considering these recent scientific results, further research is needed on the individual effects of manual therapy for patients with OA.

2Implications for Practice

The Ottawa Panel has found evidence to recommend and support the use of TE (on their own or combined with manual therapy), especially strengthening exercises and general physical activity, for patients with OA, particularly for the management of pain and improvement of functional status. These recommendations are limited by methodological considerations, such as the relatively good quality, but generally poorly reported description, of TE programs and the selection of outcomes of the included primary trials.

References

(1) Helewa A, Walker JM. Epidemiology and economics of arthritis. In: Walker JM, Helewa A, eds. Physical Therapy in Rheumatoid Arthritis. Toronto, Ontario, Canada: WB Saunders Co; 1996:9-18.

(2) Centers for Disease Control and Prevention. Arthritis prevalence and activity limitations--United States, 1990. Morbidity and Mortality Weekly Report. 1994;43:433-438.

(3) Badley EM, Wang PP. Arthritis and the aging population: projections of arthritis prevalence in Canada 1991 to 2031. J Rheumatol. 1998;25: 138-144.

(4) Badley EM, Rasooly I, Webster GK. Relative importance of musculoskeletal disorders as a cause of chronic health problems, disability and health care utilization: findings from the 1990 Ontario Health Survey. J Rheumatol. 1994;21:505-514.

(5) Coyte P, Asche C, Croxford R, Chan B. The economic cost of arthritis and rheumatism in Canada. In: Badley EM, Williams JI, eds. Patterns of Health Care in Ontario: Arthritis and Related Conditions. Toronto, Ontario, Canada: Institute for Clinical Evaluative Sciences; 1998:27-34.

(6) Maetzel A, Li LC, Pencharz J, Tomlinson G, Bombadier C; Community Hypertension and Arthritis Project Study Team. The economic burden associated with osteoarthritis, rheumatoid arthritis, and hypertension: a comparative study. Ann Rheum Dis. 2004;63:395-401.

(7) Woolf SH. Practice guidelines: a new reality in medicine, I: recent developments. Arch Intern Med. 1990;150:1811-1818.

(8) Grimshaw JM, Freemantle N, Wallace S, et al. Developing and implementing clinical practice guidelines. Qual Health Care. 1995;4:55-64.

(9) Grilli R, Magrini N, Penna A, et al. Practice guidelines developed by specialty societies: the need for a critical appraisal. Lancet. 2000; 355 (9198):103-106.

(10) Graham ID, Beardall S, Carter AO, et al. What is the quality of drug therapy clinical practice guidelines in Canada? Can Med Assoc J. 2001;165:157-163.

(11) Cluzeau F, Littlejohns P, Grimshaw J, et al. Development and application of a generic methodology to assess the quality of clinical guidelines. Int J Qual Health Care. 1999;11:21-28.

(12) Shaneyfelt T, Mayo-Smith M, Rothwangl J. Are guidelines following guidelines? The methodological quality of clinical practice guidelines in the peer-reviewed medical literature. JAMA. 1999;281: 1900-1905.

(13) Cates JR, Young DN, Guerriero DJ, et al. Evaluating the quality of clinical practice guidelines. J Manipulative Physiol Ther. 2001;24: 170-176.

(14) Cates JR, Young DN, Guerriero DJ, et al. An independent assessment of chiropractic practice guidelines. J Manipulative Physiol Ther. 2003;26:282-286.

(15) Philadelphia Panel Evidence-Based Clinical Practice Guidelines on Selected Rehabilitation Interventions: Overview and Methodology. Phys Ther. 2001;81:1629-1640.

(16) Ottawa Panel Evidence-Based Clinical Practice Guidelines for Therapeutic Exercises in the Management of Rheumatoid Arthritis in Adults. Phys Ther. 2004;84:934-972.

(17) Bell M, McConnell S, Fransen M. Exercise for Treating Osteoarthritis of the Hip or Knee (Cochrane Review) [Update software]. Oxford, United Kingdom: The Cochrane Library; 2003:4.

(18) Brosseau L, MacLeay L, Robinson V, et al. Intensity of Exercise for Osteoarthritis: A Meta-analysis (Cochrane Review) [Update software]. Oxford, United Kingdom: The Cochrane Library; 2004:2.

(19) van Baar ME, Assendelft WJJ, Dekker J, et al. Effectiveness of exercise therapy in patients with osteoarthritis of the hip or knee: a systematic review of randomized clinical trials. Arthritis Rheum. 1999; 42:1361-1369.

(20) Pelland L, Brosseau L, Casimiro L, et al. Efficacy of strengthening exercise for osteoarthritis, part I: a meta-analysis. Physical Therapy Reviews. 2004;9:77-108.

(21) Brosseau L, Pelland L, Casimiro L, et al. Efficacy of fitness exercise for osteoarthritis, part II: a meta-analysis. Physical Therapy Reviews. 2004;10:125-131.

(22) Dekker J, Mulder PH, Bijlsma JWJ, Oostendorp RAB. Exercise therapy in patients with rheumatoid arthritis and osteoarthritis: a review. Adv Behav Res Ther. 1993;15:211-238.

(23) Marks R. Quadriceps strength training for osteoarthritis of the knee: a literature review and analysis. Physiotherapy. 1993;79: 13-18.

(24) La Mantia K, Marks R. The efficacy of aerobic exercise for treating osteoarthritis of the knee. New Zealand Journal of Physiotherapy. 1995; 23(2):23-30.

(25) Minor MA. Physical activity and management of arthritis. Ann Behav Med. 1991;13:117-124.

(26) Ytterberg SR, Mahowald ML, Krug HE. Exercise for arthritis. Baillieres Best Pract Res Clin Rheumatol. 1994;8:161-189.

(27) Semble EL, Loeser RF, Wise CM. Therapeutic exercise for rheumatoid arthritis and osteoarthritis. Semin Arthritis Rheum. 1990;20: 32-40.

(28) Westby MD. A health professional's guide to exercise prescription for people with arthritis: a review of aerobic fitness activities. Arthritis Rheum. 2001;45:501-511.

(29) Stenstrom CH, Minor MA. Evidence for the benefit of aerobic and strengthening exercise in rheumatoid arthritis. Arthritis Rheum. 2003; 49:428-434.

(30) Deyle GD, Henderson NE, Matekel RL, et al. Effectiveness of manual physical therapy and exercise in osteoarthritis of the knee: a randomized, controlled trial. Ann Intern Med. 2000;132:173-181.

(31) American College of Rheumatology Subcommittee on Osteoarthritis Guidelines. Recommendations for the medical management of osteoarthritis of the hip and knee: 2000 update. Arthritis Rheum. 2000;43:1905-1915.

(32) American Geriatrics Society Panel on Exercise and Osteoarthritis. Exercise prescription for older adults with osteoarthritis pain: consensus practice recommendations. J Am Geriatr Soc. 2001;49:808-823.

(33) Guidelines for the Management of Pain, Osteoarthritis, Rheumatoid Arthritis, and Juvenile Chronic Arthritis: Clinical Practice Guideline #2. Glenview, III: American Pain Society; 2002:96-109.

(34) British Medical Journal Clinical Evidence: A Compendium of the Best Available Evidence for Effective Health Care. 9th ed. London, United Kingdom: BMJ Books; 2003.

(35) Ontario Program for Optimal Therapeutics. Ontario Treatment Guidelines for Osteoarthritis, Rheumatoid Arthritis, and Acute Musculoskeletal Injury. Toronto, Ontario, Canada: Musculoskeletal Therapeutics Review Panel; 2000.

(36) Philadelphia Panel Evidence-Based Clinical Practice Guidelines on Selected Rehabilitation Interventions for Knee Pain. Phys Ther. 2001; 81:1675-1700.

(37) Anonymous. OMERACT: Conference on Outcome Measures in Rheumatoid Arthritis Clinical Trials. J Rheumatol. 1993;20:526-591.

(38) Klippel JH, Weyand CM, Wortmann RL. Primer on the Rheumatic Diseases. 11th ed. Atlanta, Ga: Arthritis Foundation; 1997.

(39) Philadelphia Panel Evidence-Based Clinical Practice Guidelines on Selected Rehabilitation Interventions for Low Back Pain. Phys Ther. 2001;81:1641-1674.

(40) Philadelphia Panel Evidence-Based Clinical Practice Guidelines on Selected Rehabilitation Interventions for Neck Pain. Phys Ther. 2001; 81:1701-1717.

(41) D'Lima DD, Colwell CW Jr, Morris BA, et al. The effect of preoperative exercise on total knee replacement outcomes. Clin Orthop. 1996;326:174-182.

(42) Topp R, Woolley S, Hornyak J Jr, et al. The effect of dynamic versus isometric resistance training on pain and functioning among adults with osteoarthritis of the knee. Arch Phys Med Rehabil. 2002;83: 1187-1195.

(43) Frank C, Akeson WH, Woo SLY, et al. Physiology and therapeutic value of passive joint motion. Clin Orthop. 1984;185:113-125.

(44) Green J, McKenna F, Redfern EJ, Chamberlain MA. Home exercises are as effective as outpatient hydrotherapy for osteoarthritis of the hip. Br J Rheumatol. 1993;32:812-815.

(45) Messier SP, Thompson CD, Ettinger MH. Effects of long-term aerobic or weight training regimens on gait in an older, osteoarthritic population. J Appl Biomech. 1997;13:202-225.

(46) Callaghan MJ, Oldham JA, Hunt J. An evaluation of exercise regimes for patients with osteoarthritis of the knee: a single-blind randomized controlled trial. Clin Rehabil. 1995;9:213-218.

(47) Evcik D, Sonel B. Effectiveness of a home-based exercise therapy and walking program on osteoarthritis of the knee. Rheumatol Int. 2002;22:103-106.

(48) Minor MA, Hewett JE, Webel RR, et al. Efficacy of physical conditioning exercise in patients with rheumatoid arthritis and osteoarthritis. Arthritis Rheum. 1989;32:1396-1405.

(49) Sashika H, Matsuba Y, Wanabe Y. Home program of physical therapy: effect on disabilities of patients with total hip arthroplasty. Arch Phys Med Rehabil. 1996;77:273-277.

(50) MacDonald SJ, Bourne RB, Rorabeck CH, et al. Prospective randomized clinical trial of continuous passive motion after total knee arthroplasty. Clin Orthop. 2000;380:30-35.

(51) Chen B, Zimmerman JR, Soulen L, DeLisa JA. Continuous passive motion after total knee arthroplasty: a prospective study. Am J Phys Med Rehabil. 2000;79:421-426.

(52) Chiarello CM, Gundersen L, O'Halloran T. The effect of continuous passive motion duration and increment on range of motion in total knee arthroplasty patients. J Orthop Sports Phys Ther. 1997;25:127-129.

(53) Colwell CW, Morris BA. The influence of continuous passive motion on the results of total knee arthroplasty. Clin Orthop. 1992;276:225-228.

(54) Harms M, Engstrom B. Continuous passive motion as an adjunct to treatment in the physiotherapy management of the total knee arthroplasty patient. Physiotherapy. 1991;77:301-307.

(55) Kumar PJ, McPherson EJ, Dorr LD, et al. Rehabilitation after total knee arthroplasty: a comparison of 2 rehabilitation techniques. Clin Orthop. 1996;331:93-101.

(56) McInnes J, Larson MG, Daltroy LH, et al. A controlled evaluation of continuous passive motion in patients undergoing total knee arthroplasty. JAMA. 1992;268:1423-1428.

(57) Montgomery F, Eliasson M. Continuous passive motion compared to active physical therapy after knee arthroplasty: similar hospitalization times in a randomized study of 68 patients. Acta Orthop Scand. 1996;67:7-9.

(58) Nielsen PT, Rechnagel K, Nielsen SE. No effect of continuous passive motion after arthroplasty of the knee. Acta Orthop Scand. 1988;59:580-581.

(59) Pope RO, Corcoran S, McCaul K, Howie DW. Continuous passive motion after primary total knee arthroplasty: does it offer any benefits? J Bone Joint Surg Br. 1997;79:914-917.

(60) Vince KG, Kelly MA, Beck J, Insall JN. Continuous passive motion after total knee arthroplasty. J Arthroplasty. 1987;2:281-284.

(61) Walker RH, Morris BA, Angulo DL, et al. Postoperative use of continuous passive motion, transcutaneous electrical nerve stimulation, and continuous cooling pad following total knee arthroplasty. J Arthroplasty. 1991;6:151-156.

(62) Brosseau L, Milne S, Wells GA, et al. Efficacy of continuous passive motion following total knee arthroplasty: a meta-analysis. J Rheumatol. 2004;31:2251-2264.

(63) Bautch JC, Malone DG, Vailas AC. Effects of exercise on knee joints with osteoarthritis: a pilot study of biologic markers. Arthritis Care Res. 1997; 10:48-55.

(64) Borjesson M, Robertson E, Weidenhielm L, et al. Physiotherapy in knee osteoarthritis: effect on pain and walking. Physiother Res Int. 1996;1 (2):89-97.

(65) Fransen M, Crosbie J, Edmonds J. Physical therapy is effective for patients with osteoarthritis of the knee: a randomized controlled clinical trial. J Rheumatol. 2001;28:156-164.

(66) Garfinkel MS, Schumacher HR Jr, Husain AJ, et al. Evaluation of a yoga based regimen for treatment of osteoarthritis of the hands. J Rheumatol. 1994;21:2341-2343.

(67) Gur H, Cakin N, Akova B, et al. Concentric versus combined concentric-eccentric isokinetic training: effects on functional capacity and symptoms in patients with osteoarthritis of the knee. Arch Phys Med Rehabil. 2002;83:308-316.

(68) Hurley MV, Scott DL. Improvements in quadriceps sensorimotor function and disability of patients with knee osteoarthritis following a clinically practicable exercise regime. Br J Rheumatol. 1998;37: 1181-1187.

(69) Kovar PA, Allegrante JP, MacKenzie CR, et al. Supervised fitness walking in patients with osteoarthritis of the knee: a randomized, controlled trial. Ann Intern Med. 1992;116:529-534.

(70) Kreindler H, Lewis CB, Rush S, Schaefer K. Effects of three exercise protocols on strength of persons with osteoarthritis of the knee. Topics in Geriatric Rehabilitation. 1989;4:32-39.

(71) O'Reilly SC, Muir KR, Doherty M. Effectiveness of home exercise on pain and disability from osteoarthritis of the knee: a randomised controlled trial. Ann Rheum Dis. 1999;58:15-19.

(72) Peloquin L, Bravo G, Gauthier P, et al. Effects of a cross-training exercise program in persons with osteoarthritis of the knee: a randomized controlled trial. Journal of Clinical Rheumatology. 1999;5:126-136.

(73) Penninx BWJH, Messier SP, Rejeski WJ, et al. Physical exercise and the prevention of disability in activities of daily living in older persons with osteoarthritis. Arch Intern Med. 2001;161:2309-2316.

(74) Peterson MGE, Kovar-Toledano PA, Otis JC, et al. Effect of a walking program on gait characteristics in patients with osteoarthritis. Arthritis Care Res. 1993;6:11-16.

(75) Petrella RJ. Is exercise effective treatment for osteoarthritis of the knee? Br J Sports Med. 2000;34:326-331.

(76) Rejeski WJ, Ettinger WHJr, Martin K, Morgan T. Treating disability in knee osteoarthritis with exercise therapy: a central role of self-efficacy and pain. Arthritis Care Res. 1998;11:94-101.

(77) Rodgers JA, Garvin KL, Walker CW, et al. Preoperative physical therapy in primary total knee arthroplasty. J Arthroplasty. 1998;13: 414-421.

(78) Rogind H, Bibow-Nielson B, Jensen B, et al. The effects of a physical training program on patients with osteoarthritis of the knees. Arch Phys Med Rehabil. 1998;79:1421-1427.

(79) Schilke JM, Johnson GO, Housh TJ, O'Dell JR. Effects of muscle-strength training on the functional status of patients with osteoarthritis of the knee joint. Nurs Res. 1996;45:68-72.

(80) Stature TA, Machold KP, Smolen JS, et al. Joint protection and home hand exercises improve hand function in patients with hand osteoarthritis: a randomized controlled trial. Arthritis Rheum. 2002;47: 44-49.

(81) Suomi R, Lindauer S. Effectiveness of arthritis foundation aquatic program on strength and range of motion in women with arthritis. Journal of Aging and Physical Activity. 1997;5:341-351.

(82) van Baar ME, Dekker J, Oostendorp RAB, et al. The effectiveness of exercise therapy in patients with osteoarthritis of the hip or knee: a randomized clinical trial. J Rheumatol. 1998;25:2432-2439.

(83) van Baar ME, Dekker J, Oostendorp RAB, et al. Effectiveness of exercise in patients with osteoarthritis of the hip or knee: nine months follow-up. Ann Rheum Dis. 2001;60:1123-1130.

(84) Ahern M, Nicholls E, Simionata E, et al. Clinical and psychological effects of hydrotherapy in rheumatic diseases. Clin Rehab. 1995;9: 204-212.

(85) Aubriot JH, Guincestre JY, Grandbastien B. Interet des appareils asthromoteurs dans la reeducation precoce des arthroplastics totales de genou: etude prospective a propos de 120 dossiers [Rehabilitation following total knee arthroplasty: role of passive motion--a randomised study about 120 subjects]. Rev Chir Orthop Reparatrice Appar Mot. 1993;79:586-590.

(86) Baker KR, Nelson ME, Felson DT, et al. The efficacy of home based progressive strength training in older adults with knee osteoarthritis: a randomized controlled trial. J Rheumatol. 2001;28:1655-1665.

(87) Balint G, Szebenyi B. Non-pharmacological therapies in osteoarthritis. Baillieres Clin Best Pract Res Rheumatol. 1997;11:795-815.

(88) Basso DM, Knapp L. Comparison of two continuous passive motion protocols for patients with total knee implants. Phys Ther. 1987;67: 360-363.

(89) Beaupre LA, Davies DM, Jones CA, Cinats JG. Exercise combined with continuous passive motion or slider board therapy compared with exercise only: a randomized controlled trial of patients following total knee arthroplasty. Phys Ther. 2001;81:1029-1037.

(90) Belza B, Topolski T, Kinne S, et al. Does adherence make a difference? Results from a community-based aquatic program. Nurs Res. 2002;51:285-291.

(91) Beverley C. Evidence-based practice digests: topic--exercise as a treatment for osteoarthritis. Journal of Clinical Excellence. 2001;2:249-253.

(92) Boardman ND III, Cofield RH, Bengston KA, et al. Rehabilitation after total shoulder arthroplasty. J Arthroplasty. 2001;16:483-486.

(93) Bunning RD, Materson RS. A rational program of exercise for patients with osteoarthritis. Semin Arthritis Rheum. 1991;21(3 suppl 2):33-43.

(94) Burke JH, Grady JH, de Vries J, Baten CTM. Usability of thenar eminence orthoses: report of a comparative study. Clin Rehabil. 1999; 13:288-294.

(95) Chamberlain MA, Care G, Harfield B. Physiotherapy in osteoarthrosis of the knees: a controlled trial of hospital versus home exercises. Int Rehabil Med. 1982;4:101-106.

(96) Davis D. Continuous passive motion for total knee arthroplasty [abstract]. Phys Ther. 1984;64:709.

(97) Dougados M, Ravaud P. Exercise therapy in patients with osteoarthritis of the hip or knee. Current Rheumatology Reports. 2001;3:353-354.

(98) Ettinger WH Jr, Burns R, Messier SP, et al. A randomized trial comparing aerobic exercise and resistance exercise with a health education program in older adults with knee osteoarthritis: the Fitness Arthritis and Seniors Trial (FAST). JAMA. 1997;277:25-31.

(99) Eungpinichpong W. The efficacy of physical exercise programs for patients with osteoarthritis of the knee as determined by clinical and gait parameters. New Zealand Journal of Physiotherapy. 1998;26(3):5.

(100) Fisher NM, Pendergast DR, Gresham GE, Calkins E. Muscle rehabilitation: its effect on muscular and functional performance of patients with knee osteoarthritis. Arch Phys Med Rehabil. 1991;72: 367-374.

(101) Fisher NM, Gresham G, Pendergast DR. Effects of a quantitative progressive rehabilitation program applied unilaterally to the osteoarthritic knee. Arch Phys Med Rehabil. 1993;74:1319-1326.

(102) Fransen M, Margiotta E, Crosbie J, Edmonds J. A revised group exercise program for osteoarthritis of the knee. Physiother Res Int. 1997;2:30-41.

(103) Frost H, Lamb SE, Robertson S. A randomized controlled trial of exercise to improve mobility and function after elective knee arthroplasty: feasibility, results, and methodological difficulties. Clin Rehabil. 2002;16:200-209.

(104) Gerber LH. Exercise and arthritis. Bull Rheum Dis. 1990;39(6):1-9.

(105) Goletz TH, Henry JH. Continuous passive motion after total knee arthroplasty. South Med J. 1986;79:1116-1120.

(106) Gose JC. Continuous passive motion in the postoperative treatment of patients with total knee replacement: a retrospective study. Phys Ther. 1987;67:39-42.

(107) Hall KD, Hayes KW, Falconer J. Differential strength decline in patients with osteoarthritis of the knee: revision of a hypothesis. Arthritis Care Res. 1993;6:89-96.

(108) Hartman CA, Manos TM, Winter C, et al. Effects of Tai Chi training on function and quality of life indicators in older adults with osteoarthritis. J Am Geriatr Soc. 2000;48:1553-1559.

(109) Haug J, Wood LT. Efficacy of neuromuscular stimulation of the quadriceps femoris during continuous passive motion following total knee arthroplasty. Arch Phys Med Rehabil. 1988;69:423-424.

(110) Hoeksma HL, Dekker J, Ronday HK, et al. Comparison of manual therapy and exercise therapy in osteoarthritis of the hip: a randomized clinical trial. Arthritis Care Res. 2004;51:722-729.

(111) Hopman-Rock M, Westhoff MH. The effects of a health educational and exercise program for older adults with osteoarthritis of the hip or knee. J Rheumatol. 2000;27:1947-1954.

(112) Johnson DP. The effect of continuous passive motion on wound-healing and joint mobility after knee arthroplasty. J Bone Joint Surg Am. 1990;72:421-426.

(113) Johnson DP, Eastwood DM. Beneficial effects of continuous passive motion after total condylar knee arthroplasty. Ann R Coll Surg Engl. 1992;74:412-416.

(114) Kim J-M, Moon M-S. Squatting following total knee arthroplasty. Clin Orthop. 1995;313:177-186.

(115) Langeland N. Conservative treatment with active exercises in osteoarthritis of the hip. Acta Orthop Scand. 1972;43:118-125.

(116) Lau SK, Chiu KY. Use of continuous passive motion after total knee arthroplasty. J Arthroplasty. 2001;16:336-339.

(117) Leivseth G, Tortsensson J, Reikeras O. Effect of passive muscle stretching in osteoarthritis of the hip. Clin Sci. 1989;76:113-117.

(118) Lynch AF, Bourne RB, Rorabeck CH, et al. Deep-vein thrombosis and continuous passive motion after total knee arthroplasty. J Bone Joint Surg Am. 1988;70:11-14.

(119) Maloney WJ, Schurmann DJ, Hangen D, et al. The influence of continuous passive motion on outcome in total knee arthroplasty. Clin Orthop. 1990;256:162-168.

(120) Mangione KK, Axen K, Haas F. Mechanical unweighting effects on treadmill exercise and pain in elderly people with osteoarthritis of the knee. Phys Ther. 1996;76:387-394.

(121) Mangione KK, McCully K, Gloviak A, et al. The effects of high-intensity and low-intensity cycle ergometry in older adults with knee osteoarthritis. J Gerontol A Biol Sci Med Sci. 1999;54:M184-M190.

(122) Marks R, Cantin D. Symptomatic osteo-arthritis of the knee: the efficacy of physiotherapy. Physiotherapy. 1997;83:306-312.

(123) May LA, Busse W, Zayak D, Whitridge MR. Comparison of continuous passive motion (CPM) machines and lower limb mobility boards (LLiMB) in the rehabilitation of patients with total knee arthroplasty. Canadian Journal of Rehabilitation. 1999;12:257-263.

(124) Mei-Hwa J, Jin-Shin L. The effects of physiotherapy on osteoarthritic knees of females. Journal of the Formosan Medical Association. 1991;90:1008-1013.

(125) Merchan ECR, de la Corte H. The role of rehabilitation after high tibial osteotomy in patients with medial gonarthrosis. Journal of Orthopaedic Rheumatology. 1993;6:151-153.

(126) Messier SP, Loeser RF, Mitchell MN, et al. Exercise and weight loss in obese older adults with knee osteoarthritis: a preliminary study. J Am Geriatr Soc. 2000;48:1062-1072.

(127) Messier SP, Royer TD, Craven TE, et al. Long-term exercise and its effect on balance in older, osteoarthritic adults: results from fitness, arthritis, and senior trials (FAST). J Am Geriatr Soc. 2000;48: 131-138.

(128) Meyer CL, Hawley DJ. Characteristics of participants in water exercise programs compared to patients seen in a rheumatic disease clinic. Arthritis Care Res. 1994;7:85-89.

(129) Minor MA, Hewett JE, Webel RR, et al. Exercise tolerance and disease related measures in persons with rheumatoid arthritis and osteoarthritis. J Rheumatol. 1988;15:905-911.

(130) Minor MA, Brown JD. Exercise maintenance of persons with arthritis after participation in a class experience. Health Educ Q. 1993;20:83-95.

(131) Nicolakis P, Burak EC, Kollmitzer J, et al. An investigation of the effectiveness of exercise and manual therapy in treating symptoms of TMJ osteoarthritis. Cranio--The Journal of Craniomandibular Practice. 2001;19(1):26-32.

(132) Nicolakis P, Erdogmus CB, Kollmitzer J, et al. Long-term outcome after treatment of temporomandibular joint osteoarthritis with exercise and manual therapy. Cranio--The Journal of Craniomandibular Practice. 2002;20(1):23-27.

(133) Nordesjo LO, Nordgren B, Wigren A, Kolstad BL Isometric strength and endurance in patients with severe rheumatoid arthritis or osteoarthritis in the knee joints: a comparative study of healthy men and women. Scand J Rheumatol. 1983;12:152-156.

(134) Odenbring S, Lindstrand A, Egund N. Early knee mobilization after osteotomy for gonarthrosis. Acta Orthop Scand. 1989;60:699-702.

(135) Penninx BWJH, Rejeski WJ, Miller ME, et al. Exercise and depressive symptoms: a comparison of aerobic and resistance exercise effects on emotional and physical function in older persons with high and low depressive symptomatology. J Gerontol. 2002;57B(2):P124-P132.

(136) Petrella RJ, Bartha C. Home based exercise therapy for older patients with knee osteoarthritis: a randomized clinical trial. J Rheumatol. 2000;27:2215-2221.

(137) Rao A, Evans MF. Does a structured exercise program benefit elderly people with knee osteoarthritis? Canadian Family Physician. 1998;44:283-284.

(138) Rasti Z, Olsen O. Continuous Passive Motion for Rehabilitation After Total Knee Arthroplasty in Patients With Osteoarthritis (Protocol for a Cochrane Review) [Update software]. Oxford, United Kingdom: The Cochrane Library. 2001:4.

(139) Rejeski WJ, Brawley LR, Ettinger W, et al. Compliance to exercise therapy in older participants with knee osteoarthritis: implications for treating disability. Med Sci Sports Exerc. 1997;29:977-985.

(140) Rejeski WJ, Brawley LR, Ettinger W, et al. Compliance to exercise therapy in treating seniors with knee osteoarthritis. Clin J Sport Med. 1998;8:148.

(141) Simkin PA, de Lateur BJ, Alquist AD, et al. Continuous passive motion for osteoarthritis of the hip: a pilot study. J Rheumatol. 1999; 26:1987-1991.

(142) Stenstrom CH. Therapeutic exercise in rheumatoid arthritis. Arthritis Care Res. 1994;7:190-197.

(143) Sullivan T, Allegrante JP, Peterson MGE, et al. One-year follow-up of patients with osteoarthritis of the knee who participated in a program of supervised fitness walking and supportive patient education. Arthritis Care Res. 1998;11:228-233.

(144) Sylvester KL. Investigation of the effect of hydrotherapy in the treatment of osteoarthitis hips. Clin Rehabil. 1989;4:223-228.

(145) Tan J, Balci N, Sepici V, Gener FA. Isokinetic and isometric strength in osteoarthritis of the knee: a comparative study with healthy women. American Journal of Medicine and Rehabilitation. 1995;74: 364-369.

(146) Thomas KS, Muir KR, Doherty M, et al. Home based exercise programme for knee pain and knee osteoarthritis: a randomised controlled trial. BMJ. 2002;325:752-755.

(147) Tork SC, Douglas V. Arthritis water exercise program evaluation: a self-assessment survey. Arthritis Care Res. 1989;2(1):28-30.

(148) Ververeli PA, Sutton DC, Hearn SL, et al. Continuous passive motion after total knee arthroplasty: analysis of costs and benefits. Clin Orthop. 1995;321:208-215.

(149) Wasilewski SA, Woods LC, Torgerson WR Jr, Healy WL. Value of continuous passive motion in total knee arthroplasty. Orthopedics. 1990;13:291-295.

(150) Weiss S, LaStayo P, Mills A, Bramlet D. Prospective analysis of splinting the first carpometacarpal joint: an objective, subjective, and radiographic assessment. Journal of Hand Therapy. 2000;13:218-226.

(151) Worland RL, Arredondo J, Angles F, et al. Home continuous passive motion machine versus professional physical therapy following total knee replacement. J Arthroplasty. 1998;13:784-787.

(152) Yashar AA, Venn-Watson E, Welsh T, et al. Continuous passive motion with accelerated flexion after total knee arthroplasty. Clin Orthop. 1997;345:38-43.

(153) Young JS, Kroll MA. Continuous passive motion compared to active assisted range of motion [abstract]. Phys Ther. 1984;64:721.

(154) Altman R, Alarcon G, Appelrouth D, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hand. Arthritis Rheum. 1990;33:1601-1610.

(155) Cluzeau F, Littlejohns P. Appraising clinical practice guidelines in England and Wales: the development of a methodologic framework and its application to policy. Jt Comm J Qual Improv. 1999;25:514-521.

(156) Milne S, Brosseau L, Noel MJ, Robinson V, et al. Continuous Passive Motion Following Total Knee Arthroplasty (Cochrane Review) [Update software]. Oxford, United Kingdom: The Cochrane Library; 2004:2.

(157) Chang R, Roubenoff R, Mayer J, et al. Work Group Recommendations: 2002 Exercise and Physical Activity Conference, St Louis, Missouri--Session IV: exercise in the presence of rheumatic disease. Arthritis Rheum. 2003;49:280.

(158) Minor M, Stenstrom CH, Klepper SE, et al. Work Group Recommendations: 2002 Exercise and Physical Activity Conference, St Louis, Missouri--Session V: evidence of benefit of exercise and physical activity in arthritis. Arthritis Rheum. 2003;49:453-454.

(159) Deyo RA, Walsh NE, Schoenfeld LS, Ramamurthy S. Can trials of physical treatments be blinded? The example of transcutaneous electrical nerve stimulation for chronic pain. Am J Phys Med Rehabil. 1990;69:6-10.

(160) Pencharz JN, Grigoriadis E, Jansz GF, Bombardier C. A critical appraisal of clinical practice guidelines for the treatment of lower-limb osteoarthritis. Arthritis Research. 2002;4:1-9.

(161) Sharma L. Examination of exercise effects on knee osteoarthritis outcomes: why should the local mechanical environment be considered? Arthritis Rheum. 2003;49:255-260.

(162) Krebs D, Herzog W, McGibbon CA, Sharma L. Work Group Recommendations: 2002 Exercise and Physical Activity Conference, St Louis, Missouri--Session III: biomechanical considerations for exercise. Arthritis Rheum. 2003;49:261,280.

(163) Hurley MV. Muscle dysfunction and effective rehabilitation of knee osteoarthritis: what we know and what we need to find out. Arthritis Rheum. 2003;49:444-452.

(164) Eyler AA. Correlates of physical activity: who's active and who's not? Arthritis Rheum. 2003;49:136-140.

(165) Sniezek JE, Macera CA, Hootman JM, Eyler AA. Work Group Recommendations: 2002 Exercise and Physical Activity Conference, St Louis, Missouri--Session II: the problem and challenge of inactivity. Arthritis Rheum. 2003;49:141.

(166) Meenan R, Sharpe P, Boutaugh M, Brady T. Work Group Recommendations: 2002 Exercise and Physical Activity Conference, St Louis, Missouri--Session VI: population approaches to health promotion and disability prevention through physical activity. Arthritis Rheum. 2003;49: 477.

(167) Hootman JM, Macera CA, Ham SA, et al. Physical activity levels among the general US adult population and in adults with and without arthritis. Arthritis Rheum. 2003;49:129-135.

(168) Coggon D, Reading I, Croft P, et al. Knee osteoarthritis and obesity. Int J Obes Relat Metab Disord. 2001;25:622-627.

(169) Huang M-H, Chen C-H, Chen T-W, et al. The effects of weight reduction on the rehabilitation of patients with knee osteoarthritis and obesity. Arthritis Care Res. 2000;13:398-405.

(170) Macera CA, Hootman JM, Sniezek JE. Major public health benefits of physical activity. Arthritis Rheum. 2003;49:122-128.

(171) Baranowski T, Anderson C, Carmack C. Mediating variable framework in physical activity interventions. How are we doing? How might we do better? Am J Prev Med. 1998;15:266-297.

(172) Sallis J, Owen N. Ecological models. In: Glanz K, Lewis F, Rimer B, eds. Health Behavior and Health Education: Theory, Research, and Practice. San Francisco, Calif: Jossey-Bass; 1997:403-424.

(173) McLeroy KR, Bibeau D, Steckler A, Glanz K. An ecological perspective on health promotion programs. Health Educ Q. 1988;15: 351-377.

(174) Green LW, Kreuser MW. Health Promotion Planning: An Educational and Environmental Approach. 2nd ed. Mountain View, Calif: Mayfield Publishing Co; 1991.

(175) Sharpe PA. Community-based physical activity intervention. Arthritis Rheum. 2003;49:455-462.

(176) Jones CA, Rees JM, Dodds WN, Jayson MI. Changes in plasma opioid concentrations after physiotherapeutic exercises for arthritic patients. Neuropeptides. 1985;5:561-562.

(177) Jonsdottir IH, Hoffmann P, Thoren P. Physical exercise, endogenous opioids and immune function. Acta Physiol Scand Suppl. 1997;640: 47-50.

(178) Kangilaski J. Beta-endorphin levels lower in arthritis patients. JAMA. 1981;246:203.

* Cybex International Inc, 10 Trotter Dr, Medway, MA 02053.

Ottawa Panel Members:

Ottawa Methods Group:

Lucie Brosseau, PhD, University Research Chair in Evidence-Based Practice in Rehabilitation, Physiotherapy Program, School of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa, Ottawa, Ontario, Canada

George A Wells, PhD, Department of Epidemiology and Community Medicine, University of Ottawa

Peter Tugwell, MD, MSc, Centre for Global Health, Institute of Population Health, University of Ottawa

Mary Egan, PhD, Occupational Therapy Program, School of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa

Claire-Jehanne Dubouloz, PhD, Occupational Therapy Program, School of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa

Lynn Casimiro, MA, Physiotherapy Program, School of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa

Vivian A Robinson, MSc, Centre for Global Health, Institute of Population Health

Lucie Pelland, PhD, Physiotherapy Program, School of Rehabilitation Sciences, Queens' University, Kingston, Ontario, Canada

Jessie McGowan, MLIS, Director, Medical Library, Centre for Global Health, Institute of Population Health, University of Ottawa

Maria Judd, PT, MSc, Canadian Physiotherapy Association, Ottawa, Ontario, Canada

Sarah Milne, PT, MSc, Department of Epidemiology and Community Medicine, University of Ottawa

External Experts:

Mary Bell, MD (Rheumatologist), Sunnybrook and Women's College Health Sciences Centre, Toronto, Ontario, Canada

Hillel M Finestone, MD (Physiatrist), Sisters of Charity of Ottawa Health Service, Ottawa, Ontario, Canada

France Legare, MD (Evidence-Based Practice in Family Medicine), University of Laval, Quebec City, Quebec, Canada

Catherine Caron, MD (Family Physician), Sisters of Charity of Ottawa Health Service

Sydney Lineker, PT, MSc, The Arthritis Society, Ontario Division, Research Co-ordinator, Toronto, Ontario, Canada

Angela Haines-Wangda, PT, MSc, Ottawa Hospital, General Campus, Ottawa, Ontario, Canada

Marion Russell-Doreleyers, PT who practices acupuncture, MSc, Canadian Physiotherapy Association and Ottawa Arthritis Rehabilitation and Education Program, Ottawa, Ontario, Canada

Martha Hall, OT, MPA, Canadian Association of Occupational Therapists and Ottawa Arthritis Rehabilitation and Education Program

Gerry Arts, person with osteoarthritis (named with her written permission)

Assistant Manuscript Writer:

Marnie Lamb, MA, School of Rehabilitation Sciences, Faculty of Health Sciences, University of Ottawa

Address all correspondence and requests for reprints to: Lucie Brosseau, PhD, Physiotherapy Program, School of Rehabilitation Sciences, Faculty of Health Sciences, 451 Smyth Rd, University of Ottawa, Ottawa, Ontario, Canada K1H 8M5 ([e-mail]).

This study was financially supported by The Arthritis Society (Canada) (Grant TAS-319); the Ontario Ministry of Health and Long-Term Care (Canada) (Grant HRPD-05225); the Career Scientist Salary Support Program (HRPD-05225), for Dr Brosseau; the University Research Chair Program, for research staff salary support for Dr Brosseau; the Centre National de Formation en Sante/Health Canada; and the Ministry of Human Resources, Summer Students Program, Government of Canada.

Acknowledgments: The Ottawa Panel is indebted to Ms Catherine Lamothe, Ms Gabriele Wieschollek, Ms Judith Robitaille, Ms Lucie Lavigne, Mr Michel Boudreau, Mr Guillaume Michaud, Ms Michelle Vaillant, Ms Chantal Lavoie, and Mr Guillaume Lemieux for their technical support and help in data extraction.

Table 1.

Inclusion and Exclusion Criteria for the Ostecorthritis Project (a)

Inclusion                            Exclusion

Study Designs                        Study Designs

* Randomized controlled trial        * Case series/case report
* Controlled clinical trial          * Uncontrolled cohort studies
* Cohort study                       * Data (graphic) without a mean
* Case-control study                   and SD
* English and French articles only   * Sample size of fewer than 5
  Population                           patients per treatment group
* Outpatients/inpatients             * Studies with more than 20%
* OA of all human joints               dropout rate Population
  (including temporomandibular       * Cancer (and other oncologic
  joint) except vertebral column       conditions)
* Chronic and acute conditions       * Cardiac conditions
* Age groups >18 y                   * Dermatologic conditions
                                     * No known pathology or
                                       impairments
                                     * Juvenile arthritis
                                     * Mixed population (other than OA
                                       and RA)
                                     * Multiple conditions
                                     * Neurologic conditions
                                     * Other rheumatologic or
                                       musculoskeletal conditions
                                     * Pediatric conditions (no
                                       juvenile arthritis)
                                     * Psychiatric conditions
                                     * Pulmonary conditions
                                     * Scoliosis

Intervention                         Intervention

* Eligible control groups: place-    * Bilateral interventions (if
  bo, untreated, sham, routine         systemic effects)
  conventional therapy, active       * Neck and back interventions
  physical therapy treatments, and   * Multidisciplinary, functional
  educational pamphlets (no            restoration programs
  surgery, drugs, or injections)     * Surgery of shoulder, knee, neck,
  * Eligible interventions:            or low back
  1. Chiropractic interventions      * Medication leg, phonophoresis
     (manipulation, mobilization,      with medications)
     manual therapy)                 * Thermal biofeedback
  2. Intensity of rehabilitation     * Psychosocial interventions
  3. Therapeutic exercises,          * Therapeutic exercises, including
     including swimming pool           postsurgery and CPM
     exercise

Outcomes                             Outcomes

* Absenteeism, sick leave, return    * Biochemical measures
  to work (if available)             * Patient adherence to medication
* Balance status                     * Psychosocial measures
* Cardiopulmonary functions            (depression, home and community
* Coordination status                  activities, leisure, social
* Costs (economics)                    roles, sexual functions)
* Disease activity                   * Serum markers (except ESR)
* Edema
* EMG activity
* Functional status, activities of
  daily living (self-care
  activities)
* Gait status
* Girth, volume
* Inflammation
* Joint imaging
* Medication intake (if reported)
* Muscle force, endurance, and
  power
* Pain
* Patient satisfaction
* Postural assessment
* Quality of life
* Range of motion, flexibility,
  mobility
* Side effects (if reported)
* Swelling
* Weight loss

(a) CPM=continous passive motion. EMG=electromyographic,
ESR=crythrocyte sedimentation rate, OA=ostctrarthritis, RA=rheumatoid
arthritis, SD=standard deviation.

Table 2.

Excluded Studies for Therapeutic Exercises (N=89) (a)

Study                             Reason for Exclusion

AGS(32)                           Not an RCT
Ahern et al (84)                  No statistical data for control
                                    group
Aubriot et al (85)                No standard deviation
Baker et al (86)                  Head-to-head study
Balint and Szebenyi (87)          Not an RCT
Basso and Knapp (88)              Not an RCT
Beaupre et al (89)                Mixed population
Belza et al (90)                  More than 20% dropout rate
Beverley (91)                     Review
Boardman et al (92)               Mixed population
Bunning and Materson (93)         Review
Burke et al (94)                  No control group
Callaghan et al (46)              Combined interventions
Chamberlain et al (95)            No statistical data
Chen et al (51)                   CPM
Chiarello et al (52)              CPM
Colwell and Morris (53)           CPM
Davis (96)                        Not enough statistical data
D'Lima et al (41)                 Larger proportion of patients
                                    with RA
Dougados and Ravaud (97)          Overview
Ettinger et al (98)               More than 20% dropout rate
Eungpinichpong (99)               No statistical data
Fisher et al (100)                More than 20% dropout rate
Fisher et al (101)                No control group
Frank et al (43)                  Head-to-head study
Fransen et al (102)               More than 20% dropout rate
Frost et al (103)                 More than 20% dropout rate
Gerber (104)                      Review
Goletz and Henry (105)            Not an RCT
Gose (106)                        Not an RCT
Green et al (44)                  Head-to-head study
Hall et al (108)                  No intervention
Harms and Engstrom (54)           CPM
Hartman et al (108)               Majority spine OA
Haug and Wood (109)               Combined electrical stimulation
Hoeksma et al (110)               Head-to-head study
Hopman-Rock and Westhoff (111)    Education on exercise (wrong
                                    intervention)
Johnson (112)                     Head-to-head study
Johnson and Eastwood (113)        No standard deviation
Kim and Moon (114)                Head-to-head study
Kumar et al (55)                  CPM
Langeland (115)                   No control group
Lou and Chiu (116)                Number of patients in each
                                    group missing
Leivseth et al (117)              Biochemical data
Lynch et al (118)                 No standard deviation
MacDonald et al (50)              CPM
Maloney et al (119)               Mixed population
Mangione et al (120)              Patients as their own control
Mangione et al (121)              Wrong intervention
Marks and Cantin (122)            No control group
Mays et al (123)                  Head-to-head study
Mei-Hwa and Jin-Shin (124)        No time period for the
                                    outcomes
Merchan and de la Corte (125)     No statistical data
Messier et al (126)               Nutrition was the main
                                    intervention
Messier et al (127)               Subjects without known
                                    pathology or impairments
Meyer and Hawley (128)            No time period for the
                                    outcomes
McInnes et al (56)                CPM
Minor et al (29)                  No control group
Minor and Brown (130)             No statistical data for control
Montgomery and Eliasson (57)      CPM
Nielsen et al (58)                CPM
Nicolakis et al (131)             No control group
Nicolakis et al (132)             No control group
Nordesjo et al (133)              Subjects without known
                                    pathology or impairments
Odenbring et al (134)             Not subjects with TKA
Penninx et al (135)               No statistical data
Petrella and Bartha (136)         Review
Pope et al (59)                   CPM
Rao and Evans (137)               No significant data
Rasti and Olsen (138)             Literature review
Rejeski et al (139)               Outcome was adherence
Rejeski et al (140)               More than 20% dropout rate
Sashika et al (49)                Postsurgery
Simkin et al (141)                Not enough statistical data
Stenstrom (142)                   Review
Sullivan et al (143)              More than 20% dropout rate
Sylvester (144)                   Head-to-head study
Tan et al (145)                   Subjects without known
                                    pathology or impairments
Thomas et al (146)                Not specific to OA
Tork and Douglas (147)            No control group
van Baar et al (19)               Systematic review
Ververeli et al (148)             Not an RCT
Vince et al (60)                  CPM
Walker et al (61)                 CPM
Wasilewski et al (149)            Not an RCT (retrospective
                                    study)
Weiss et al (150)                 Multiple conditions
Worland et al (151)               Both groups received CPM
Yashar et al (152)                Mixed population
Young and Kroll (153)             Not enough statistical data

(a) AGS=American Geriatrics Societv Panel on Exercise and
Osteoarthritis, RCT=ru:dontizecl controlled trial, CPM=continuous
passive motion, RA=rheumatoid arthritis, OA=osteoarthtitis, TKA=total
knee arthroplasty.

Table 3.

Clinical Relevance: Lower-Extremity Strengthening Versus Control (a)

                      Treatment                                No. of
Study                 Group        Outcome                     Patients

Topp et al (42)       Exercise     Pain, getting up from
                                     floor  (0-16)             35
                      Control      Pain, getting up from
                                     floor  (0-16)             35
Topp et al (42)       Exercise     Pain, getting up from
                                     floor  (0-16)             35
                      Control      Pain, getting up from
                                     floor  (0-16)             35
Topp et al (42)       Exercise     Pain, going up stairs
                                     (0-16)                    35
                      Control      Pain, going up stairs
                                     (0-16)                    35
Topp et al (42)       Exercise     Pain, going down stairs
                                     (0-16)                    35
                      Control      Pain, going down stairs
                                     (0-16)                    35
Topp et al (42)       Exercise     Pain (WOMAC)                35
                      Control      Pain (WOMAC)                35
Topp et al (42)       Exercise     Time to get down to
                                     floor (s)                 35
                      Control      Time to get down to
                                     floor (s)                 35
Topp et al (42)       Exercise     Time to get up off floor
                                     (s)                       35
                      Control      Time to get up off floor
                                     (s)                       35
Topp et al (42)       Exercise     Time to go up stairs (s)    35
                      Control      Time to go up stairs (s)    35
Topp et al (42)       Exercise     Time to go down stairs
                                     (s)                       35
                      Control      Time to go down stairs
                                     (s)                       35
Topp et al (42)       Exercise     Functional limitation
                                     (WOMAC)                   35
                      Control      Functional limitation
                                     (WOMAC)                   35
Schilke et al (79)    Exercise     Peak torque, right knee
                                     extensors                 10
                      Control      Peak torque, right knee
                                     extensors                 10
Schilke et al (79)    Exercise     OASI-mobility               10
                      Control      OASI-mobility               10

                      Treatment    Baseline    End-of-Study    Absolute
Study                 Group        Mean        Mean            Benefit

Topp et al (42)       Exercise      6.44        2.67           -4.27
                      Control       4.53        5.03
Topp et al (42)       Exercise      4.96        2.86           -2.45
                      Control       3.54        3.89
Topp et al (42)       Exercise      5.61        4.03           -1.65
                      Control       4.59        4.66
Topp et al (42)       Exercise      5.30        3.71           -1.52
                      Control       4.47        4.40
Topp et al (42)       Exercise     12.40       10.71           -1.71
                      Control      10.75       10.77
Topp et al (42)       Exercise      4.72        3.89           -1.18
                      Control       4.98        5.33
Topp et al (42)       Exercise      7.16        5.71           -1.57
                      Control       8.04        8.16
Topp et al (42)       Exercise     18.85       16.33           -1.2
                      Control      18.85       17.53
Topp et al (42)       Exercise     19.29       15.96           -1.63
                      Control      18.04       16.34
Topp et al (42)       Exercise     41.09       35.30           -6.62
                      Control      38.87       39.70
Schilke et al (79)    Exercise     52.50       67.40           14.50
                      Control      35.90       36.30
Schilke et al (79)    Exercise     11.30        7.64           -3.24
                      Control       9.90        9.48

                                   Relative
                                   Difference
                                   in Change
                      Treatment    From
Study                 Group        Baseline      WMD (95% CI)

Topp et al (42)       Exercise     -78%          -2.36 (-4.22, -0.5)
                      Control
Topp et al (42)       Exercise     -58%          -1.03 (-2.57, 0.51)
                      Control
Topp et al (42)       Exercise     -32%          -0.63 (-2.32, 1.06)
                      Control
Topp et al (42)       Exercise     -31%          -0.69 (-2.27, 0.89)
                      Control
Topp et al (42)       Exercise     -15%          -0.06 (-1.54, 1.42)
                      Control
Topp et al (42)       Exercise     -24%          -1.44 (-3.15, 0.27)
                      Control
Topp et al (42)       Exercise     -21%          -2.45 (-5.32, 0.27)
                      Control
Topp et al (42)       Exercise      -6%          -1.2 (-4.5, 2.1)
                      Control
Topp et al (42)       Exercise      -9%          -0.38 (-3.57, 2.81)
                      Control
Topp et al (42)       Exercise     -17%          -4.4 (-9.47, 0.67)
                      Control
Schilke et al (79)    Exercise      33%          31.1 (6.4, 55.8)
                      Control
Schilke et al (79)    Exercise       4%          -5.14 (-7.37, -2.91)
                      Control

(a) WOMAC=Western Ontario and McMaster Universities Osteoarthfitis
Index, OASI=Osteoarthritis Screening Index, WMD=weighted mean
difference, CI=confidence interval.

Table 4.

Clinical Relevance: Lower-Extremity Isometric Strengthening Versus
Control (a)

                                                               No. of
Study             Treatment Group         Outcome              Patients

Topp et al (42)   LE isometric            Pain, getting up
                  strengthening Control   from floor (0-16)    32
                                          Pain, getting up
                                          from floor (0-16)    35
Topp et al (42)   LE isometric            Pain, getting down   32
                  strengthening Control   to floor (0-16)
                                          Pain, getting down   35
                                          to floor (0-16)
Topp et al (42)   LE isometric            Pain, going up       32
                  strengthening Control   stairs (0-16)
                                          Pain, going up       35
                                          stairs (0-16)
Topp et al (42)   LE isometric            Pain, going down     32
                  strengthening Control   stairs (0-16)
                                          Pain going down      35
                                          stairs (0-16)
Topp et al (42)   LE isometric            Pain (WOMAC)         32
                  strengthening Control   Pain (WOMAC)         35
Topp et al (42)   LE isometric            Time to get down     32
                  strengthening Control   to floor (s)
                                          Time to get down     35
                                          to floor (s)
Topp et al (42)   LE isometric            Time to get up off   32
                  strengthening Control   floor (s)
                                          Time to get down     35
                                          to floor (s)
Topp et al (42)   LE isometric            Time to go up        32
                  strengthening Control   stairs (s)
                                          Time to go up        35
                                          stairs (s)
Topp et al (42)   LE isometric            Time to go down      32
                  strengthening Control   stairs (s)
                                          Time to go down      35
                                          stairs (s)
Topp et al (42)   LE isometric            Stiffness (WOMAC)    32
                  strengthening Control   Stiffness (WOMAC)    35
Topp et al (42)   LE isometric            Functional           32
                  strengthening Control   limitation (WOMAC)
                                          Functional           35
                                          limitation (WOMAC)

                                                     End-of-
                                          Baseline   Study    Absolute
Study             Treatment Group         Mean       Mean     Benefit

Topp et al (42)   LE isometric             5.39       2.89    -3
                  strengthening Control    4.53       5.03
Topp et al (42)   LE isometric             4.20       1.84    -2.71
                  strengthening Control    3.54       3.89
Topp et al (42)   LE isometric             5.19       2.98    -2.28
                  strengthening Control    4.59       4.66
Topp et al (42)   LE isometric             4.70       2.78    -1.85
                  strengthening Control    4.47       4.40
Topp et al (42)   LE isometric            11.75      10.38    -1.39
                  strengthening Control   10.75      10.77
Topp et al (42)   LE isometric             5.56       4.31    -1.6
                  strengthening Control    4.98       5.33
Topp et al (42)   LE isometric             8.26       6.37    -2.01
                  strengthening Control    8.04       8.16
Topp et al (42)   LE isometric            17.99      15.15    -1.52
                  strengthening Control   18.85      17.53
Topp et al (42)   LE isometric            16.86      13.95    -1.21
                  strengthening Control   18.04      16.34
Topp et al (42)   LE isometric             5.13       5.03    -0.37
                  strengthening Control    5.23       5.50
Topp et al (42)   LE isometric            38.13      35.97    -2.99
                  strengthening Control   38.87      39.70

                                          Relative Difference
                                          in Change From
Study             Treatment Group         Baseline

Topp et al (42)   LE isometric            -61%
                  strengthening Control
Topp et al (42)   LE isometric            -70%
                  strengthening Control
Topp et al (42)   LE isometric            -47%
                  strengthening Control
Topp et al (42)   LE isometric            -40%
                  strengthening Control
Topp et al (42)   LE isometric            -12%
                  strengthening Control
Topp et al (42)   LE isometric            -30%
                  strengthening Control
Topp et al (42)   LE isometric            -25%
                  strengthening Control
Topp et al (42)   LE isometric             -8%
                  strengthening Control
Topp et al (42)   LE isometric             -7%
                  strengthening Control
Topp et al (42)   LE isometric             -7%
                  strengthening Control
Topp et al (42)   LE isometric             -8%
                  strengthening Control

Study             Treatment Group         WMD (95% CI)

Topp et al (42)   LE isometric            -2.14 (-4.01, -0.27)
                  strengthening Control
Topp et al (42)   LE isometric            -2.05 (-3.62, -0.48)
                  strengthening Control
Topp et al (42)   LE isometric            -1 .68 (-3.41, 0.05)
                  strengthening Control
Topp et al (42)   LE isometric            -1.62 (-3.24, 0)
                  strengthening Control
Topp et al (42)   LE isometric            -0.39 (-1.91, 1.13)
                  strengthening Control
Topp et al (42)   LE isometric            -1 .02 (-2.75, 0.71)
                  strengthening Control
Topp et al (42)   LE isometric            -1 .79 (-4.67, 1.09)
                  strengthening Control
Topp et al (42)   LE isometric            -2.38 (-5.75, 0.99)
                  strengthening Control
Topp et al (42)   LE isometric            -2.39 (-5.66, 0.88)
                  strengthening Control
Topp et al (42)   LE isometric            -0.47 (-1 .22, 0.28)
                  strengthening Control
Topp et al (42)   LE isometric            -3.73 (-8.91, 1.45)
                  strengthening Control

(a) LE=lower extremity, WOMAC=Western Ontario and McMaster Universities
Osteoarthritis Index, WMD=weighted mean difference, CI=confidence
interval.

Table 5.

Clinical Relevance: Concentric Versus Control (a)

            Treatment                              No. of     Baseline
Study       Group        Outcome                   Patients     Mean

Gur et      Concentric   Pain at night (0-10),
  al (67)                  end Tx: 8 wk            9            4.4
            Control      Pain at night (0-10),
                           end Tx: 8 wk            6            3.2
Gur et      Concentric   Pain after inactivity
  al (67)                  (0-10), end Tx: 8 wk    9            4.1
            Control      Pain after inactivity
                           (0-10), end Tx: 8 wk    6            3.8
Gur et      Concentric   Pain sitting (0-10),
  al (67)                  end Tx: 8 wk            9            3.4
            Control      Pain sitting (0-10),
                           end Tx: 8 wk            6            2.5
Gur et      Concentric   Pain rising from chair,
  al (67)                  end Tx: 8 wk            9            5.2
            Control      Pain rising from chair,
                           end Tx: 8 wk            6            4.7
Gur et      Concentric   Pain standing, end
  al (67)                  Tx: 8 wk                9            4.0
            Control      Pain standing, end
                           Tx: 8 wk                6            4.0
Gur et      Concentric   Pain climbing stairs,
  al (67)                  end Tx: 8 wk            9            5.8
            Control      Pain climbing stairs,
                           end Tx: 8 wk            6            4.8
Gur et      Concentric   Pain descending stairs,
  al (67)                  end Tx: 8 wk            9            5.8
            Control      Pain descending stairs,
                           end Tx: 8 wk            6            5.0
Gur et      Concentric   Pain, total score, end
  al (67)                  Tx: 8 wk                9           33.9
            Control      Pain. total score, end
                           Tx: 8 wk                6           27.3

                                                   End-of-
            Treatment                              Study      Absolute
Study       Group        Outcome                   Mean       Benefit

Gur et      Concentric   Pain at night (0-10),
  al (67)                  end Tx: 8 wk             1.4         -3.3
            Control      Pain at night (0-10),
                           end Tx: 8 wk             3.5
Gur et      Concentric   Pain after inactivity
  al (67)                  (0-10), end Tx: 8 wk     1.2         -2.9
            Control      Pain after inactivity
                           (0-10), end Tx: 8 wk     3.8
Gur et      Concentric   Pain sitting (0-10),
  al (67)                  end Tx: 8 wk             0.9         -2.7
            Control      Pain sitting (0-10),
                           end Tx: 8 wk             2.7
Gur et      Concentric   Pain rising from chair,
  al (67)                  end Tx: 8 wk             2.0         -3.2
            Control      Pain rising from chair,
                           end Tx: 8 wk             4.7
Gur et      Concentric   Pain standing, end
  al (67)                  Tx: 8 wk                 1.4         -2.6
            Control      Pain standing, end
                           Tx: 8 wk                 4.0
Gur et      Concentric   Pain climbing stairs,
  al (67)                  end Tx: 8 wk             1.7         -4.3
            Control      Pain climbing stairs,
                           end Tx: 8 wk             5.0
Gur et      Concentric   Pain descending stairs,
  al (67)                  end Tx: 8 wk             1.7         -4.1
            Control      Pain descending stairs,
                           end Tx: 8 wk             5.0
Gur et      Concentric   Pain, total score, end
  al (67)                  Tx: 8 wk                10.3        -24.3
            Control      Pain. total score, end
                           Tx: 8 wk                28.0

                                                   Relative Difference
            Treatment                              in Change From
Study       Group        Outcome                   Baseline

Gur et      Concentric   Pain at night (0-10),
  al (67)                  end Tx: 8 wk            -84%
            Control      Pain at night (0-10),
                           end Tx: 8 wk
Gur et      Concentric   Pain after inactivity
  al (67)                  (0-10), end Tx: 8 wk    -73%
            Control      Pain after inactivity
                           (0-10), end Tx: 8 wk
Gur et      Concentric   Pain sitting (0-10),
  al (67)                  end Tx: 8 wk            -89%
            Control      Pain sitting (0-10),
                           end Tx: 8 wk
Gur et      Concentric   Pain rising from chair,
  al (67)                  end Tx: 8 wk            -64%
            Control      Pain rising from chair,
                           end Tx: 8 wk
Gur et      Concentric   Pain standing, end
  al (67)                  Tx: 8 wk                -65%
            Control      Pain standing, end
                           Tx: 8 wk
Gur et      Concentric   Pain climbing stairs,
  al (67)                  end Tx: 8 wk            -80%
            Control      Pain climbing stairs,
                           end Tx: 8 wk
Gur et      Concentric   Pain descending stairs,
  al (67)                  end Tx: 8 wk            -75%
            Control      Pain descending stairs,
                           end Tx: 8 wk
Gur et      Concentric   Pain, total score, end
  al (67)                  Tx: 8 wk                -78%
            Control      Pain. total score, end
                           Tx: 8 wk

            Treatment
Study       Group        Outcome                   WMD (95% CI)

Gur et      Concentric   Pain at night (0-10),
  al (67)                  end Tx: 8 wk            -2.1 (-3.64, -0.56)
            Control      Pain at night (0-10),
                           end Tx: 8 wk
Gur et      Concentric   Pain after inactivity
  a (67)                  (0-10), end Tx: 8 wk     -2.6 (-3.56, -1.64)
            Control      Pain after inactivity
                           (0-10), end Tx: 8 wk
Gur et      Concentric   Pain sitting (0-10),
  a1 (67)                  end Tx: 8 wk            -1.8 (-2.51, -1.09)
            Control      Pain sitting (0-10),
                           end Tx: 8 wk
Gur et      Concentric   Pain rising from chair,
  al (67)                  end Tx: 8 wk            -2.7 (-3.83, -1.57)
            Control      Pain rising from chair,
                           end Tx: 8 wk
Gur et      Concentric   Pain standing, end
  al (67)                  Tx: 8 wk                -2.6 (-3.46, -1.74)
            Control      Pain standing, end
                           Tx: 8 wk
Gur et      Concentric   Pain climbing stairs,
  al (67)                  end Tx: 8 wk            -3.3 (-4.17, -2.43)
            Control      Pain climbing stairs,
                           end Tx: 8 wk
Gur et      Concentric   Pain descending stairs,
  al (67)                  end Tx: 8 wk            -3.3 (-4.27, -2.33)
            Control      Pain descending stairs,
                           end Tx: 8 wk
Gur et      Concentric   Pain, total score, end
  al (67)                  Tx: 8 wk                -17.7 (-22.79,
                                                     -12.61)
            Control      Pain. total score, end
                           Tx: 8 wk

(a) Tx=treatment, WMD=weighted mean difference, CI=confidence interval.

Table 6.

Clinical Relevance: Concentric-Eccentric Versus Control (a)

            Treatment                               No. of     Baseline
Study       Group         Outcome                   Patients   Mean

Gur et      Concentric-   Pain at night (0-10),     8           3.8
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain at night (0-10),     6           3.2
                            end Tx: 8 wk
Gur et      Concentric-   Pain sitting (0-10),      8           3.5
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain sitting (0-10),      6           2.5
                            end Tx: 8 wk
Gur et      Concentric-   Pain rising from chair,   8           5.4
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain rising from chair,   6           4.7
                            end Tx: 8 wk
Gur et      Concentric-   Pain standing, end
  al (67)     eccentric     Tx: 8 wk                8           3.5
            Control       Pain standing, end
                            Tx: 8 wk                6           4.0
Gur et      Concentric-   Pain climbing stairs,
  al (67)     eccentric     end Tx: 8 wk            8           6.5
            Control       Pain climbing stairs,
                            end Tx: 8 wk            6           4.8
Gur et      Concentric-   Pain, total score,
  al (67)     eccentric     end Tx: 8 wk            8          35.8
            Control       Pain, total score,
                            end Tx: 8 wk            6          27.3
Gur et      Concentric-   15-m walk (s), end        8           3.9
  al (67)     eccentric     Tx: 8 wk
            Control       15-m walk (s), end
                            Tx: 8 wk                6           3.5
Gur et      Concentric-   Stair-climbing time
  al (67)     eccentric     (s), end Tx: 8 wk       8           4.6
            Control       Stair-climbing time
                            (s), end Tx: 8 wk       6           3.7
Gur et      Concentric-   Stair-descending time
  al (67)     eccentric     (s), end Tx: 8 wk       8           5.3
            Control       Stair-descending time
                            (s), end Tx: 8 wk       6           4.7
Gur et      Concentric-   Function, total score,
  al (67)     eccentric     end Tx: 8 wk            8          19.1
            Control       Function, total score,
                            end Tx: 8 wk            6          15.5

                                                    End-of-
            Treatment                               Study      Absolute
Study       Group         Outcome                   Mean       Benefit

Gur et      Concentric-   Pain at night (0-10),      1.5        -2.6
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain at night (0-10),      3.5
                            end Tx: 8 wk
Gur et      Concentric-   Pain sitting (0-10),       1.1        -2.6
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain sitting (0-10),       2.7
                            end Tx: 8 wk
Gur et      Concentric-   Pain rising from chair,    2.5        -2.9
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain rising from chair,    4.7
                            end Tx: 8 wk
Gur et      Concentric-   Pain standing, end
  al (67)     eccentric     Tx: 8 wk                 1.6        -1.9
            Control       Pain standing, end
                            Tx: 8 wk                 4.0
Gur et      Concentric-   Pain climbing stairs,
  al (67)     eccentric     end Tx: 8 wk             3.5        -3.2
            Control       Pain climbing stairs,
                            end Tx: 8 wk             5.0
Gur et      Concentric-   Pain, total score,
  al (67)     eccentric     end Tx: 8 wk            16.6       -19.9
            Control       Pain, total score,
                            end Tx: 8 wk            28.0
Gur et      Concentric-   15-m walk (s), end         1.0        -2.9
  al (67)     eccentric     Tx: 8 wk
            Control       15-m walk (s), end
                            Tx: 8 wk                 3.5
Gur et      Concentric-   Stair-climbing time
  al (67)     eccentric     (s), end Tx: 8 wk        1.5        -4.1
            Control       Stair-climbing time
                            (s), end Tx: 8 wk        4.7
Gur et      Concentric-   Stair-descending time
  al (67)     eccentric     (s), end Tx: 8 wk        1.8        -4.1
            Control       Stair-descending time
                            (s), end Tx: 8 wk        5.3
Gur et      Concentric-   Function, total score,
  al (67)     eccentric     end Tx: 8 wk             6.0       -15.6
            Control       Function, total score,
                            end Tx: 8 wk            18.0

                                                    Relative Difference
            Treatment                               in Change From

Study       Group         Outcome                   Baseline

Gur et      Concentric-   Pain at night (0-10),     -73%
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain at night (0-10),
                            end Tx: 8 wk
Gur et      Concentric-   Pain sitting (0-10),      -85%
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain sitting (0-10),
                            end Tx: 8 wk
Gur et      Concentric-   Pain rising from chair,   -57%
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain rising from chair,
                            end Tx: 8 wk
Gur et      Concentric-   Pain standing, end
  al (67)     eccentric     Tx: 8 wk                -51%
            Control       Pain standing, end
                            Tx: 8 wk
Gur et      Concentric-   Pain climbing stairs,
  al (67)     eccentric     end Tx: 8 wk            -55%
            Control       Pain climbing stairs,
                            end Tx: 8 wk
Gur et      Concentric-   Pain, total score,
  al (67)     eccentric     end Tx: 8 wk            -62%
            Control       Pain, total score,
                            end Tx: 8 wk
Gur et      Concentric-   15-m walk (s), end        -78%
  al (67)     eccentric     Tx: 8 wk
            Control       15-m walk (s), end
                            Tx: 8 wk
Gur et      Concentric-   Stair-climbing time
  al (67)     eccentric     (s), end Tx: 8 wk       -97%
            Control       Stair-climbing time
                            (s), end Tx: 8 wk
Gur et      Concentric-   Stair-descending time
  al (67)     eccentric     (s), end Tx: 8 wk       -81%
            Control       Stair-descending time
                            (s), end Tx: 8 wk
Gur et      Concentric-   Function, total score,
  al (67)     eccentric     end Tx: 8 wk            -89%
            Control       Function, total score,
                            end Tx: 8 wk

            Treatment
Study       Group         Outcome                   WMD (95% CI)

Gur et      Concentric-   Pain at night (0-10),     -2 (-3.4, -0.6)
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain at night (0-10),
                            end Tx: 8 wk
Gur et      Concentric-   Pain sitting (0-10),      -1.6 (-2.4, -0.8)
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain sitting (0-10),
                            end Tx: 8 wk
Gur et      Concentric-   Pain rising from chair,   -2.2 (-3.43, -0.97)
  al (67)     eccentric     end Tx: 8 wk
            Control       Pain rising from chair,
                            end Tx: 8 wk
Gur et      Concentric-   Pain standing, end
  al (67)     eccentric     Tx: 8 wk                -2.4 (-3.54, -1.26)
            Control       Pain standing, end
                            Tx: 8 wk
Gur et      Concentric-   Pain climbing stairs,
  al (67)     eccentric     end Tx: 8 wk            -1.5 (-2.78, -0.22)
            Control       Pain climbing stairs,
                            end Tx: 8 wk
Gur et      Concentric-   Pain, total score,
  al (67)     eccentric     end Tx: 8 wk            -11.4 (-17.95,
                                                      -4.85)
            Control       Pain, total score,
                            end Tx: 8 wk
Gur et      Concentric-   15-m walk (s), end        -0.23 (-1 .45,
  al (67)     eccentric     Tx: 8 wk                  0.99)
            Control       15-m walk (s), end
                            Tx: 8 wk
Gur et      Concentric-   Stair-climbing time
  al (67)     eccentric     (s), end Tx: 8 wk       -0.07 (-0.92, 0.78)
            Control       Stair-climbing time
                            (s), end Tx: 8 wk
Gur et      Concentric-   Stair-descending time
  al (67)     eccentric     (s), end Tx: 8 wk       -0.86 (-2.52, 0.8)
            Control       Stair-descending time
                            (s), end Tx: 8 wk
Gur et      Concentric-   Function, total score,
  al (67)     eccentric     end Tx: 8 wk            -0.86 (-2.52, 0.8)
            Control       Function, total score,
                            end Tx: 8 wk

(a) Tx=treatment, WMD=weighted mean difference, CI=confidence interval.

Table 7.

Clinical Relevance: Home Strengthening Program Versus Control (a)

               Treatment                       No. of         Baseline
Study          Group          Outcome          Patients       Mean

Evcik and      Home streng-   WOMAC-physical
  Sonel (47)     thening        function       27             25.4
               Control        WOMAC-physical
                                function       26             25.2
Evcik and      Home streng-   NHP-physical
  Sonel (47)     thening        mobility       27             40.2
               Control        NHP-physical
                                mobility       26             44.1
Evcik and      Home streng-   WOMAC-pain       27              6.6
  Sonel (47)     thening
               Control        WOMAC-pain       26              6.6
Evcik and      Home streng-   Pain-VAS         27              7.2
  Sonel (47)     thening
               Control        Pain-VAS         26              7.0
Evcik and      Home streng-   NHP-pain         27             40.0
  Sonel (47)     thening
               Control        NHP-pain         26             40.9
Evcik and      Home streng-   NHP-energy       27             53.2
  Sonel (47)     thening
               Control        NHP-energy       26             52.9

               Treatment                       End-of-Study   Absolute
Study          Group          Outcome          Mean           Benefit

Evcik and      Home streng-   WOMAC-physical
  Sonel (47)     thening        function       10.8           -10.1
               Control        WOMAC-physical
                                function       20.7
Evcik and      Home streng-   NHP-physical
  Sonel (47)     thening        mobility       29.5            -3.2
               Control        NHP-physical
                                mobility       36.6
Evcik and      Home streng-   WOMAC-pain        3.0            -3
  Sonel (47)     thening
               Control        WOMAC-pain        6.0
Evcik and      Home streng-   Pain-VAS          3.5            -3.5
  Sonel (47)     thening
               Control        Pain-VAS          6.8
Evcik and      Home streng-   NHP-pain          9.8            -9.7
  Sonel (47)     thening
               Control        NHP-pain         20.4
Evcik and      Home streng-   NHP-energy       33.4           -16.2
  Sonel (47)     thening
               Control        NHP-energy       49.3

                                               Relative Difference
               Treatment                       in Change From
Study          Group          Outcome          Baseline

Evcik and      Home streng-   WOMAC-physical
  Sonel (47)     thening        function       -40%
               Control        WOMAC-physical
                                function
Evcik and      Home streng-   NHP-physical
  Sonel (47)     thening        mobility        -8%
               Control        NHP-physical
                                mobility
Evcik and      Home streng-   WOMAC-pain       -45%
  Sonel (47)     thening
               Control        WOMAC-pain
Evcik and      Home streng-   Pain-VAS         -49%
  Sonel (47)     thening
               Control        Pain-VAS
Evcik and      Home streng-   NHP-pain         -24%
  Sonel (47)     thening
               Control        NHP-pain
Evcik and      Home streng-   NHP-energy       -31%
  Sonel (47)     thening
               Control        NHP-energy

               Treatment
Study          Group          Outcome          WMD (95% CI)

Evcik and      Home streng-   WOMAC-physical
  Sonel (47)     thening        function       -9.9 (-11.72, -8.08)
               Control        WOMAC-physical
                                function
Evcik and      Home streng-   NHP-physical
  Sonel (47)     thening        mobility       -7.1 (-10.06, -4.14)
               Control        NHP-physical
                                mobility
Evcik and      Home streng-   WOMAC-pain       -3 (-4.42, -1.58)
  Sonel (47)     thening
               Control        WOMAC-pain
Evcik and      Home streng-   Pain-VAS         -3.3 (-3.98, -2.62)
  Sonel (47)     thening
               Control        Pain-VAS
Evcik and      Home streng-   NHP-pain         -10.6 (-12.3, -8.9)
  Sonel (47)     thening
               Control        NHP-pain
Evcik and      Home streng-   NHP-energy       -15.9 (-16.93, -14.87)
  Sonel (47)     thening
               Control        NHP-energy

(a) WOMAC=Western Ontario and McMaster Universities Osteoarthritis
Index, NHP=Nottingham Profile VAS=visual analog scale WMD=weighted
mean difference, CI=confidence interval.

Table 8.

Clinical Relevance: General Lower-Extremity Exercise Program Versus
Control (a)

                      Treatment                                No. of
Study                 Group       Outcome                      Patients

Rogind et al (78)     Exercises   ROM in knee flexion           12
                                    ([degrees]) (most
                                    affected knee), end Tx:
                                    10-12 wk
                      Control     ROM in knee flexion           13
                                    ([degrees]) (most
                                    affected knee), end Tx:
                                    10-12 wk
Rogind et al (78)     Exercises   ROM in knee flexion           12
                                    ([degrees]) (least
                                    affected knee), end Tx:
                                    10-12 wk
                      Control     ROM in knee flexion           13
                                    ([degrees]) (least
                                    affected knee), end Tx:
                                    10-12 wk
Rogind et al (78)     Exercises   ROM in knee flexion           12
                                    ([degrees]) (least
                                    affected knee), FU: 12
                                    mo
                      Control     ROM in knee flexion           13
                                    ([degrees]) (least
                                    affected knee), FU: 12
                                    mo
van Baar et al (82)   Exercises   Hip muscle force, change,     93
                                    FU: 24 wk
                      Control     Hip muscle force, change,     98
                                    FU: 24 wk
Fransen et al (65)    Exercises   WOMAC-pain, change            83
                                    (0-100), end Tx: 8 wk
                      Control     WOMAC-pain, change            43
                                    (0-100), end Tx: 8 wk
van Baar et al (82)   Exercises   Pain past week-VAS            93
                                    (0-100), change, end Tx:
                                    10-12 wk
                      Control     Pain past week-VAS            98
                                    (0-100), change, end Tx:
                                    10-12 wk
van Baar et al (82)   Exercises   Pain-VAS (0-100), FU: 24      93
                                    wk
                      Control     Pain-VAS (0-100), FU: 24      98
                                    wk
van Baar et al (83)   Exercises   Pain-VAS (0-100), FU: 24      98
                                    wk
                      Control     Pain-VAS (0-100), FU: 24     102
                                    wk
Rogind et al (78)     Exercises   Pain at night (0-10), FU:     12
                                    12 mo
                      Control     Pain at night (0-10), FU:     13
                                    12 mo
Fransen et al (65)    Exercises   Muscle force, isometric       83
                                    knee extensors (N), end
                                    Tx: 8 wk
                      Control     Muscle force, isometric       43
                                    knee extensors (N), end
                                    Tx: 8 wk
Fransen et al (65)    Exercises   Muscle force, isometric       83
                                    knee flexors (N), end
                                    Tx: 8 wk
                      Control     Muscle force, isometric       43
                                    knee flexors (N), end
                                    Tx: 8 wk
Fransen et al (65)    Exercises   Fast speed (cm/s), end Tx:    83
                                    8 wk
                      Control     Fast speed (cm/s), end Tx:    43
                                    8 wk
Fransen et al (65)    Exercises   Fast cadence (steps/min),     83
                                    end Tx: 8 wk
                      Control     Fast cadence (steps/min),     43
                                    end Tx: 8 wk
Fransen et al (65)    Exercises   Fast stride length (cm),      83
                                    end Tx: 8 wk
                      Control     Fast stride length (cm),      43
                                    end Tx: 8 wk
Fransen et al (65)    Exercises   WOMAC--function, change       83
                                    (0-100), end Tx: 8 wk
                      Control     WOMAC--function, change       43
                                    (0-100), end Tx: 8 wk
van Barr et al (82)   Exercises   Self-reported disability      93
                                    (IRLG), FU: 24 wk
                      Control     Self-reported disability      98
                                    (IRLG), FU: 24 wk
Fransen et al (65)    Exercises   SF-36 quality of life,        83
                                    change, end Tx: 8 wk
                      Control     SF-36 quality of life,        43
                                    change, end Tx: 8 wk

                      Treatment   Baseline   End-of-Study   Absolute
Study                 Group       Mean       Mean           Benefit

Rogind et al (78)     Exercises   130        135              3
                      Control     123        125
Rogind et al (78)     Exercises   132        135              4
                      Control     126        125
Rogind et al (78)     Exercises   132        133              6
                      Control     126        121
van Baar et al (82)   Exercises   N/A          0.22         WMD=0.18
                      Control     N/A          0.04
Fransen et al (65)    Exercises   N/A         10.6          WMD=12.1
                      Control     N/A         -1.5
van Baar et al (82)   Exercises    46.9       24.1          -17.1
                      Control      43.1       37.4
van Baar et al (82)   Exercises    34.0       23.5          -11.2
                      Control      28.7       29.4
van Baar et al (83)   Exercises    34.0       16.8          -11.6
                      Control      28.7       23.1
Rogind et al (78)     Exercises     4.0        2.0           -3
                      Control       5.0        6.0
Fransen et al (65)    Exercises   169.45     180.25          13.2
                      Control     173.3      170.9
Fransen et al (65)    Exercises    94.65     103.05           9
                      Control     100.1       99.5
Fransen et al (65)    Exercises   128.6      135.7            6.7
                      Control     127.8      128.2
Fransen et al (65)    Exercises   119.9      121.8            1.6
                      Control     117.6      117.9
Fransen et al (65)    Exercises   128.2      132.9            4.3
                      Control     130.3      130.7
Fransen et al (65)    Exercises    60.8       68.5            7.8
                      Control      60.0       59.9
van Barr et al (82)   Exercises   -20.0       -1.1           -1.7
                      Control     -20.6        0.0
Fransen et al (65)    Exercises    32.85      36.45           3.1
                      Control      34.8       35.3

                                  Relative
                                  Difference
                                  in Change
                      Treatment   From
Study                 Group       Baseline     WMD (95% CI)

Rogind et al (78)     Exercises     2%         -10 (-14.09, -5.91)
                      Control
Rogind et al (78)     Exercises     3%         -10 (-12.25, -7.75)
                      Control
Rogind et al (78)     Exercises     5%         -12 (-16.94, -7.06)
                      Control
van Baar et al (82)   Exercises                 -0.18 (-0.34, -0.02)
                      Control
Fransen et al (65)    Exercises                -12.1 (-14.24, -9.96)
                      Control
van Baar et al (82)   Exercises   -38%         -17.1 (-29.99, -4.21)
                      Control
van Baar et al (82)   Exercises   -36%         -11.2 (-18.59, -3.81)
                      Control
van Baar et al (83)   Exercises   -37%         -11.6 (-19.5, -3.7)
                      Control
Rogind et al (78)     Exercises   -66%          -4 (-5.94, -2.06)
                      Control
Fransen et al (65)    Exercises     8%         -13.2 (-14.44, -11.96)
                      Control
Fransen et al (65)    Exercises     9%          -9 (-9.96, -8.04)
                      Control
Fransen et al (65)    Exercises     5%          -6.7 (-7.06, -6.34)
                      Control
Fransen et al (65)    Exercises     1%          -1.6 (-1.8, -1.4)
                      Control
Fransen et al (65)    Exercises     3%          -4.3 (-4.61, -3.99)
                      Control
Fransen et al (65)    Exercises    13%          -7.8 (-8.48, -7.12)
                      Control
van Barr et al (82)   Exercises     8%          -1.1 (-1.91, -0.29)
                      Control
Fransen et al (65)    Exercises     9%          -3.1 (-3.44, -2.76)
                      Control

(a) FU=follow-up, N/A=not available, ROM=range of motion, Tx=Treatment,
WMD=weighted mean difference, CI=confidence interval, VAS=visual analog
scale, WOMAC=Western Ontario and McMaster Universities Osteoarthritis
Index, IRLG=Influence of Rheumatic Disease on General Health and
Lifestyle, SF-36=Medical Outcomes Study 36-Item Short-Form Health
Survey questionnaire.

Table 9.

Clinical Relevance: Lower-Extremity Strengthening Exercise Program
Versus Control (a)

                                                          No.
Study                  Group       Outcome                Improved   N

Borjesson et al (64)   Exercises   Ability to step down   13         34
                       Control     Ability to step down    4         34

                                   Risk     Risk
                                   Occur-   Differ-
Study                  Group       rence    ence      WMD (95% GI)

Borjesson et al (64)   Exercises   38%      26%       3.25 (1.18, 8.97)
                       Control     12%

(a) WMD=weighted mean difference, CI-confidence interval.

Table 10.

Clinical Relevance: Progression Versus No Progression (a)

           Treatment                             No. of      Baseline
Study      Group       Outcome                   Patients    Mean

Petrella   Exercises   Pain at rest--VAS         91          N/A
  (75)                   (0-100), change,
                         end Tx: 8 wk
           Control     Pain at rest--VAS         88          N/A
                         (0-100), change,
                         end Tx: 8 wk
Petrella   Exercises   Pain after step           91          N/A
  (75)                   test--VAS (0-100),
                         change, end Tx: 8 wk
           Control     Pain after step           88          N/A
                         test--VAS (0-100),
                         change, end Tx: 8 wk
Petrella   Exercises   ROM in knee flexion       91          N/A
  (75)                   ([degrees]), end
                         Tx: 8 wk
           Control     ROM in knee flexion       88          N/A
                         ([degrees]), end
                         Tx: 8 wk
                                                 End-of-
           Treatment                             Study      Absolute
Study      Group       Outcome                   Mean       Benefit

Petrella   Exercises   Pain at rest-VAS          N/A        -23 WMD
  (75)                   (0-100), change,
                         end Tx: 8 wk
           Control     Pain at rest--VAS         N/A
                         (0-100), change,
                         end Tx: 8 wk
Petrella   Exercises   Pain after step           N/A        11.3 WMD
  (75)                   test--VAS (0-100),
                         change, end Tx: 8 wk
           Control     Pain after step           N/A
                         test--VAS (0-100),
                         change, end Tx: 8 wk
Petrella   Exercises   ROM in knee flexion       N/A        13[degrees]
  (75)                   ([degrees]), end
                         Tx: 8 wk
           Control     ROM in knee flexion       N/A
                         ([degrees]), end
                         Tx: 8 wk

           Treatment
Study      Group       Outcome                   WMD (95% CI)

Petrella   Exercises   Pain at rest--VAS         -23 (-24.03, -21.97)
  (75)                   (0-100), change, end
                         Tx: 8 wk
           Control     Pain at rest--VAS
                         (0-100), change,
                         end Tx: 8 wk
Petrella   Exercises   Pain after step           -11.3 (-12.04, -10.56)
  (75)                   test--VAS (0-100),
                         change, end Tx: 8 wk
           Control     Pain after step
                         test--VAS (0-100),
                         change, end Tx: 8 wk
Petrella   Exercises   ROM in knee flexion       -13 (-14.45, -11.55)
  (75)                   ([degrees]), end
                         Tx: 8 wk
           Control     ROM in knee flexion
                         ([degrees]), end
                         Tx: 8 wk

(a) N/A=not available, ROM=range of motion, Tx=treatment, VAS=visual
analog scale. WMD=weighted mean difference. CI=confidence interval.

Table 11.

Clinical Relevance: Hand Strengthening Versus Control (a)

            Treatment                            No. of      Baseline
Study       Group             Outcome            Patients    Mean

Stamm et    Hand functional   Grip force,        20          0.43
  al (80)     strengthening     change, right,
                                end Tx: 3 mo
            Control           Grip force,        20          0.54
                                change, right,
                                end Tx: 3 mo
Stamm et    Hand functional   Grip force,        20          0.44
  al (80)     strengthening     change, left,
                                end Tx: 3 mo
            Control           Grip force,        20          0.53
                                change, left,
                                end Tx: 3 mo

                                                 End-of-
            Treatment                            Study      Absolute
Study       Group             Outcome            Mean       Benefit

Stamm et    Hand functional   Grip force,        0.55       0.09
  al (80)     strengthening     change, right,
                                end Tx: 3 mo
            Control           Grip force,        0.57
                                change, right,
                                end Tx: 3 mo
Stamm et    Hand functional   Grip force,        0.55       0.08
  al (80)     strengthening     change, left,
                                end Tx: 3 mo
            Control           Grip force,        0.56
                                change, left,
                                end Tx: 3 mo

                                                 Relative
                                                 Difference
                                                 in Change
            Treatment                            From
Study       Group             Outcome            Baseline

Stamm et    Hand functional   Grip force,        19%
  al (80)     strengthening     change, right,
                                end Tx: 3 mo
            Control           Grip force,
                                change, right,
                                end Tx: 3 mo
Stamm et    Hand functional   Grip force,        16%
  al (80)     strengthening     change, left,
                                end Tx: 3 mo
            Control           Grip force,
                                change, left,
                                end Tx: 3 mo

            Treatment
Study       Group             Outcome            WMD (95% CI)

Stamm et    Hand functional   Grip force,        -0.11 (-0.13, -0.09)
  al (80)     strengthening     change, right,
                                end Tx: 3 mo
            Control           Grip force,
                                change, right,
                                end Tx: 3 mo
Stamm et    Hand functional   Grip force,        -0.1 (-0.11, -0.09)
  al (80)     strengthening     change, left,
                                end Tx: 3 mo
            Control           Grip force,
                                change, left,
                                end Tx: 3 mo

(a) Tx=treatment, WMD=weighted mean difference,
CI=confidence interval.

Table 12.

Clinical Relevance: Hand Strengthening Versus Control (a)

                                                  No.
Study      Group             Outcome              Observed    N

Stamm et   Hand functional   Pain-VAS, no.        13          20
  al (80)    strengthening     improved, end
                               Tx: 3 mo
           Control           Pain-VAS, no.         4          20
                               improved, end
                               Tx: 3 mo

                                                Risk         Risk
Study      Group             Outcome            Occurrence   Difference

Stamm et   Hand functional   Pain-VAS, no.      65%          45%
  al (80)    strengthening     improved, end
                               Tx: 3 mo
           Control           Pain-VAS, no.      20%
                               improved, end
                               Tx: 3 mo

Study      Group             Outcome            WMD (95% CI)

Stamm et   Hand functional   Pain-VAS, no.      3.25 (1.28, 8.27)
  al (80)    strengthening     improved, end
                               Tx: 3 mo
           Control           Pain-VAS, no.
                               improved, end
                               Tx: 3 mo

(a) VAS=visual analog scale, Tx=treatment, WMD=weighted mean
difference, CI=confidence interval.

Table 13.

Clinical Relevance: Walking Program Versus Control (a)

               Treatment                            No. of     Baseline
Study          Group       Outcome                  Patients   Mean

Rejeski et     Walking     Stair-climbing time      357 (b)    N/A
  al (76)        program     (s), FU: 18 mo
               Control     Stair-climbing time      357 (b)    N/A
                             (s), FU: 18 mo
Rejeski et     Walking     Climbing self-efficacy   357 (b)    N/A
  al (76)        program     score (0-10),
                             FU: 18 mo
               Control     Climbing self-efficacy   357 (b)    N/A
                             score (0-10),
                             FU: 18 mo
Rejeski et     Walking     General health status    357 (b)    N/A
  al (76)        program     (0-100), FU: 18 mo
               Control     General health status    357 (b)    N/A
                             (0-100), FU: 18 mo
Kovar et       Walking     6-min walk test, end      47        381
  al (69)        program     Tx: 8 wk
               Control     6-min walk test, end      45        356
                             Tx: 8 wk
Peterson et    Walking     6-min walk test, end      47        390
  al (74)        program     Tx: 8 wk
               Control     6-min walk test, end      44        357
                             Tx: 8 wk
Peterson et    Walking     Free speed (m/min),       47         56
  al (74)        program     end Tx: 8 wk
               Control     Free speed (m/min),       44         51
                             end Tx: 8 wk
Peterson et    Walking     Free stride (m),          47          1.1
  al (74)        program     end Tx: 8 wk
               Control     Free stride (m),          44          1.1
                             end Tx: 8 wk
Peterson et    Walking     Fast speed (m/min),       47         76
  al (74)        program     end Tx: 8 wk
               Control     Fast speed (m/min),       44         70
                             end Tx: 8 wk
Peterson et    Walking     Fast stride (m),          47          1.2
  al (74)        program     end Tx: 8 wk
               Control     Fast stride (m),          44          1.2
                             end Tx: 8 wk
Peterson et    Walking     AIMS-physical activity    47         N/A
  al (74)        program     (0-10), end Tx: 8 wk
               Control     AIMS-physical activity    44         N/A
                             (0-10), end Tx: 8 wk
Minor et       Walking     AIMS-physical             36         N/A
  al (48)        program     activity (0-10),
                             end Tx: 12 wk
               Control     AIMS-physical             32         N/A
                             activity (0-10),
                             end Tx: 12 wk
Evcik and      Walking     NHP-physical mobility     28         41.3
  Sone1 (47)     program     (0-100), FU: 6 mo
               Control     NHP-physical mobility     26         44.1
                             (0-100), FU: 6 mo
Evcik and      Walking     WOMAC-physical            28         23.9
  Sonel (47)     program     function (0-68),
                             FU: 6 mo
               Control     WOMAC-physical            26         25.2
                             function (0-68),
                             FU: 6 mo
Peterson et    Walking     AIMS-pain (0-10), end     47         N/A
  al (74)        program     Tx: 8 wk
               Control     AIMS-pain (0-10), end     44         N/A
                             Tx: 8 wk
Evcik and      Walking     WOMAC-pain (0-10),        28          6.9
  Sonel (47)     program     FU: 6 mo
               Control     WOMAC-pain (0-10),        26          6.6
                             FU: 6 mo
Evcik and      Walking     Pain-VAS (0-10),          28          7.1
  Sonel (47)     program     FU: 6 mo
               Control     Pain-VAS (0-10),          26          7.0
                             FU: 6 mo
Evcik and      Walking     NHP-pain (0-100),         28         41.3
                 program     FU: 6 mo
  Sonel (47)   Control     NHP-pain (0-100),         26         40.9
                             FU: 6 mo
Evcik and      Walking     NHP-energy (0-100),       28         50.7
  Sonel (47)     program     FU: 6 mo
               Control     NHP-energy (0-100),       26         52.9
                             FU: 6 mo
Evcik and      Walking     NHP-sleep (0-100),        28         44.9
  Sonel (47)     program     FU: 6 mo
               Control     NHP-sleep (0-100),        26         45.0
                             FU: 6 mo
Peterson et    Walking     AIMS-medication use       47         N/A
  al (74)        program     (0-6), end Tx: 8 wk
               Control     AIMS-medication use       44         N/A
                             (0-6), end Tx: 8 wk
Minor et       Walking     15.2-m (50-ft) walking    36          9.9
  al (48)        program     time (s), end
                             Tx: 12 wk
               Control     15.2-m (50-ft) walking    32         11.0
                             time (s), end
                             Tx: 12 wk
Minor et       Walking     Aerobic capacity          36         18.9
  al (48)        program     (mL/kg
                             [min.sup.-1]),
                             end Tx: 12 wk
               Control     Aerobic capacity          32         17.4
                             (ml/kg min-1), end
                             Tx: 12 wk
Minor et       Walking     Exercise endurance        36         11.9
  al (48)        program     (min), end Tx: 12 wk
               Control     Exercise endurance        32         10.0
                             (min), end Tx: 12 wk
Minor et       Walking     Exercise heart rate       36        149
  al (48)        program     (bpm), end Tx: 12 wk
               Control     Exercise heart rate       32        138
                             (bpm), end Tx: 12 wk
Messier et     Exercises   Walking speed (cm/s),     34        109.42
  al (45)                    mid-Tx: 3 mo
               Control     Walking speed (cm/s),     36        109.42
                             mid-Tx: 3 mo
Messier et     Exercises   Walking speed (cm/s),     34        109.42
  al (45)                    mid-Tx: 9 mo
               Control     Walking speed (cm/s),     36        109.42
                             mid-Tx: 9 mo
Messier et     Exercises   Walking speed (cm/s),     34        109.42
  al (45)                    FU: 18 mo
               Control     Walking speed (cm/s),     36        109.42
                             FU: 18 mo
Messier et     Exercises   Cadence (steps/min),      34        106.45
  al (45)                    mid-Tx: 3 mo
               Control     Cadence (steps/min),      36        106.45
                             mid-Tx: 3 mo
Messier et     Exercises   Cadence (steps/min),      34        106.45
  al (45)                    mid-Tx: 9 mo
               Control     Cadence (steps/min),      36        106.45
                             mid-Tx: 3 mo
Messier et     Exercises   Cadence (steps/min),      34        106.45
  al (45)                    FU: 18 mo
               Control     Cadence (steps/min),      36        106.45
                             FU: 18 mo
Messier et     Exercises   Stride length (cm),       34        120.19
                             mid-Tx: 3 mo
  al (45)      Control     Stride length (cm),       36        120.19
                             mid-Tx: 3 mo
Messier et     Exercises   Stride length (cm),       34        120.19
  al (45)                    mid-Tx: 9 mo
               Control     Stride length (cm),       36        120.19
                             mid-Tx: 9 mo
Messier et     Exercises   Stride length (cm),       34        120.19
  al (45)                    FU: 18 mo
               Control     Stride length (cm),       36        120.19
                             FU: 18 mo
Messier et     Exercises   % swing, mid-Tx: 3 mo     34         34.36
  al (45)      Control     % swing, mid-Tx: 3 mo     36         34.36
Messier et     Exercises   % swing, mid-Tx: 9 mo     34         34.36
  al (45)      Control     % swing, mid-Tx: 9 mo     36         34.36
Messier et     Exercises   % swing, FU: 18 mo        34         34.36
  al (45)      Control     % swing, FU: 18 mo        36         34.36

                                                    End-of-
               Treatment                            Study     Absolute
Study          Group       Outcome                  Mean      Benefit

Rejeski et     Walking     Stair-climbing time        9.08      N/A
  al (76)        program     (s), FU: 18 mo
               Control     Stair-climbing time       10.49
                             (s), FU: 18 mo
Rejeski et     Walking     Climbing self-efficacy    66.06      N/A
  al (76)        program     score (0-10),
                             FU: 18 mo
               Control     Climbing self-efficacy    58.06
                             score (0-10),
                             FU: 18 mo
Rejeski et     Walking     General health status     75.56      N/A
  al (76)        program     (0-100), FU: 18 mo
               Control     General health status     71.97
                             (0-100), FU: 18 mo
Kovar et       Walking     6-min walk test, end     451         87
  al (69)        program     Tx: 8 wk
               Control     6-min walk test, end     339
                             Tx: 8 wk
Peterson et    Walking     6-min walk test, end     449         78
  al (74)        program     Tx: 8 wk
               Control     6-min walk test, end     338
                             Tx: 8 wk
Peterson et    Walking     Free speed (m/min),       61          2
  al (74)        program     end Tx: 8 wk
               Control     Free speed (m/min),       54
                             end Tx: 8 wk
Peterson et    Walking     Free stride (m),           1.2        0.2
  al (74)        program     end Tx: 8 wk
               Control     Free stride (m),           1.0
                             end Tx: 8 wk
Peterson et    Walking     Fast speed (m/min),       83          9
  al (74)        program     end Tx: 8 wk
               Control     Fast speed (m/min),       68
                             end Tx: 8 wk
Peterson et    Walking     Fast stride (m),           1.4        0.2
  al (74)        program     end Tx: 8 wk
               Control     Fast stride (m),           1.2
                             end Tx: 8 wk
Peterson et    Walking     AIMS-physical activity     3.74       2.22
  al (74)        program     (0-10), end Tx: 8 wk
               Control     AIMS-physical activity     5.96
                             (0-10), end Tx: 8 wk
Minor et       Walking     AIMS-physical activity     3.6        1.3
  al (48)        program     (0-10), end
                             Tx: 12 wk
               Control     AIMS-physical activity     4.9
                             (0-10), end
                             Tx: 12 wk
Evcik and      Walking     NHP-physical mobility      8.6      -25.2
  Sone1 (47)     program     (0-100), FU: 6 mo
               Control     NHP-physical mobility     36.6
                             (0-100), FU: 6 mo
Evcik and      Walking     WOMAC-physical            10.2       -9.2
  Sonel (47)     program     function (0-68),
                             FU: 6 mo
               Control     WOMAC-physical            20.7
                             function (0-68),
                             FU: 6 mo
Peterson et    Walking     AIMS-pain (0-10), end      3.77        N/A
  al (74)        program     Tx: 8 wk
               Control     AIMS-pain (0-10), end      4.77
                             Tx: 8 wk
Evcik and      Walking     WOMAC-pain (0-10),         3.4       -2.9
  Sonel (47)     program     FU: 6 mo
               Control     WOMAC-pain (0-10),         6.0
                             FU: 6 mo
Evcik and      Walking     Pain-VAS (0-10),           3.6       -3.3
  Sonel (47)     program     FU: 6 mo
               Control     Pain-VAS (0-10),           6.8
                             FU: 6 mo
Evcik and      Walking     NHP-pain (0-100),          9.0      -11.8
                 program     FU: 6 mo
  Sonel (47)   Control     NHP-pain (0-100),         20.4
                             FU: 6 mo
Evcik and      Walking     NHP-energy (0-100),       14.6      -32.5
  Sonel (47)     program     FU: 6 mo
               Control     NHP-energy (0-100),       49.3
                             FU: 6 mo
Evcik and      Walking     NHP-sleep (0-100),        19.6      -15.6
  Sonel (47)     program     FU: 6 mo
               Control     NHP-sleep (0-100),        35.3
                             FU: 6 mo
Peterson et    Walking     AIMS-medication use        3.64       N/A
  al (74)        program     (0-6), end Tx: 8 wk
               Control     AIMS-medication use        0.9
                             (0-6), end Tx: 8 wk
Minor et       Walking     15.2-m (50-ft) walking     8.7       -0.9
  al (48)        program     time (s), end
                             Tx: 12 wk
               Control     15.2-m (50-ft) walking    10.7
                             time (s), end
                             Tx: 12 wk
Minor et       Walking     Aerobic capacity          22.4        3.6
  al (48)        program     (mL/kg
                             [min.sup.-1]), end
                             Tx: 12 wk
               Control     Aerobic capacity          17.3
                             (ml/kg min-1), end
                             Tx: 12 wk
Minor et       Walking     Exercise endurance        14.8        1.4
  al (48)        program     (min), end Tx: 12 wk
               Control     Exercise endurance        11.5
                             (min), end Tx: 12 wk
Minor et       Walking     Exercise heart rate      158          5
  al (48)        program     (bpm), end Tx: 12 wk
               Control     Exercise heart rate      142
                             (bpm), end Tx: 12 wk
Messier et     Exercises   Walking speed (cm/s),    114.64       3.77
  al (45)                    mid-Tx: 3 mo
               Control     Walking speed (cm/s),    110.87
                             mid-Tx: 3 mo
Messier et     Exercises   Walking speed (cm/s),    116.30       4.37
  al (45)                    mid-Tx: 9 mo
               Control     Walking speed (cm/s),    111.93
                             mid-Tx: 9 mo
Messier et     Exercises   Walking speed (cm/s),    115.20       7.79
  al (45)                    FU: 18 mo
               Control     Walking speed (cm/s),    107.41
                             FU: 18 mo
Messier et     Exercises   Cadence (steps/min),     109.31       0.87
  al (45)                    mid-Tx: 3 mo
               Control     Cadence (steps/min),     108.44
                             mid-Tx: 3 mo
Messier et     Exercises   Cadence (steps/min),     109.42       0.94
  al (45)                    mid-Tx: 9 mo
               Control     Cadence (steps/min),     108.48
                             mid-Tx: 3 mo
Messier et     Exercises   Cadence (steps/min),     109.31       2.08
  al (45)                    FU: 18 mo
               Control     Cadence (steps/min),     107.23
                             FU: 18 mo
Messier et     Exercises   Stride length (cm),      122.76       2.17
                             mid-Tx: 3 mo
  al (45)      Control     Stride length (cm),      120.59
                             mid-Tx: 3 mo
Messier et     Exercises   Stride length (cm),      123.54       2.84
  al (45)                    mid-Tx: 9 mo
               Control     Stride length (cm),      120.70
                             mid-Tx: 9 mo
Messier et     Exercises   Stride length (cm),      123.92       6.49
  al (45)                    FU: 18 mo
               Control     Stride length (cm),      117.43
                             FU: 18 mo
Messier et     Exercises   % swing, mid-Tx: 3 mo    33.85        0.39
  al (45)      Control     % swing, mid-Tx: 3 mo    33.46
Messier et     Exercises   % swing, mid-Tx: 9 mo    33.67       -0.36
  al (45)      Control     % swing, mid-Tx: 9 mo    34.03
Messier et     Exercises   % swing, FU: 18 mo       35.63        0.54
  al (45)      Control     % swing, FU: 18 mo       35.09

                                                    Relative Difference
               Treatment                            in Change From
Study          Group       Outcome                  Baseline

Rejeski et     Walking     Stair-climbing time       14%
  al (76)        program     (s), FU: 18 mo
               Control     Stair-climbing time
                             (s), FU: 18 mo
Rejeski et     Walking     Climbing self-efficacy    13%
  al (76)        program     score (0-10),
                             FU: 18 mo
               Control     Climbing self-efficacy
                             score (0-10),
                             FU: 18 mo
Rejeski et     Walking     General health status      5%
  al (76)        program     (0-100), FU: 18 mo
               Control     General health status
                             (0-100), FU: 18 mo
Kovar et       Walking     6-min walk test, end      24%
  al (69)        program     Tx: 8 wk
               Control     6-min walk test, end
                             Tx: 8 wk
Peterson et    Walking     6-min walk test, end      21%
  al (74)        program     Tx: 8 wk
               Control     6-min walk test, end
                             Tx: 8 wk
Peterson et    Walking     Free speed (m/min),        4%
  al (74)        program     end Tx: 8 wk
               Control     Free speed (m/min),
                             end Tx: 8 wk
Peterson et    Walking     Free stride (m),          18%
  al (74)        program     end Tx: 8 wk
               Control     Free stride (m),
                             end Tx: 8 wk
Peterson et    Walking     Fast speed (m/min),       12%
  al (74)        program     end Tx: 8 wk
               Control     Fast speed (m/min),
                             end Tx: 8 wk
Peterson et    Walking     Fast stride (m),          17%
  al (74)        program     end Tx: 8 wk
               Control     Fast stride (m),
                             end Tx: 8 wk
Peterson et    Walking     AIMS-physical activity    46%
  al (74)        program     (0-10), end Tx: 8 wk
               Control     AIMS-physical activity
                             (0-10), end Tx: 8 wk
Minor et       Walking     AIMS-physical activity    31%
  al (48)        program     (0-10), end
                             Tx: 12 wk
               Control     AIMS-physical activity
                             (0-10), end
                             Tx: 12 wk
Evcik and      Walking     NHP-physical mobility    -59%
  Sone1 (47)     program     (0-100), FU: 6 mo
               Control     NHP-physical mobility
                             (0-100), FU: 6 mo
Evcik and      Walking     WOMAC-physical           -38%
  Sonel (47)     program     function (0-68),
                             FU: 6 mo
               Control     WOMAC-physical
                             function (0-68),
                             FU: 6 mo
Peterson et    Walking     AIMS-pain (0-10), end     23.5%
  al (74)        program     Tx: 8 wk
               Control     AIMS-pain (0-10), end
                             Tx: 8 wk
Evcik and      Walking     WOMAC-pain (0-10),       -43%
  Sonel (47)     program     FU: 6 mo
               Control     WOMAC-pain (0-10),
                             FU: 6 mo
Evcik and      Walking     Pain-VAS (0-10),         -47%
  Sonel (47)     program     FU: 6 mo
               Control     Pain-VAS (0-10),
                             FU: 6 mo
Evcik and      Walking     NHP-pain (0-100),        -29%
                 program     FU: 6 mo
  Sonel (47)   Control     NHP-pain (0-100),
                             FU: 6 mo
Evcik and      Walking     NHP-energy (0-100),      -63%
  Sonel (47)     program     FU: 6 mo
               Control     NHP-energy (0-100),
                             FU: 6 mo
Evcik and      Walking     NHP-sleep (0-100),       -35%
  Sonel (47)     program     FU: 6 mo
               Control     NHP-sleep (0-100),
                             FU: 6 mo
Peterson et    Walking     AIMS-medication use      116%
  al (74)        program     (0-6), end Tx: 8 wk
               Control     AIMS-medication use
                             (0-6), end Tx: 8 wk
Minor et       Walking     15.2-m (50-ft) walking    -9%
  al (48)        program     time (s), end
                             Tx: 12 wk
               Control     15.2-m (50-ft) walking
                             time (s), end
                             Tx: 12 wk
Minor et       Walking     Aerobic capacity          20%
  al (48)        program     (mL/kg
                             [min.sup.-1]), end
                             Tx: 12 wk
               Control     Aerobic capacity
                             (ml/kg min-1), end
                             Tx: 12 wk
Minor et       Walking     Exercise endurance        13%
  al (48)        program     (min), end Tx: 12 wk
               Control     Exercise endurance
                             (min), end Tx: 12 wk
Minor et       Walking     Exercise heart rate        3%
  al (48)        program     (bpm), end Tx: 12 wk
               Control     Exercise heart rate
                             (bpm), end Tx: 12 wk
Messier et     Exercises   Walking speed (cm/s),      3%
  al (45)                    mid-Tx: 3 mo
               Control     Walking speed (cm/s),
                             mid-Tx: 3 mo
Messier et     Exercises   Walking speed (cm/s),      4%
  al (45)                    mid-Tx: 9 mo
               Control     Walking speed (cm/s),
                             mid-Tx: 9 mo
Messier et     Exercises   Walking speed (cm/s),      7%
  al (45)                    FU: 18 mo
               Control     Walking speed (cm/s),
                             FU: 18 mo
Messier et     Exercises   Cadence (steps/min),       1%
  al (45)                    mid-Tx: 3 mo
               Control     Cadence (steps/min),
                             mid-Tx: 3 mo
Messier et     Exercises   Cadence (steps/min),       1%
  al (45)                    mid-Tx: 9 mo
               Control     Cadence (steps/min),
                             mid-Tx: 3 mo
Messier et     Exercises   Cadence (steps/min),       2%
  al (45)                    FU: 18 mo
               Control     Cadence (steps/min),
                             FU: 18 mo
Messier et     Exercises   Stride length (cm),        2%
                             mid-Tx: 3 mo
  al (45)      Control     Stride length (cm),
                             mid-Tx: 3 mo
Messier et     Exercises   Stride length (cm),        2%
  al (45)                    mid-Tx: 9 mo
               Control     Stride length (cm),
                             mid-Tx: 9 mo
Messier et     Exercises   Stride length (cm),        5%
  al (45)                    FU: 18 mo
               Control     Stride length (cm),
                             FU: 18 mo
Messier et     Exercises   % swing, mid-Tx: 3 mo      1%
  al (45)      Control     % swing, mid-Tx: 3 mo
Messier et     Exercises   % swing, mid-Tx: 9 mo     -1%
  al (45)      Control     % swing, mid-Tx: 9 mo
Messier et     Exercises   % swing, FU: 18 mo         2%
  al (45)      Control     % swing, FU: 18 mo

               Treatment
Study          Group       Outcome                  WMD (95% CI)

Rejeski et     Walking     Stair-climbing time      -1.41 (-1.51,
  al (76)        program     (s), FU: 18 mo          -1.31)
               Control     Stair-climbing time
                             (s), FU: 18 mo
Rejeski et     Walking     Climbing self-efficacy   -8 (-8.45, -7.55)
  al (76)        program     score (0-10),
                             FU: 18 mo
               Control     Climbing self-efficacy
                             score (0-10),
                             FU: 18 mo
Rejeski et     Walking     General health status    -3.59 (-3.72, 3.46)
  al (76)        program     (0-100), FU: 18 mo
               Control     General health status
                             (0-100), FU: 18 mo
Kovar et       Walking     6-min walk test, end     -112 (-161.72,
  al (69)        program     Tx: 8 wk                 -62.28)
               Control     6-min walk test, end
                             Tx: 8 wk
Peterson et    Walking     6-min walk test, end     -111 161.02,
  al (74)        program     Tx: 8 wk                 -60.98)
               Control     6-min walk test, end
                             Tx: 8 wk
Peterson et    Walking     Free speed (m/min),      -7 (-13.66, -0.34)
  al (74)        program     end Tx: 8 wk
               Control     Free speed (m/min),
                             end Tx: 8 wk
Peterson et    Walking     Free stride (m),         -0.2 (-0.32, -0.08)
  al (74)        program     end Tx: 8 wk
               Control     Free stride (m),
                             end Tx: 8 wk
Peterson et    Walking     Fast speed (m/min),      -15 (-23.48, -6.53)
  al (74)        program     end Tx: 8 wk
               Control     Fast speed (m/min),
                             end Tx: 8 wk
Peterson et    Walking     Fast stride (m),         -0.2 (-0.37, -0.03)
  al (74)        program     end Tx: 8 wk
               Control     Fast stride (m),
                             end Tx: 8 wk
Peterson et    Walking     AIMS-physical activity   -2.22 (-3.25,
  al (74)        program     (0-10), end Tx: 8 wk     -1.19)
               Control     AIMS-physical activity
                             (0-10), end Tx: 8 wk
Minor et       Walking     AIMS-physical activity   -1.3 (-2.48, -0.12)
  al (48)        program     (0-10), end
                             Tx: 12 wk
               Control     AIMS-physical activity
                             (0-10), end
                             Tx: 12 wk
Evcik and      Walking     NHP-physical mobility    -28 (-30.77,
  Sone1 (47)     program     (0-100), FU: 6 mo        -25.23)
               Control     NHP-physical mobility
                             (0-100), FU: 6 mo
Evcik and      Walking     WOMAC-physical           -10.5 (-12.41,
  Sonel (47)     program     function (0-68),         -8.59)
                             FU: 6 mo
               Control     WOMAC-physical
                             function (0-68),
                             FU: 6 mo
Peterson et    Walking     AIMS-pain (0-10), end    -1 (-1.79, -0.21)
  al (74)        program     Tx: 8 wk
               Control     AIMS-pain (0-10), end
                             Tx: 8 wk
Evcik and      Walking     WOMAC-pain (0-10),       -2.6 (-3.96, -1.24)
  Sonel (47)     program     FU: 6 mo
               Control     WOMAC-pain (0-10),
                             FU: 6 mo
Evcik and      Walking     Pain-VAS (0-10),         -3.2 (-3.84, -2.56)
  Sonel (47)     program     FU: 6 mo
               Control     Pain-VAS (0-10),
                             FU: 6 mo
Evcik and      Walking     NHP-pain (0-100),        -11.4 (-13.13,
                 program     FU: 6 mo                 -9.67)
  Sonel (47)   Control     NHP-pain (0-100),
                             FU: 6 mo
Evcik and      Walking     NHP-energy (0-100),      -34.7 (-35.51,
  Sonel (47)     program     FU: 6 mo                 -33.89)
               Control     NHP-energy (0-100),
                             FU: 6 mo
Evcik and      Walking     NHP-sleep (0-100),       -15.7 (-17.95,
  Sonel (47)     program     FU: 6 mo                 -13.45)
               Control     NHP-sleep (0-100),
                             FU: 6 mo
Peterson et    Walking     AIMS-medication use      -2.74 (-3.55,
  al (74)        program     (0-6), end Tx: 8 wk      -1.93)
               Control     AIMS-medication use
                             (0-6), end Tx: 8 wk
Minor et       Walking     15.2-m (50-ft) walking   -2 (-2.97, -1.03)
  al (48)        program     time (s), end
                             Tx: 12 wk
               Control     15.2-m (50-ft) walking
                             time (s), end
                             Tx: 12 wk
Minor et       Walking     Aerobic capacity         -5.1 (-7.32, -2.88)
  al (48)        program     (mL/kg
                             [min.sup.-1]), end
                             Tx: 12 wk
               Control     Aerobic capacity
                             (ml/kg min-1), end
                             Tx: 12 wk
Minor et       Walking     Exercise endurance       -3.3 (-5.48, -1.12)
  al (48)        program     (min), end Tx: 12 wk
               Control     Exercise endurance
                             (min), end Tx: 12 wk
Minor et       Walking     Exercise heart rate      -16 (-28.06, -3.94)
  al (48)        program     (bpm), end Tx: 12 wk
               Control     Exercise heart rate
                             (bpm), end Tx: 12 wk
Messier et     Exercises   Walking speed (cm/s),    -3.69 (-4.91,
  al (45)                    mid-Tx: 3 mo             -2.47)
               Control     Walking speed (cm/s),
                             mid-Tx: 3 mo
Messier et     Exercises   Walking speed (cm/s),    -10.29 (-11.53,
  al (45)                    mid-Tx: 9 mo             -9.05)
               Control     Walking speed (cm/s),
                             mid-Tx: 9 mo
Messier et     Exercises   Walking speed (cm/s),    -10.29 (-11.51,
  al (45)                    FU: 18 mo                -9.07)
               Control     Walking speed (cm/s),
                             FU: 18 mo
Messier et     Exercises   Cadence (steps/min),     -3.56 (-4.12, -3)
  al (45)                    mid-Tx: 3 mo
               Control     Cadence (steps/min),
                             mid-Tx: 3 mo
Messier et     Exercises   Cadence (steps/min),     -3.69 (-4.26,
  al (45)                    mid-Tx: 9 mo             (-3.12)
               Control     Cadence (steps/min),
                             mid-Tx: 3 mo
Messier et     Exercises   Cadence (steps/min),     -3.77 (-4.33,
  al (45)                    FU: 18 mo                (-3.21)
               Control     Cadence (steps/min),
                             FU: 18 mo
Messier et     Exercises   Stride length (cm),      0.32 (-0.71, 1.35)
                             mid-Tx: 3 mo
  al (45)      Control     Stride length (cm),
                             mid-Tx: 3 mo
Messier et     Exercises   Stride length (cm),      -7.53 (-8.58,
  al (45)                    mid-Tx: 9 mo             -6.48)
               Control     Stride length (cm),
                             mid-Tx: 9 mo
Messier et     Exercises   Stride length (cm),      -7.54 (-8.56,
  al (45)                    FU: 18 mo                -6.52)
               Control     Stride length (cm),
                             FU: 18 mo
Messier et     Exercises   % swing, mid-Tx: 3 mo    -0.86 (-1.08,
  al (45)      Control     % swing, mid-Tx: 3 mo      -0.64)
Messier et     Exercises   % swing, mid-Tx: 9 mo    -0.37 (-1.7, 0.96)
  al (45)      Control     % swing, mid-Tx: 9 mo
Messier et     Exercises   % swing, FU: 18 mo       -0.54 (-0.76,
  al (45)      Control     % swing, FU: 18 mo         -0.32)

(a) FU=follow-up, N/A=not available, NHP=Nottingham Health Profile,
Tx=treatment, VAS=visual analog scale, WMD=weighted mean difference,
CI=confidence interval, WOMAC=Western Ontario and McMaster Universities
Osteoarthritis Index, AIMS=Arthritis Impact Measurement Scales.

(b) Total number of patients in the study. Study did not specify
number of patients per group.

Table 14.

Walking Program Versus Control (a)

                                               No.
Study        Group       Outcome               Observed     N

Penninx et   Walking     Incidence of          32           88
  al (73)      program     disability in
                           ADL, FU: 18 mo
             Control     Incidence of          42           80
                           disability in
                           ADL, FU: 18 mo
Penninx et   Walking     Disability in         26           88
  al (73)      program     transferring
                           from bed to
                           chair, FU: 18 mo
             Control     Disability in         40           80
                           transferring
                           from bed to
                           chair, FU: 18 mo
Penninx et   Walking     Disability in         11           88
  al (73)      program     bathing, FU:
                           18 mo
             Control     Disability in         22           80
                           bathing, FU:
                           18 mo

                                               Risk         Risk
Study        Group       Outcome               Occurrence   Difference

Penninx et   Walking     Incidence of          36%          -16%
  al (73)      program     disability in
                           ADL, FU: 18 mo
             Control     Incidence of          53%
                           disability in
                           ADL, FU: 18 mo
Penninx et   Walking     Disability in         30%          -20%
  al (73)      program     transferring
                           from bed to
                           chair, FU: 18 mo
             Control     Disability in         50%
                           transferring
                           from bed to
                           chair, FU: 18 mo
Penninx et   Walking     Disability in         13%          -15%
  al (73)      program     bathing, FU:
                           18 mo
             Control     Disability in         28%
                           bathing, FU:
                           18 mo

Study        Group       Outcome               WMD (95% CI)

Penninx et   Walking     Incidence of          0.69 (0.49, 0.98)
  al (73)      program     disability in
                           ADL, FU: 18 mo
             Control     Incidence of
                           disability in
                           ADL, FU: 18 mo
Penninx et   Walking     Disability in         0.59 (0.4, 0.87)
  al (73)      program     transferring
                           from bed to
                           chair, FU: 18 mo
             Control     Disability in
                           transferring
                           from bed to
                           chair, FU: 18 mo
Penninx et   Walking     Disability in         0.45 (0.24, 0.88)
  al (73)      program     bathing, FU:
                           18 mo
             Control     Disability in
                           bathing, FU:
                           18 mo

(a) ADL=activities of daily living, FU=follow-up, WMD=weighted
mean difference, CI=confidence interval.

Table 15.

Clinical Relevance: Jogging in Water Versus Control (a)

            Treatment                             No. of      Baseline
Study       Group        Outcome                  Patients    Mean

Minor et    Jogging in   AIMS-physical            47            4.9
  al (48)     water        activity (0-10),
                           end Tx: 12 wk
            Control      AIMS-physical            32            4.0
                           activity (0-10),
                           end Tx: 12 wk
Minor et    Jogging in   15.2-m (50-ft)           47           10.9
  al (48)     water        walking time (s),
                           end Tx: 12 wk
            Control      15.2-m (50-ft)           32           11.0
                           walking time (s),
                           end Tx: 12 wk
Minor et    Jogging in   Exercise endurance       47           11.5
  al (48)     water        (min), end
                           Tx: 12 wk
            Control      Exercise endurance       32           10.0
                           (min), end Tx: 12 wk
Minor et    Jogging in   Exercise heart rate      47          145
  al (48)     water        (bpm), end Tx: 12 wk
            Control      Exercise heart rate      32          138
                           (bpm), end Tx: 12 wk

                                                  End-of-
            Treatment                             Study       Absolute
Study       Group        Outcome                  Mean        Benefit

Minor et    Jogging in   AIMS-physical              3.7        -2.1
  al (48)     water        activity (0-10),
                           end Tx: 12 wk
            Control      AIMS-physical              4.9
                           activity (0-10),
                           end Tx: 12 wk
Minor et    Jogging in   15.2-m (50-ft)             9.6        -1
  al (48)     water        walking time (s),
                           end Tx: 12 wk
            Control      15.2-m (50-ft)            10.7
                           walking time (s),
                           end Tx: 12 wk
Minor et    Jogging in   Exercise endurance        14.3         1.3
  al (48)     water        (min), end
                           Tx: 12 wk
            Control      Exercise endurance        11.5
                           (min), end Tx: 12 wk
Minor et    Jogging in   Exercise heart rate      155           6
  al (48)     water        (bpm), end Tx: 12 wk
            Control      Exercise heart rate      142
                           (bpm), end Tx: 12 wk

                                                  Relative Difference
            Treatment                             in Change From
Study       Group        Outcome                  Baseline

Minor et    Jogging in   AIMS-physical            -46%
  al (48)     water        activity (0-10),
                           end Tx: 12 wk
            Control      AIMS-physical
                           activity (0-10),
                           end Tx: 12 wk
Minor et    Jogging in   15.2-m (50-ft)             9%
  al (48)     water        walking time (s),
                           end Tx: 12 wk
            Control      15.2-m (50-ft)
                           walking time (s),
                           end Tx: 12 wk
Minor et    Jogging in   Exercise endurance        12%
  al (48)     water        (min), end
                           Tx: 12 wk
            Control      Exercise endurance
                           (min), end Tx: 12 wk
Minor et    Jogging in   Exercise heart rate        4%
  al (48)     water        (bpm), end Tx: 12 wk
            Control      Exercise heart rate
                           (bpm), end Tx: 12 wk

            Treatment
Study       Group        Outcome                  WMD (95% CI)

Minor et    Jogging in   AIMS-physical            -1.2 (-2.29, -0.11)
  al (48)     water        activity (0-10),
                           end Tx: 12 wk
            Control      AIMS-physical
                           activity (0-10),
                           end Tx: 12 wk
Minor et    Jogging in   15.2-m (50-ft)           -1.1 (-2.12, -0.08)
  al (48)     water        walking time (s),
                           end Tx: 12 wk
            Control      15.2-m (50-ft)
                           walking time (s),
                           end Tx: 12 wk
Minor et    Jogging in   Exercise endurance       -2.8 (-5.37, -0.23)
  al (48)     water        (min), end
                           Tx: 12 wk
            Control      Exercise endurance
                           (min), end Tx: 12 wk
Minor et    Jogging in   Exercise heart rate      -13 (-24.68, -1.32)
  al (48)     water        (bpm), end Tx: 12 wk
            Control      Exercise heart rate
                           (bpm), end Tx: 12 wk

(a) AIMS=Arthritis Impact Measurement Scales, Tx=treatment,
WMD=weighted mean difference, CI-confidence interval.

Table 16.

Clinical Relevance: Manual Therapy and Exercise Versus Control (a)

                                                 No. of      Baseline
Study       Treatment Group   Outcome            Patients    Mean

Deyle et    Manual therapy    WOMAC-pain,        42          1,046.7
  al (30)     and exercise      end Tx: 4 wk
            Subtherapeutic    WOMAC-pain,        41          1,093.5
              ultrasound        end Tx: 4 wk
              (placebo)
Deyle et    Manual therapy    WOMAC-pain,        42          1,046.7
  al (30)     and exercise      end Tx: 8 wk
            Subtherapeutic    WOMAC-pain,        41          1,093.5
              ultrasound        end Tx: 8 wk
              (placebo)
Deyle et    Manual therapy    6-min walk test,   42            431.0
  al (30)     and exercise      end Tx: 4 wk
            Subtherapeutic    6-min walk test,   41            402.9
              ultrasound        end Tx: 4 wk
              (placebo)
Deyle et    Manual therapy    6-min walk test,   42            431.0
  al (30)     and exercise      FU: 8 wk
            Subtherapeutic    6-min walk test,   41            402.9
              ultrasound        FU: 8 wk
              (placebo)

                                                 End-of-
                                                 Study       Absolute
Study       Treatment Group   Outcome            Mean        Benefit

Deyle et    Manual therapy    WOMAC-pain,        505.2       -369.2
  al (30)     and exercise      end Tx: 4 wk
            Subtherapeutic    WOMAC-pain,        921.2
              ultrasound        end Tx: 4 wk
              (placebo)
Deyle et    Manual therapy    WOMAC-pain,        462.4       -425.1
  al (30)     and exercise      end Tx: 8 wk
            Subtherapeutic    WOMAC-pain,        934.30
              ultrasound        end Tx: 8 wk
              (placebo)
Deyle et    Manual therapy    6-min walk test,   484.0         53.8
  al (30)     and exercise      end Tx: 4 wk
            Subtherapeutic    6-min walk test,   402.1
              ultrasound        end Tx: 4 wk
              (placebo)
Deyle et    Manual therapy    6-min walk test,   487.4         49.6
  al (30)     and exercise      FU: 8 wk
            Subtherapeutic    6-min walk test,   409.7
              ultrasound        FU: 8 wk
              (placebo)

                                                 Relative Difference
                                                 in Change From
Study       Treatment Group   Outcome            Baseline

Deyle et    Manual therapy    WOMAC-pain,        -34%
  al (30)     and exercise      end Tx: 4 wk
            Subtherapeutic    WOMAC-pain,
              ultrasound        end Tx: 4 wk
              (placebo)
Deyle et    Manual therapy    WOMAC-pain,        -40%
  al (30)     and exercise      end Tx: 8 wk
            Subtherapeutic    WOMAC-pain,
              ultrasound        end Tx: 8 wk
              (placebo)
Deyle et    Manual therapy    6-min walk test,    13%
  al (30)     and exercise      end Tx: 4 wk
            Subtherapeutic    6-min walk test,
              ultrasound        end Tx: 4 wk
              (placebo)
Deyle et    Manual therapy    6-min walk test,    12%
  al (30)     and exercise      FU: 8 wk
            Subtherapeutic    6-min walk test,
              ultrasound        FU: 8 wk
              (placebo)

Study       Treatment Group   Outcome            WMD (95% CI)

Deyle et    Manual therapy    WOMAC-pain,        -416 (-618.15,
  al (30)     and exercise      end Tx: 4 wk       -213.85)
            Subtherapeutic    WOMAC-pain,
              ultrasound        end Tx: 4 wk
              (placebo)
Deyle et    Manual therapy    WOMAC-pain,        -471.9 (-732.74,
  al (30)     and exercise      end Tx: 8 wk       -211.06)
            Subtherapeutic    WOMAC-pain,
              ultrasound        end Tx: 8 wk
              (placebo)
Deyle et    Manual therapy    6-min walk test,   -81.9 (-141.04,
  al (30)     and exercise      end Tx: 4 wk       -22.76)
            Subtherapeutic    6-min walk test,
              ultrasound        end Tx: 4 wk
              (placebo)
Deyle et    Manual therapy    6-min walk test,   -77.7 (-136.81,
  al (30)     and exercise      FU: 8 wk           -18.59)
            Subtherapeutic    6-min walk test,
              ultrasound        FU: 8 wk
              (placebo)

(a) FU=follow-up, WMD=weighted mean difference, CI=confidence interval,
WOMAC=Western Ontario and McMaster Universities Osteoarthritis Index,
Tx=treatment.