Osteoarthritis is the most common type of arthritis.  It mainly affects the knees, hips, back, neck and fingers. People with symptoms complain of deep, aching, pain in their joints. The pain is usually occasional and mostly associated with physical activity.  The pain can, however, become more persistent, affecting daily activities and ability to sleep. Affected joints become stiff, which makes it difficult to start moving after a period of rest. With advanced disease, a joint may have a limited range of movement and deformities, such as bowing of the legs, may occur. The symptoms may make it very difficult to carry out daily functional activities, cause stress and emotional difficulties and reduce quality of life. Physical fitness is compromised because of the reduced ability to move about, making it difficult to control body weight and increasing the risk of cardiovascular disease.

The most common features of osteoarthritis that appear on x-rays are bone spurs at the joint margins, joint space narrowing as the protective cartilage layer breaks down and changes in the bone underneath the cartilage layer.

The aim of treatment is to reduce the pain and improve mobility. Drug therapy alone is often not effective and may cause adverse effects, such as stomach upsets with non-steroidal anti-inflammatory drugs (NSAIDs), and kidney problems with persistent high use of conventional painkillers (paracetamol and codeine). Application of heat or ice packs, exercise and physical therapies, glucosamine and complementary therapies can, therefore, become an important part of day-to-day management of osteoarthritis.

We present some of the evidence from Cochrane systematic reviews about complementary and alternative treatments related to arthritis. This evidence comes from carefully researched reviews of information about clinical trials done to evaluate medical treatments. Studies are only included in these reviews if they meet pre-defined criteria.


Altered forces applied to the bones and joints by muscles and gravity, poor joint alignment and reduced muscle strength all contribute to increased joint loading and stress and are important in both the initiation and progression of osteoarthritis. Non-drug treatments that are currently considered to have sufficient level of scientific evidence for their use are education, exercise, appliances (canes or sticks, insoles) and weight reduction. 

A wide range of therapeutic exercise programs are used. Exercises vary from simple thigh (quadriceps) muscle strengthening using straight leg raises, walking through to very complex programs that include manual therapy, upper limb and trunk muscle strengthening, balance and coordination training. Exercise dosage is an important factor that defines the frequency of exercising, the intensity, and the exercise program duration.

Long-term adherence to exercise or physical activity regimens is needed to maintain the benefits of exercise. This means that regular supervision or monitoring may be required. A group format could be accessible if introduced to community centres or gymnasiums and would provide social contact that encourages exercising.

It has been demonstrated that regular exercise has many other physical and mental health benefits apart from those related to arthritis.

What the synthesised research says

For pain

Land-based therapeutic exercise was shown to reduce self-reported pain and improve physical function for people with osteoarthritis of the knee. Supervised exercise classes appeared to be as beneficial as treatments provided on a one-to-one basis. The effect size on self-reported pain was small for the four studies assessing home programs.

Seventeen controlled studies provided data on 2562 people with osteoarthritis of the knee who were randomly assigned to either an exercise program delivered individually (five trials), a group exercise program (7 trials), or a home based, individual exercise program (5 trials) compared with no exercise program (standard treatment). Only 2 trials (around 100 participants) provided results for osteoarthritis of the hip.

Pain and physical function

A small beneficial effect was seen with exercise. Group programs appeared to be as effective as programs provided on a one-to-one basis.

The overall, average effect size for both outcomes was small compared to reported estimates for current drug treatments.

How it was tested

The researchers thoroughly searched the medical literature for the 17 studies that met the inclusion criteria. Some studies had less than 50 participants and some were conducted less well than others.

There was a wide variability between studies in the severity of osteoarthritis (from the x-ray and symptoms reported) and the use of non-steroidal anti-inflammatory drugs (NSAIDs); some studies allowed people to continue to take anti-inflammatory drugs and others did not.

Side effects and general cautions

Many of the included studies primarily involved participants with early or mild symptomatic disease. This alters a person’s perception of improvement and the degree to which they can benefit from an exercise program. One of the potential benefits of exercise in people with early disease is increasing their physiological reserve capacity and this will not be captured by questionnaires used in the studies.

The amount of exercise varied considerably between the studies included. A certain dosage is not universally beneficial for all people with osteoarthritis. It also depends on factors such as age, degree of musculoskeletal damage and presence of other illnesses such as a heart condition.


Fransen M, McConnell S, Bell M. Exercise for osteoarthritis of the hip or knee. The Cochrane Database of Systematic Reviews 2001, Issue 2. Art. No.: CD004376. DOI: 10.1002/14651858.CD004376.


Painkillers and anti-inflammatory drugs alone often do not control symptoms adequately and can cause adverse effects such as stomach upsets and constipation. This has prompted the evaluation of a number of alternative treatments for arthritis, including pulsed electromagnetic field (PEMF) and pulsed electrical stimulation or electrotherapy. Different physiotherapy treatments help improve clinical symptoms and people’s ability to carry out activities of daily living.

What is known

For pulsed electromagnetic field stimulation (PEMF), electrodes are placed over the affected area (with or without direct skin contact) and an induced current created through magnetic impulses. In pulsed electrical stimulation, or electrotherapy, electrodes are placed over the affected area and an electrical field is applied directly.

People became interested in PEMF for osteoarthritis when scientists discovered that tiny electric currents are present in bones and joints when they are moving or in use. As a result PEMF stimulation has been used for bone repair. For joints, PEMF might stimulate re-growth of cartilage (the protective cushioning covering the opposing bone surfaces), and increase the lubrication of joints (by increasing synthesis of proteoglycans). The joints are in a cavity that contains slippery, viscous synovial fluid, which also acts as a lubricant within the cartilage and forms a major part of its matrix.

What the synthesised research says

Both PEMF treatment and electrical stimulation therapy improved knee pain (by about 1 cm on a 10 cm Visual Analogue Scale). Also improved were the joint pain with passive movement, tenderness of the joint, difficulty in carrying out daily activities and the pain caused, to a similar degree. This meant that the overall disease activity was assessed as better whether it was the individual or a physician who made the assessment, in two studies.

Improvements of osteoarthritis in the neck (cervical spine) were less than for the knee (in one study that looked at both the knee and the neck).

How it was tested

The researchers made a thorough search of the medical literature and found 3 studies that randomly assigned 259 people to receive either active electrotherapy or a sham procedure. The treatments for two PEMF trials lasted approximately a month, giving a total of eighteen 30-minute treatments. The treatments in the third trial were more intensive, involving 4 weeks of 6 hour-a-day treatments.

Electrotherapy was not given in the same way in these studies. The two PEMF studies used a non-contact device that delivered three signals in stepwise fashion (5 to 12 hertz at 10 G to 25 G of magnetic energy). The third study used a pulsed electric device to deliver a 100 hertz low-amplitude signal to the knee via skin surface electrodes.

Side effects and general cautions

Similar numbers of people withdrew from the studies whether or not they were receiving active or sham stimulation. In one study, skin reactions were reported in both groups and appeared to be caused by the conducting gel applied to the skin.

The studies were well conducted. Two of the studies were performed by the same group and funded by a manufacturer of PEMF. How much the measured changes are of benefit to the person with osteoarthritis on a daily basis is difficult to assess.


Hulme J, Robinson V, DeBie R, Wells G, Judd M, Tugwell P. Electromagnetic fields for the treatment of osteoarthritis. The Cochrane Database of Systematic Reviews 2002, Issue 1. Art. No.: CD003523. DOI: 10.1002/14651858.CD003523.


Osteoarthritis (OA) is the most common form of arthritis affecting many joints including the hands, hips, shoulders and knees. It is a degenerative process involving breakdown of the cartilage that protects the ends of the bones at the joints. This causes pain, swelling, reduced range of motion and stiffness. Drug and non-drug treatments are used to relieve the pain and swelling.

What is known

Glucosamine can be found naturally in the body and is one of the building blocks of cartilage. Taking glucosamine supplements may slow or stop cartilage breakdown.

What the synthesised research says

Glucosamine has some ability to reduce the pain of arthritis in the hip or knee.

In an earlier version of this Cochrane review, glucosamine taken over six weeks decreased pain and improved physical ability (joint function) in people with osteoarthritis. This update analysed newer studies that were more strictly carried out over a treatment period of two to three months. The pain relief with glucosamine was less than seen previously. Ability to carry out activities (that is, function) may not improve as much, if at all (and varied with the different scales to measure function).

How it was tested

The researchers made a thorough search of the medical literature and found 20 controlled studies. Over 2500 people with osteoarthritis of the knee or hip were randomly assigned to receive either glucosamine (by mouth or injection), non-active treatment (placebo) or a non-steroidal anti-inflammatory drug (NSAID). Most of the studies were two to three months long.

For pain, glucosamine improved pain more than placebo (by 13 points on a scale of 0 to 100). Considering the one high quality studies only, pain improved about the same on both glucosamine and placebo. The Italian Rotta brand of glucosamine improved pain more than other brands.

Side effects and general cautions

The number of people who reported side effects were about the same whether they were taking glucosamine or the non-active placebo. These side effects included stomach upset and pain in other joints.


Towheed TE, Maxwell L, Anastassiades TP, Shea B, Houpt J, Robinson V, Hochberg MC, Wells G. Glucosamine therapy for treating osteoarthritis. The Cochrane Database of Systematic Reviews 2005, Issue 2. Art. No.: CD002946.pub2. DOI: 10.1002/14651858.CD002946.pub2.


Osteoarthritis is a common condition and can cause progressive disability in some people. People often find conventional treatments for arthritis to have limitations in their effectiveness and unpleasant side effects, which leads to widespread use of complementary and alternative medicines.

What is known

People with chronic diseases often look to complementary medicine. In the treatment of osteoarthritis, plant and herbal therapies (phytotherapy) may have benefit. The Chinese herb Thunder God vine (Radix Tripterygium wilfordii hook F) has been used for rheumatoid arthritis for centuries whilst European herbalists claim that several different herbs and popular arthritis remedies that include devil's claw (Harpagophytum species), plant seed oils such as evening primrose and borage, and Salix species are effective. An avocado and soybean mixture, willow bark and tipi tea are others

What the synthesised research says

An avocado and soybean mixture taken over three to six months provided long-term symptomatic relief for people with osteoarthritis of the hip or knee, particularly on ability to function. Pain was less and people could reduce their consumption of non-steroidal anti-inflammatory drugs (NSAIDs). People with osteoarthritis of the hip consistently improved more than people with osteoarthritis of the knee, in two studies.

Single studies of a willow bark preparation (Reumalex), capsaicin cream and tipi tea were insufficient to either recommend or discourage their use.

Reumalex is a licensed over-the-counter willow bark herbal remedy. One study compared Reumalex to a calcium phosphate placebo over two months and reported a slight improvement in chronic arthritic pain relief with Reumalex. People continued to take any other self medication they were on.

In one study, capsaicin cream (0.025%) gave pain relief to people with moderate to severe knee pain caused by osteoarthritis or rheumatoid arthritis. The cream was applied to the skin four times daily over four weeks. The people also kept on their normal arthritis medications, taken by mouth.

A 200 ml drink of tipi tea (Petiveria alliacea) drunk three times daily for a week improved pain and the time to walk 15 metres, as did the alternative tea.

How it was tested

The researchers made a thorough search of the medical literature and found five randomised studies with four different herbal interventions meeting the review criteria.

Two studies compared a total of 327 people with osteoarthritis of the hip or knee randomly assigned to receive the avocado-soybean mixture or non-active treatment (placebo).

Of the 72 participants who completed the study on Reumalex, 52 had osteoarthritis and 20 had rheumatoid arthritis. People received either two Reumalex tablets 'at a time' or two placebo tablets.

A total of 101 people with osteoarthritis (a total of 70) or rheumatoid arthritis (31 people) were randomly assigned to receive the active capsaicin cream or a non-active (placebo) treatment. The capsaicin cream caused a sensation of burning skin where it was applied in less than half of people who received active treatment; this did not affect the treatment outcome.

In the one study on tipi tea, 20 participants with hip or knee osteoarthritis drank either tipi tea (9 g per day) or Imperata exaltata (Sape) as placebo tea. They then had no tea for a week before crossing over to the other tea, also for a week.

Side effects and general cautions

No serious side effects were reported.

In general, the herbal interventions seemed to be relatively well tolerated. In one of the avocado and soya studies most side effects were of a gastrointestinal nature and occurred during the first 45 days, whilst NSAIDs were still being taken, suggesting a likelihood of side effects being related to the NSAID rather that the avocado and soya.

Herbal therapy in conjunction with other treatments or combined with a non-herbal substance was excluded from this review. 'Herbal intervention' included any whole plant extract but excluded homeopathy, aromatherapy, or any preparation of synthetic origin or consisting only of plant derivative(s).


Little CV, Parsons T, Logan S. Herbal therapy for treating osteoarthritis. The Cochrane Database of Systematic Reviews 2000, Issue 4. Art. No.: CD002947. DOI: 10.1002/14651858.CD002947.


Management of knee osteoarthritis aims to relieve pain, maintain or improve mobility and minimise disability. Treatment options include non-drug interventions including exercise, drug therapy and surgery.

What is known

Different physiotherapy treatments help improve clinical symptoms and people’s ability to carry out activities of daily living. A patient may receive TENS (transcutaneous electrical nerve stimulation) as part of the treatment. TENS machines are also available outside of physical therapy. Where TENS is self-administered, a person has a portable unit and places the electrodes on the painful areas, and carries on with his or her daily activities.

TENS therapy is used to treat a variety of painful conditions, both sudden in onset (acute) and longer-term or chronic. Its effectiveness in relieving pain is thought to be due to its ability to modulate nerve impulses, including blocking nerve fibre pain signals (to the spinal cord). TENS may also stimulate our natural painkillers (the body’s own morphine-like substances).

A TENS device can be set to give different types of stimulation sent through electrodes placed on the skin. It is important that these electrodes are carefully positioned to give maximum benefit. In clinical practice, stimulation can be high frequency (40 to 150 hertz) or low frequency (1 to 4 hertz), also in bursts or with long pulse widths (hyperstimulation). Strong burst mode TENS stimulates people’s natural painkiller secretion and causes rhythmic muscle contractions. Acupuncture-like TENS (AL-TENS) is low frequency, high intensity pulses applied to acupuncture points. AL-TENS has been shown to increase a person’s threshold to pain.  

What the synthesised research says

Receiving conventional TENS over at least four weeks effectively controls knee pain and relieves the stiffness associated with osteoarthritis. Repeated applications of conventional TENS were required to show benefit.

The participants in the TENS treatment group were at least twice as likely to have pain improvement as those receiving sham TENS (placebo). This benefit continued after therapy had finished. Different modes of TENS settings (high frequency and strong burst mode) all gave pain relief in osteoarthritis of the knee.

Acupuncture-like TENS is effective in pain relief, decreasing stiffness and walking time, improving quadriceps muscle strength and bending of the knee (flexion) after two weeks.

How it was tested

The researchers made a thorough search of the medical literature and identified 7 controlled studies. These involved 294 participants who were randomly assigned to receive active TENS or sham treatment (placebo). Four studies used conventional high frequency and 2 used strong burst mode TENS as the active treatment; and one study used acupuncture-like TENS (AL-TENS). The pain relief of strong burst mode TENS and AL-TENS was approximately two times better than with high frequency TENS.

When AL-TENS group compared to placebo, pain, knee stiffness, the time taken to walk a short distance (50 feet), thigh (quadriceps) muscle strength and knee flexion were all improved with active treatment.

No clear overall benefit was seen on other outcomes with other forms of TENS but this might be explained by the low quality of the methods used in the included studies, a wide variety of TENS devices used and application protocols, or inadequate intervention periods (short duration studies).

Although the included studies were all published articles, the number with results that could have occurred by chance (4 studies) was comparable to those with positive results (3 studies). Pain relief was apparent in the studies with higher methodological quality, both modes of TENS device settings, repeated TENS application, and an intervention period of TENS application of at least four weeks.

Side effects and general cautions

No study reported side effects of the TENS treatment. This might be because TENS appears to be relatively safe.

People with surgery on the affected knee were excluded from the studies.

Variability in the disease (the stage and severity of knee osteoarthritis) and the people (such as their lifestyle, having other diseases and use of various medications) might exist. TENS was applied with varying intensities.


Osiri M, Brosseau L, McGowan J, Robinson VA, Shea BJ, Tugwell P, Wells G. Transcutaneous electrical nerve stimulation for knee osteoarthritis. The Cochrane Database of Systematic Reviews 2000, Issue 4. Art. No.: CD002823. DOI: 10.1002/14651858.CD002823.


Arthritic conditions are painful. People often find conventional treatments for arthritis to have limitations in their effectiveness and unpleasant side effects, which leads to widespread use of complementary and alternative medicines. Although the source of inflammatory pain is in the joints, ligaments, and muscles, it is the central nervous system where the perception and interpretation of pain takes place. Different physiotherapy treatments are used to help improve clinical symptoms and people’s ability to carry out activities of daily living.

What is known

Low level laser therapy is a light source that generates extremely pure light of a single wavelength. It is a non-invasive therapy applied to the skin and used in medicine and physiotherapy in the treatment of osteoarthritis to reduce inflammation and pain. The effect is not thermal but rather related to photochemical reactions in cells. Laser therapy may positively modify the sensory input to the central nervous system and provide an improvement in the perception of pain localized to the area of treatment, which would mean that it is most effective when nerves as well as joints are treated.

What the synthesised research says

Laser therapy was not uniformly effective in relieving the pain of osteoarthritis. Furthermore, the laser therapy was applied differently in the different studies.

Three trials which assessed people with osteoarthritis in various joints found no improvement in the pain; two found an improvement; one reported an improvement in pain, morning stiffness and ability to bend the knee (range of motion). In this last study people also undertook an exercise program over the twelve weeks of treatment, which is likely to have benefit of its own.

No clear benefits were seen on joint mobility or muscle strength with laser therapy. This may be due to lack of data. Only four of the seven studies measured muscle strength and range of motion, with one study finding an increased knee range of motion of some 11 degrees.

How it was tested

The researchers made a thorough search of the medical literature and found seven controlled studies in which a total of 345 people were randomly assigned to receive active laser or sham treatment (placebo). Their average age ranged from 56 to 74 years. Studies included people with osteoarthritis of the thumb (one study), hand (one study), knee (four studies) and at unspecified joints (one study).

The length of the treatment ranged from 4 to 12 weeks.

Side effects and general cautions

One person reported an area of severe redness of the skin (erythema) with treatment.

The studies were small – the largest study randomised only 47 people to receive laser therapy.

The most effective way of delivering laser therapy could not be obtained from these studies.


Brosseau L, Gam A, Harman K, Morin M, Robinson VA, Shea BJ, Tugwell P, Wells G, de Bie RA. Low level laser therapy (Classes I, II and III) for treating osteoarthritis. The Cochrane Database of Systematic Reviews 2004, Issue 3. Art. No.: CD002046. DOI: 10.1002/14651858.CD002046.pub2.


Chronic knee pain is one of the most common reasons for visits to a general practitioner and is often the result of osteoarthritis. Rehabilitation programs are recommended to many osteoarthritis patients. Therapeutic ultrasound is one of many physical therapy treatments available and is often used as part of such a rehabilitation program.

What is known

Therapeutic ultrasound is a mechanical energy consisting of high frequency vibrations that can be continuous or pulsed and is applied to the affected area. Pulsed ultrasound produces non-warming (thermal) effects and is used to help reduce inflammation during a flare of inflammation. Continuous ultrasound generates thermal effects and is for chronic conditions. Therapeutic ultrasound that penetrates deeply enough to increase connective tissue (collagen) elasticity may be useful in the early stages of a flexibility program.

What the synthesised research says

Ultrasound therapy did not appear to give pain relief, or improve range of motion or functional status. Ultrasound therapy appears to have no benefit over sham treatment (placebo) or short wave diathermy for people with hip or knee osteoarthritis, in three studies with a total of 147 participants.

Only one study with 74 participants compared therapeutic ultrasound to placebo. This trial showed no difference in range of motion, pain or walking speed after four weeks of therapeutic ultrasound.

Therapeutic ultrasound and two other electrotherapy modalities, short wave diathermy and galvanic current, were equivalent in reducing pain and an individual assessment of improvement. This conclusion is from two studies (220 participants) that compared therapeutic ultrasound to these other active therapies.

For continuous therapeutic ultrasound versus galvanic current, there was greater pain relief with therapeutic ultrasound but no difference in the doctor’s assessment of a patient.

How it was tested

The researchers made a thorough search of the medical literature and found 3 controlled studies (undertaken between 1975 and 1992). These included 294 participants with hip or knee osteoarthritis. People were aged from 40 to 85 years, with an average age of around 65 years in 2 of the studies.

Only one study with 74 participants compared therapeutic ultrasound to placebo. This trial showed no difference in range of motion, pain or walking speed after four weeks of therapeutic ultrasound. Treatment was given two to three times per week for four to six weeks. All participants in this study were prescribed stretching and strengthening exercises.

Two other studies compared therapeutic ultrasound to an active therapy. One assigned 180 people with hip or knee osteoarthritis to therapeutic ultrasound, galvanic current or short-wave diathermy. The other randomly assigned 60 participants to continuous therapeutic ultrasound or short wave diathermy.

Side effects and general cautions

The review of evidence was limited by poor reporting of how the ultrasound was delivered, the people who were included in the studies and the severity of their symptoms, what other treatments they were on and how the studies were carried out. One study used no other treatment; another included exercises, education and manual therapy; and the third allowed the use of the NSAID ibuprofen.

Pulsed therapeutic ultrasound is used for an acute condition in order to reduce inflammation while continuous therapeutic ultrasound is recommended for more persistent (chronic) conditions.

No conclusions can be drawn about the use of ultrasound in smaller joints such as the wrist or hands.


Robinson VA, Brosseau L, Peterson J, Shea BJ, Tugwell P, Wells G. Therapeutic ultrasound for osteoarthritis of the knee. The Cochrane Database of Systematic Reviews 2001, Issue 3. Art. No.: CD003132. DOI: 10.1002/14651858.CD003132.