Can Magnetic Bracelets and Magnetic Therapy help with Pain Relief?

Can Magnetic Bracelets and Magnetic Therapy help with Pain Relief?

Is this an Approach Worth Trying? Each Individual Must Decide for Themselves

A Balanced View Point?

Our Personal Opinion: In judging whether a particular treatment is worth trying we believe in taking a broad view of the possible sources of evidence. From our point of view a balanced view is an inclusive approach, this means taking into account information, evidence and opinions from a wide variety of sources. These sources of information are, western medical research, anecdotal reports from actual users, complementary and alternative health understandings and feedback from therapists in the field. When we look at all these sources our personal view is that magnetic therapy is a therapy worth trying.

For example, Acupuncture has an extremely long history within China that goes back thousands of years. It eventually came to the West and was initially viewed with much scepticism. Treatments were tried for a whole range of conditions, and over time a tremendous amount of positive anecdotal evidence was accumulated that prompted reseasrch. Initial research was conducted, mostly with negative results, latter more extensive and substantial research was undertaken. Over time here in the UK Acupuncture has become an accepted treatement for many conditions. It is also offered in the NHS, and for example many people regard it as one of the treatments of choice for back problems. Patients at the beggining of this process relied on evidence of anecdotal reports from others to embark on a course of treatment, if they would have waited for research confirmation, they may well have missed out on an effective treatment. They would have waited in fact for decades if they based treatment choice only on reasearch.

The western medical approach relies completely on peer reviewed double blind trials. This is as it should be especially if treatments are funded by the state via the NHS. Why should the NHS support and fund a treatment unless there is reasonable evidence of a positive outcome. There is already too much competition for new treatments on limited budgets.

But when it comes to a patients individual choice of possible treatments which they are willing to pay for. Then within the limits of the law and as long as it is safe, it should be up to the individual what treatment they choose. This has been the situation in relation to the complementary and alternative therapies, many of which have limited research. Therapies such as Acupuncture, Homeopathy, Osteopathy, Chiropractic and all the various Herbal therapies and in this case also Magnetic Therapy and the use of Magnetic Bracelets.

A Summary of View Points

Western Research: The biggest proportion of the western medical research budget is spent on the drug biochemical approach. Generally speaking, most companies do not want to undertake research unless the end result is a product that has the possibility of a unique money making patent.  In general it is not possible to apply for patents on products with standard magnets as the level of invention is not regard as sufficiently novel or new. As a result there is not a huge amount of available research on the complementary health approaches including magnetic therapy and magnetic bracelets. However there has been some. In our view there is no one definitive piece of research, the information must be looked at as a whole where possible. An example of one piece of research is given below.

Research into Magnetic Bracelets and Magnetic Therapy can be complex as there are a large number of variables. Many of these variables are not considered in the research undertaken and a simplistic approach is taken. Variables such as number of patients in study (often very small) - patient disease type, length of disease, age, individual constitution (strong, weak), therapy variables, length of past treatments, methods of pain measurement etc. In terms of the magnetic therapy only limited attention is given to magnet strength, magnetic size (volume, areal extent, thickness), magnet position, magnet polarity and number of magnets.

Anecdotal Reports: Some discrimination can be given to anecdotal reports. Quite often they are completely dismissed by the scientific community. Patients often know there symptoms very well, and know if a treatment is working well, This should not be dismissed. Levels of improvement  in relation to placebo,(no change, small/moderate, significant). Patients should be give people credit for knowing there own symptoms and changes. Anybody who has worked in the magnetic therapy field, has seen patients who have had dramatic improvements. But like any therapeutic intervention, results can vary from no improvement to a significant improvement. Each individual must decide for themselves.

Complementary Medicine view points: For more information on some of the complementary health viewpoints on magnetic therapy see our information page below:

Click here for how magnetic therapy may work

Research: In response to a large number of people with arthritis reporting pain relief from using a magnetic bracelet, the Arthritis Research Council decided to fund a "Randomised controlled trial of magnetic bracelets for relieving pain in osteoarthritis of hip and knee." The BMJ published an article 18 December 2004 with reference to this trial. The outcome of this trial was widely reported in the press at the time.

What this Study Adds: The research indicated encouraging results. Bracelets with static magnets decrease the pain from osteoarthritis of the hip and knee, over and above the effects of placebo.

Anecdotal Evidence: Large amounts of positive anecdotal evidence from users of magnetic bracelets and magnetic therapy around the world should in our view not be ignored.

Magnetic Bracelets can give Pain Relief to Many Areas of the Body: As indicated in the research above and anecdotal evidence, our personal opinion is that magnetic bracelets can help alleviate pain not only in the wrist area, but in many other areas of the body including knees, hip and other joints. This view point is not accepted by all medical scientists.

How do they Work? Although not fully understood, a common theory to explain how magnetic bracelets work is the ionising effect of the magnets on blood flowing through the wrist arteries, the improved blood quality flows throughout the body with resultant healing effects. There are also other theories, some based on complementary health theories.

Further research: More research is needed to confirm the initial findings and determine other key variables such as the importance of magnetic strength, size and position.

Disclaimer: We make no specific medical claims. It is always best to seek professional medical advice on an individual basis for your particular circumstances.

The Actual Research: Below is a detailed copy of the research paper published by the BMJ on 18th December 2004. We have included this reaseach paper as it uses nealy 200 patients, and the magnetic bracelets used contained high power neodymium magnets, although no atention has been given to the position of the magnets or the physical size of the magnets.

BMJ 2004;329:1450-1454 (18 December), doi:10.1136/bmj.329.7480.1450

Randomised controlled trial of magnetic bracelets for relieving pain in osteoarthritis of the hip and knee

Tim Harlow, general practitioner1, Colin Greaves, research fellow2, Adrian White, senior research fellow3, Liz Brown, research assistant4, Anna Hart, statistician5, Edzard Ernst, professor of complementary medicine4

1 College Surgery, Cullompton, Devon EX15 1TG, 2 Peninsula Medical School (Primary Care), Exeter EX2 5DW, 3 Peninsula Medical School, Tamar Science Park, Plymouth PL6 8BX, 4 Peninsula Medical School (Complementary Medicine), Exeter EX2 4NT, 5 Lancashire School of Health and Postgraduate Medicine, University of Central Lancashire, Preston PR1 2HE

Objective To determine the effectiveness of commercially available magnetic bracelets for pain control in osteoarthritis of the hip and knee.

Design Randomised, placebo controlled trial with three parallel groups. Setting Five rural general practices.

Participants 194 men and women aged 45-80 years with osteoarthritis of the hip or knee.

Intervention Wearing a standard strength static bipolar magnetic bracelet, a weak magnetic bracelet, or a non-magnetic (dummy) bracelet for 12 weeks.

Main outcome measures Change in the Western Ontario and McMaster Universities osteoarthritis lower limb pain scale (WOMAC A) after 12 weeks, with the primary comparison between the standard and dummy groups. Secondary outcomes included changes in WOMAC B and C scales and a visual analogue scale for pain.

Results Mean pain scores were reduced more in the standard magnet group than in the dummy group (mean difference 1.3 points, 95% confidence interval 0.05 to 2.55). Self reported blinding status did not affect the results. The scores for secondary outcome measures were consistent with the WOMAC A scores.

Conclusion Pain from osteoarthritis of the hip and knee decreases when wearing magnetic bracelets. It is uncertain whether this response is due to specific or non-specific (placebo) effects.

Manufacturers of permanent static magnet devices claim that they reduce pain in various conditions, including osteoarthritis. 1 Worldwide sales were estimated at $5bn (£2.6bn, 3.8bn) in 1999.2 Osteoarthritis affects around 760 000 people in the United Kingdom, producing an estimated 3.02 million general practice consultations in 2000. 3 If magnets were effective they would offer a cheap and probably safe treatment option.

Some studies of permanent static magnets have found significant pain reduction2 4-9 whereas others reported no effect.10-12 Major differences exist in the type and strength of magnets used, the conditions treated, and treatment times. There are also methodological concerns about small sample size and difficulties in maintaining blinding.2 We therefore aimed to conduct an adequately powered trial testing the hypothesis that magnetic bracelets, as used in the consumer market, reduce pain in osteoarthritis of the hip and knee.

Participants and methods

Between December 2001 and December 2003, we recruited 194 participants aged 45-80 years with osteoarthritis of the hip or knee from five rural general practices in Mid Devon (see Osteoarthritis was diagnosed by a consultant (orthopaedic surgeon or rheumatologist) or a general practitioner, and we sought confirmatory radiological evidence for participants who had none recorded in their general practice notes. Participants had to score 8-20 points on the Western Ontario and McMaster Universities osteoarthritis index (WOMAC A) on entry.13 14 We excluded people with a cardiac pacemaker, current magnetic bracelet, surgery to the index joint (excluding arthroscopy), or haemophilia and women who were pregnant or breast feeding.

Recruitment was by referral from doctors, advertising, or invitation after a search of practice records. Trial nurses arranged radiological confirmation of diagnosis if needed, and they collected data in surgery based clinics at 0, 4, and 12 weeks. Participants were given a full strength bracelet at the end of the trial.

Intervention and randomisation

The participants, trial nurse, and healthcare providers, were blinded to treatment allocation. Treatments consisted of identical looking bracelets containing three different components. The manufacturer's specifications were:

Group A?Standard neodymium magnets set in a steel backing cup, with the open side facing the ventral wrist, creating a fluctuating magnetic pattern across the bracelet (fig 1). The field strength at the wrist contact surface was 170-200 mTesla.

Group B?Weak magnets with no backing plate. The field was strong enough to seem magnetic on testing (21-30 mTesla), but previous research suggests this is insufficient to be therapeutic.15 This was intended to provide an undetectable placebo.

Group C?Non-magnetic steel washers.

The National Physical Laboratory tested five bracelets of each type before the study, confirming the manufacturer's specification.

An independent researcher randomised participants in blocks of 15 (five of each bracelet type per block), using random numbers generated in Microsoft Excel. A decode sheet was sealed and locked away. A second researcher checked the procedure. On enrolment, participants were told that they would receive either an active or an inactive bracelet.

Outcome measures

The predefined primary outcome measure was change in WOMAC A score after 12 weeks' follow up.13 14 Secondary outcomes were a visual analogue scale asking, "How bad was the pain from your arthritis in the last week when it was at its worst?" with verbal and numerical anchors from none (0) to worst imaginable (100) 16; WOMAC B and C scores, measuring leg stiffness and functioning13 14; the number of days participants had used analgesics in the past week; and perceived monetary value of the bracelet.

We assessed compliance with wearing the bracelet at 4 and 12 weeks using a visual analogue scale. Blinding was assessed at 12 weeks by asking whether participants thought they had an active bracelet and the reason for such belief.

The estimated effect size was based on a 20% differential reduction in WOMAC A score, which was considered commensurate with effect sizes in studies of analgesics and osteoarthritis.13 17 A sample size of 52 in each of the groups would have 80% power to detect a difference in one way analysis of variance of change scores, assuming mean changes of 3, 1.5, and 1 and a common standard deviation of 3.4. Assuming 15% dropout, we planned to recruit 64 subjects to each group.18 We checked the suitability of these numbers for an analysis of variance across the three groups by using a range of estimated small average changes for the weak magnet group.


The analysis was specified in advance of the study as follows. Last value carried forward was used to impute missing values for subsequent visits. The blinded statistician conducted analysis of variance on all three groups using SPSS version 11.5, with change in WOMAC A score at 12 weeks as the response. The robustness of the results was checked with analysis of covariance on the WOMAC A score at 12 weeks with baseline WOMAC A as covariate, and checking sensitivity to baseline imbalances. Dunnett's test was then used to compare the means for the dummy and weak magnet group separately with the mean for the standard magnet group. The protocol specified that the primary comparison was dummy versus standard magnets, the other comparison being secondary, unless a high degree of unblinding was observed. Models were checked by examination of residuals and sensitivity to imputed values.

Subsequent analyses were unblinded. We used general linear models on all subjects to explore the association between outcomes and magnetic strength of individual bracelets. Similar analyses were then carried out, where appropriate, for WOMAC B and C and the global pain score.


Response rates and sample properties

Of the 391 people assessed for eligibility, 144 did not satisfy the inclusion criteria and 194 (78.5%) of the remaining 247 accepted entry into the trial (see Group baseline characteristics were similar (table 1). Very few participants were lost to follow up. These were evenly spread across the three groups, and their baseline WOMAC A scores were not markedly different from those of participants with complete data. Reported compliance was high, with most wearing the bracelets for 100% of waking hours.

After the trial, we tested all the returned bracelets using a calibrated Hall effect probe. This showed that the standard magnets had a mean strength of 186 (range 134-197) mTesla (only one was outside the specified range) and the non-magnetic group all had zero strength. Because of a manufacturing error, only 28 of the weak magnets were within the specified range (21-30 mTesla). The mean for these 28 magnets was 26 mTesla; 34 magnets had a strength of 69-196 (mean 128) mTesla, and two were not returned (these were assumed to be in the specified range as they were part of a good batch).

Analysis of outcomes

Table 2 shows the scores for the three groups at baseline and after 4 and 12 weeks. Analysis of variance between the three groups on the change in WOMAC A from baseline to 12 weeks showed a difference that was just non-significant (F = 2.90, df = 2, 190; P = 0.057). Results from analysis of covariance on the score at 12 weeks (with baseline WOMAC A score entered as a covariate) were significant (F = 3.24, df = 2, 189; P = 0.041).

The planned comparison (Dunnett's test) showed a significant mean difference in change in WOMAC A score of 1.3 between the standard and dummy magnet groups (95% confidence interval 0.09 to 2.60; P = 0.03), but not between the standard and weak groups (mean difference 0.81, -0.44 to 2.07; P = 0.26). A similar pattern was observed for the change in WOMAC C score. The overall analysis of variance gave significant results (F = 4.45, df = 2, 190; P = 0.013), and Dunnett's test showed a significant mean difference between the standard and dummy groups (4.4, 95% confidence interval 1.0 to 7.9; P = 0.01) but not between the standard and weak groups (3.3, -0.2 to 6.7; P = 0.07). Analysis of the visual analogue pain score showed a significant mean difference between the standard and dummy groups of 11.4 (95% confidence interval 3.0 to 19.8). Change in WOMAC B scores did not differ between groups (F = 0.73, df = 2, 190; P = 0.48). No important differences in these results emerged when either sex or analgesic use (at 12 weeks) was included as a covariate.

Table 3 provides data on participants' beliefs about group allocation and the reasons given for their beliefs. Around a third of participants in the standard and dummy groups were correct in their beliefs about their bracelet, although the reasons differed between groups. In the standard group beliefs were mainly based on noticing the magnetic force?for example, bracelets were often reported to stick to keys in pockets?or on improved symptoms. In the dummy group, few noticed the magnetic force and beliefs were most commonly based on a lack of symptom improvement.

Comparing the outcomes for the different belief groups is not appropriate because belief may follow benefit or lack of it, and any differences would therefore be hard to interpret.19 However, we have a more direct way of estimating the effect of unblinding, as participants reported whether they had noticed the magnetic strength of their bracelets (table 3). The overall pattern of results was replicated in the subgroup of 97 participants (41 (63%) in standard group v 56 (88%) in dummy group) who did not report noticing or testing the magnetic strength of their bracelets at week 12. Results from analysis of covariance estimated the mean difference in WOMAC A between the standard and dummy groups as 1.3 (95% confidence interval 0.003 to 2.62).

To examine the impact of the contamination of the weak magnets on the trial, we analysed data from only the bracelets that met the defined specification (30 weak magnets, 64 dummy magnets, and 64 standard magnets). Analysis of variance showed a significant difference for change in WOMAC A score (F = 3.73, df = 2, 155; P = 0.026). The post-hoc Dunnett's test showed a significant difference between real and dummy magnets (mean difference 1.39, 95% confidence interval 0.11 to 2.68), and a non-significant difference between real and weak magnets, although there was a strong numerical trend (mean difference 1.52, -0.09 to 3.13, P = 0.067).

Table 4 gives data on individual responses to treatments categorised according to predefined criteria for improvement.17 20 Participants' estimate of the monetary worth of the bracelet did not differ significantly. Adverse reactions were rare, with two participants in each group reporting dizziness, increased pain, or stiffness.


We found evidence of a beneficial effect of magnetic wrist bracelets on the pain of osteoarthritis of the hip and knee. Self reported unblinding to treatment group did not substantially affect the results. Although there were problems with the weak magnets, a per-specification analysis suggested (but could not confirm) a specific effect of magnetic bracelets over and above placebo. Other reasons for suspecting a specific effect are that the data on belief show a low level of unblinding in the dummy group and the data on individual responses (table 4) show that more people achieve high levels of improvement in the standard magnet group. The results for two of the secondary outcome measures (WOMAC C and visual analogue pain scores) were consistent with this pattern. No change was seen in WOMAC B score, but this measure has been found to lack sensitivity.18

The findings are consistent with previous studies on magnetic therapies and pain. Studies that have failed to show an effect on pain10 12 generally used weaker magnets (19 to 50 mTesla). Studies that have shown an effect used stronger magnets (47 to 180 mTesla), which were comparable with our standard strength magnets.2 4-8 Together these findings suggest that field strength is important.

Is the effect real?

Our study has not entirely resolved the extent to which the effect of magnetic bracelets is specific or due to placebo. Blinding did not affect the pattern of results, but the validity of the self reporting of blinding status could be questioned. Although the analysis of per-specification bracelets also suggests a specific effect, the result is only a trend and needs confirmation. Therefore, we cannot be certain whether our data show a specific effect of magnets, a placebo effect, or both.

Whatever the mechanism, the benefit from magnetic bracelets seems clinically useful. The mean reduction in WOMAC A scores in the intervention group of 2.9 (27% change from baseline score) and the difference above placebo (1.3 points) is similar to that found in trials of frontline osteoarthritis treatments, including non-steroidal topical creams,21 oral nonsteroidal drugs (including cyclo-oxygenase 2 inhibitors),17 and exercise therapy.22 In a pivotal trial of cyclooxygenase 2 inhibitors17 in osteoarthritic patients with similar baseline pain (mean WOMAC A score 10.7) and the same follow up period (12 weeks), the treatment effects (change above placebo) were 0.8, 1.5, and 1.9 points for the three doses studied. The difference we found in physical function scores (WOMAC C) also compares well with the above trials.

Furthermore, the effects seem additive to those of the participants' usual treatment. The (one off) cost of bracelets (around £30-£50 ($58-$96, 43- 92)), compares well with that of analgesics (paracetamol £20 a year, newer non-steroidal anti-inflammatories £250 a year).23 Larger investigations should now test the safety of magnets relative to the well known risks of analgesics.17 23 24

The low refusal rate favours generalisability of our findings. However, the sample selected was predominantly white with a minimum WOMAC A score of 8. Our results may thus not translate to other ethnic populations or people with milder osteoarthritis. Further work is needed to replicate our findings and determine whether the effect extends beyond 12 weeks. The contamination of group B with stronger magnets prevented a more objective estimation of any-placebo effect. However, our design seems in principle a feasible way to allow for placebo effects in future studies.

A chart showing flow of participants is on

The Mid Devon general practices involved in the study and the research nurses. Mid Devon Primary Care Research Group provided support and advice throughout the study.

Contributors: TH had the original idea, led the writing, participated in planning, and was responsible for the overall management of study. CG and AW participated in planning and design, management of study (via steering group), and writing the manuscript. LB implemented the randomisation procedure and contributed to planning and design and writing the manuscript. AH took part in planning and design, analysis and interpretation, and writing the manuscript. EE participated in planning and design, management of study (via steering group), and writing. Mike Dixon took part in planning, design, and recruitment, and commented on the manuscript. Judith Mathie and Chris Rushton participated in management of study (via steering group), data collection, and coordination of trial nurses. Mark Taylor suggested the use of weak bracelets. All authors approved the final manuscript. EE is guarantor.

Funding: Arthritis Research Campaign.

Competing interests: None declared.

Ethical approval: North and East Devon local research ethics committee and West Somerset local ethics research committee.


1. Ecoflow. Single module magnetic bracelets. (accessed 24 Feb 2004).

2. Weintraub M. Magnetic Bio-stimulation in painful diabetic peripheral neuropathy: a novel intervention-a randomized, double-placebo crossover study. Am J Pain Manage 1999;9: 8-17.

3. Arthritis Research Campaign. Arthritis: the big picture. Chesterfield: ARC, 2002.

4. Hinman R, Ford J, Heyl H. Effects of static magnets on chronic knee pain and physical function: A double-blind study. Altern Ther 2002;8(4): 50-4.[CrossRef]

5. Alfano AP, Taylor AG, Foresman PA, Dunkl PR, McConnell GG, Conaway MR, et al. Static magnetic fields for treatment of fibromyalgia: a randomized controlled trial. J Altern Complement Med 2001;7: 53-4.[ISI][Medline]

6. Man D, Man B, Plosker H. The influence of permanent magnetic field therapy on wound healing in suction lipectomy patients: a double-blind study. Plast Reconstr Surg 1999;104: 2267-8.

7. Brown CS, Ling FW, Wan JY, Pilla AA. Efficacy of static magnetic field therapy in chronic pelvic pain: a double-blind pilot study. Am J Obstet Gynecol 2002;187: 1581-7.[CrossRef][ISI][Medline]

8. Wolsko PM, Eisenberg DM, Simon LS, Davis RB, Wallaczek J, Mayo-Smith M, et al. Double-blind placebo-controlled trial of static magnets for the treatment of osteoarthritis of the knee: results of a pilot study. Altern Ther 2004;10: 36-43.

9. Weintraub MI, Wolfe GI, Barohn RA, Cole SP, Parry GJ, Hayat G, et al. Static magnetic field therapy for symptomatic diabetic neuropathy: a randomised, double blind, placebo-controlled trial. Arch Phys Med Rehabil 2003;5: 736-46.

10. Collacott EA, Zimmerman JT, White PT, Rindone JP. Bipolar permanent magnets for the treatment of chronic low back pain. JAMA 2000;283: 1322-5.[Abstract/Free Full Text]

11. Caselli MA, Clark N, Lazarus S, Velez Z, Venegas L. Evaluation of magnetic foil and PPT insoles in the treatment of heel pain. J Am Podiatr Med Assoc 1997;87: 11-6.[Abstract]

12. Winemiller MH, Billow RG, Laskoswki ER, Harmsen WS. Effect of magnetic v sham-magnetic insoles on plantar heel pain: a randomized controlled trial. JAMA 2003;290: 45-56.

13. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol 1988;15: 1833-40.[ISI][Medline]

14. Roos EM, Klassbo M, Lohmander LS. WOMAC osteoarthritis index. Reliability, validity, and responsiveness in patients with arthroscopically assessed osteoarthritis. Scand J Rheumatol 1999;28: 210-5.[CrossRef][ISI][Medline]

15. Vallbona C, Richards T. Evolution of magnetic therapy from alternative to traditional medicine. Phys Med Rehabil Clin N Am 1999;10: 729-54.[Medline]

16. Price DD, McGrath PA, Rafii A, Buckingham B. The validation of the visual analogue scales as ratio scale measures for chronic and experimental pain. Pain 1983;17: 45-56.[CrossRef][ISI][Medline]

17. Zhao SZ, McMillen JI, Markenson JA, Dedhiya SD, Zhao WW, Osterhaus JT, et al. Evaluation of the functional status aspects of health-related quality of life of patients with osteoarthritis treated with celecoxib. Pharmacother 1999;19: 1269-78.

18. Angst F, Aeschlimann A, Steiner W, Stucki G. Responsiveness of the WOMAC osteoarthritis index as compared with the SF-36 in patients with osteoarthritis of the legs undergoing a comprehensive rehabilitation intervention. Ann Rheum Dis 2001;60: 834-40.[Abstract/Free Full Text]

19. Schulz KF, Grimes DA. Blinding in randomised trials: hiding who got what. Lancet 2002;359: 696-700.[CrossRef][ISI][Medline]

20. Pham T, Van Der Heijde D, Lassere M, Altman RD, Anderson JJ, Bellamy N, et al. Outcome variables for osteoarthritis clinical trials: the OMERACT-OARSI set of responder criteria. J Rheumatol 2003;30: 1648-54.[ISI][Medline]

21. Bookman AAM, Williams KSA, Shainhouse JZ. Effect of a topical diclofenac solution for relieving symptoms of primary osteoarthritis of the knee: a randomized controlled trial. CMAJ 2004;171: 333-8.[Abstract/Free Full Text]

22. 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: 15-9.[Abstract/Free Full Text]

23. British Medical Association, Royal Pharmaceutical Society of Great Britain. British National Formulary. London: BMA, RPS, 2004. (No 47.)

24. Langman MJ. Adverse effects of conventional non-steroidal anti-inflammatory drugs on the upper gastrointestinal tract. Fundam Clin Pharmacol 2003;17: 393-403.[CrossRef][ISI][Medline]