Conservative treatment for repetitive strain injury

Conservative treatment for repetitive strain injury. Various conservative treatment options for repetitive strain injury are widely used, despite questionable evidence of their effectiveness. This systematic review evaluates the effectiveness of these treatment options for relieving symptoms of repetitive strain injury and improving activities of daily living. Searches in Medline and Embase, with additional reference checking resulted in 15 eligible trials for this review. Methodological quality was assessed, and data-extraction was performed. With the use of a “best-evidence synthesis”, no strong evidence was found for the effectiveness of any of the treatment options. There is limited evidence that multidisciplinary rehabilitation, ergonomic intervention measures, exercises, and spinal manipulation combined with soft tissue therapy are effective in providing symptom relief or improving activities of daily living. There is conflicting evidence for the effectiveness of behavioral therapy. In conclusion, little is known about the effectiveness of conservative treatment options for repetitive strain injury. To establish strong evidence, more high-quality trials are needed.

In the past two decades, repetitive strain injury (RSI) has become a major topic, both socially and scientifically. The prevalence of musculoskeletal disorders reported in the United States has increased steadily, accounting for more than 65% of all occupational illnesses (1). In 1995, 62% of all cases of work-related illness in the private sector, included in a report from the Bureau of Labor Statistics, were due to disorders associated with repeated trauma, such as typing, repetitive tool use, and repetitive grasping or moving of objects other than tools (2). The prevalence of repetitive strain injury in the entire Dutch labor force has been reported to be approximately 19% (3). Repetitive strain injury is not only a major health problem, but also a huge socioeconomic one. The total compensable cost for upper extremity, work-related musculoskeletal disorders in the United States in 1993 was USD 563 million (2).
Musculoskeletal disorders of the upper limbs and neck in association with repetitive or continuous strain in work situations have been described in different countries under various synonyms, such as occupational cervicobrachial disorder (OCD), upper-extremity cumulative trauma disorder (UECTD), work-related upper-limb disorder (WRULD), and occupational overuse syndrome (OOS) (4)(5)(6).
In The Netherlands, repetitive strain injury is the most frequently used term for these broad ranges of symptoms and disorders. It is a collective term for syndromes characterized by several complaints, such as pain, tingling, numbness, loss of coordination and loss of force, generally affecting the upper extremities and neck and caused by repetitive or continuous strain.
It is important to keep in mind that terms like repetitive strain injury and upper-extremity cumulative trauma disorder are not diagnoses, but merely statements of causation (7). The etiology of repetitive strain injury is usually multifactorial, in many cases the etiology is still obscure (5,6,8,9), and for most cases no specific diagnosis can be made. There are no pathological or radiological characteristics to support the diagnosis, and also there is often no manifestation of neurological signs. In the literature, consistent findings concerning risk factors for repetitive strain injury have been described. Recommendations to prevent these risk factors are logical, but studies on the effectiveness of intervention programs are still awaited. Many patients with repetitive strain injury are treated in primary care. There is a wide variety in the treatment options, ranging from ergonomic advice and workload restrictions to physiotherapy and the prescription of medical aids (10). There is also still little available research on the effectiveness of these types of therapeutic intervention.
This systematic review aims at evaluating the effectiveness of conservative treatment options for repetitive strain injury.

Selection criteria
Types of studies. Both randomized controlled trials (RCT) and (nonrandomized) controlled clinical trials (CCT) on conservative treatment options for repetitive strain injury were included. The studies had to be published as full reports (no abstracts) written in English, German, Dutch, French or Finnish.
Types of participants. Randomized controlled trials and controlled clinical trials were included if they reported on subjects (male or female) with repetitive strain injury. Repetitive strain injury was defined as any work-related disorder of the upper extremity, neck or thoracic region in work-age adults (18 to 65 years of age), due to repetitive work or continuous strain at work. Complaints were characterized by discomfort, impairment, disability, or persistent pain in joints, muscles, tendons, and other soft tissues, with or without physical manifestations (11). All occupational groups were included. Excluded were studies reporting on patients with acute trauma, neoplasm, metastatic diseases, and fractures. Patients with complaints of the upper limbs, neck, and thoracic region that were not work-related were also excluded.
Types of intervention. Randomized controlled trials and controlled clinical trials in which one form of intervention consisted of conservative therapy (ie, "nonsurgical" therapy) were considered for inclusion. All types of conservative intervention that were prescribed or performed in the treatment of repetitive strain injury were included, such as occupational therapy, physiotherapy, behavioral therapy, multidisciplinary treatment, or medication. Ergonomic measures used in the treatment of repetitive strain injury were also included, for example, specially designed office furniture, computer keyboards, and computer mice.
Types of outcome measures. The primary outcome measures were self-reported pain intensity expressed on a visual analogue scale (VAS) or a numerical rating scale (NRS), a global measure of improvement (overall improvement, percentage of patients recovered, subjective improvement of symptoms), return to work (sickness absence, days off work), and generic functional status (SF-36, Nottingham Health Profile, Sickness Impact Profile). Randomized controlled trials and controlled clinical trials that did not include any type of patientspecific outcome measure were excluded.

Search strategy
In an attempt to identify relevant studies for this review, the following literature search strategy was used: (i) randomized controlled trials and controlled clinical trials were identified through a computer-aided search of Medline (from 1966 to April 2000), Embase (from 1988 to April 2000), and the Cochrane Controlled Trials Register (Issue 2, 2000) and (ii) references given in retrieved articles and other relevant publications and reviews were screened.

Study selection
One reviewer (HK) generated the Medline and Embase search strategies and downloaded the relevant data (author, title, keywords and abstract) of all the identified studies into a computer file. Two reviewers (AG and MvT) independently applied the selection criteria to the studies that were retrieved through the literature search. The abstracts were reviewed or, in case of any doubt, a copy of the full article was considered. Consensus was achieved to resolve disagreements concerning the selection and inclusion of studies.

Assessment of methodological quality
The methodological quality of the controlled clinical trials and randomized controlled trials was independently assessed by three reviewers (HK, HdV, NdW). Quality assessment was pilot-tested in two trials from a comparable field of research (neck pain) not included in the review. An adjusted version of the criteria list recommended by the Cochrane Back Review Group (13) was used to assess the methodological quality of the randomized controlled trials and controlled clinical trials (appendix). Modifications were made for repetitive strain injury and nonsurgical treatment options regarding prognostic indicators (c) and outcome measures (j). Items could be scored as positive ("yes"), negative ("no"), or unclear ("don't know"). The list consisted of internal validity criteria (items b 1 , b 2 , c, e, f, g, h, i, l 1 , l 2 , n, p), descriptive or external validity criteria (items a, d 1 , d 2 , j, k, m 1 , m 2 ), and two statistical criteria (items o, q).
If the article did not provide sufficient information on one of the methodological quality criteria and the criterion was scored as unclear, the authors were contacted for additional information. This information was used for a final quality assessment made by two reviewers independently. The final scores of the quality assessment were reached in consensus.

Data extraction
Two reviewers (HK and NdW) independently extracted the data, using a standardized form. Information was collected on participants (age, gender), type of repetitive strain injury (criteria on which diagnosis was made, duration of symptoms, location of disorder), interventions (type, duration, treatment schedule), outcome measures, timing of the posttreatment and follow-up measurements, and results (number of patients, point estimates and measures of variability). Although data on all reported outcomes were extracted, outcome measures on subjective symptoms and activities of daily living, including return to work, were considered primary outcomes.

Data analysis
Clinical homogeneity was evaluated by exploring the differences between the studies with regard to study population, types of treatment, and outcome measures. The studies were considered to be heterogeneous with regard to all three criteria. Therefore, it was decided not to pool data from the studies statistically but, instead, to summarize the results according to a rating system for the strength of the scientific evidence, consisting of the following four levels: strong evidence -provided by generally consistent findings in multiple high-quality randomized clinical trials (randomized controlled trials) (level 1), moderate evidence -provided by generally consistent findings in multiple low-quality randomized controlled trials or multiple controlled clinical trials (level 2), limited evidence -only one randomized controlled trial (either high or low quality) or controlled clinical trial (level 3a), conflicting evidence -inconsistent findings in multiple randomized controlled trials or controlled clinical trials (level 3b), and no evidence -no randomized controlled trials or controlled clinical trials (level 4).
Conclusions with regard to the findings of the studies were based on the statistical significance (P < 0.05) of the outcome measures on symptoms and activities of daily living, as assessed by the reviewers. Both posttreatment and follow-up results were taken into consideration. Findings were considered to be "generally consistent" if at least 75% of the studies showed similar results.
A study was (arbitrarily) considered to be of high quality if at least 50% (6 out of 12) of the internal validity criteria were fulfilled. A sensitivity analysis was performed to explore the results when high quality was defined as fulfilling 5 and 7 or more of the 12 validity criteria.

Study selection
The literature search resulted in the identification of 986 publications. Fifteen studies in 16 publications met the inclusion criteria, 5 of which were identified in Medline. Searching Embase and the Cochrane database resulted in the identification of 2 additional studies in each (a total of 4 studies). Reference checking resulted in 6 additional studies. In summary, a total of 15 identified trials (12 randomized controlled trials and 3 controlled clinical trials) on conservative treatment for repetitive strain injury met the criteria for inclusion in this review. These studies evaluated the efficacy of physiotherapy (14)(15)(16)(17), exercise (18,19), behavioral therapies (20)(21)(22)(23)(24), chiropractic treatment (25), a multidisciplinary rehabilitation program (26), an energized splint (27), and an ergonomic approach (28,29).

Methodological quality
The results of the methodological quality assessment are shown in table 1. Applying a cut-off point of ≥ 50% positive items resulted in only five high-quality trials (15,17,18,28,29). Requests to comment on the quality score attributed to the study and to provide more information were sent to all the authors but one, whose recent address could not be traced. Only 5 out of 14 authors responded (15,(17)(18)(19)28). As a result of their comments and information, we decided to change the assessments of 18 items on the criteria list as follows: 9 from unclear to positive, 5 from unclear to negative, 1 from negative to unclear, and 3 from negative to positive. Table 1 presents the studies for each type of intervention in hierarchical order, according to their internal validity score.
Overall, the methodological quality of the studies was poor, although the scores of the randomized controlled trials were somewhat higher than the scores of the controlled clinical trials. Many studies reported random treatment allocation but failed to describe whether the method of randomization was concealed (b2). Information on co-interventions (f) and blinding of the careprovider (e), patient (h), or outcome assessor (i) was often lacking. Other prevalent shortcomings were that no intention-to-treat analyses (p) were performed and that very few studies stated whether compliance was acceptable (g). Table 2 presents the study characteristics (study design, study population, interventions, and results). Only the outcome measures that were relevant to patients are reported. The outcomes were subjective symptoms [eg, levels of pain (VAS, Pain Beliefs Questionnaire, West Haven-Yale Multidimensional Pain Inventory), numbness or stiffness], and activities of daily living (eg, functional status of the hand, return to work, and interference of symptoms with daily living). Other outcome measures reported in the studies were strength (14,15,17,27), tender points (15,17,18,19), sensibility (27)(28)(29), and depression, anxiety, coping strategies or personal harmony (16,19,(21)(22)(23)(24), but they were all considered to be secondary outcomes.

Efficacy of conservative treatment options
The differences between the groups are presented using 95% confidence intervals when data on exact    (2) week for a total of 10 treatments (N=12) improvement in pain of 50% (1) and (2) (19) work-related neck-overload on air-machines giving resistance in current pain and pain at worst in shoulder pain recruited the concentric part of the movements); 1 hour/ (1), (2) and (3) 22) RSI in which pain condi-relaxation training, cognitive skills for coping with than in (3) after the treatment; tion was due to repetitive pain, etc); 1 time/week, 1.5 hours for 9 weeks significantly more improvement in work, subjects replied to (N=12) degree of interference of pain in daily media publicity or were (2) individually conducted cognitive-behavior ther-living in (2) than in (3) after the referred to study by their apy (same components as in group therapy); treatment; no differences between the medical practitioners; no 1 time/week, 1.5 hours for 9 weeks (N=13) groups in the 6-month follow-up; 2data on (mean) age (3) (23) history of musculoskeletal taught a range of relaxation techniques; method found between groups for the pain problems in upper relied on subjective awareness of muscular tension measures of WHYMPI, self-monitored limbs, neck or shoulders and relaxation); 8 sessions of 1.5 hours for pain index or Pain Beliefs Questionin association with repe-4-6 weeks (N=11) naire posttreatment or in the 6titive tasks in the work-   overall pain severity after 6 months in (29) method carpal-tunnel syndrome or (3) Microsoft natural keyboard (N=19) (3) than in (4); significantly more wrist or forearm tendonitis (4) placebo (participants' own keyboard was improvement in overall functional as determined by review cleaned) (N=20) status in (3) than in (4) after 6 months of the workers' compensa-Participants were asked to use the assigned tion injury and illness data-keyboards in their workplace for 6 months base; participants were full-time employed and used a computer keyboard 4 hours/day or 20 hours/ week or more; mean age ranged from 40 to 45 years Rempel et al, RCT, no 13 females & 7 males (1) keyboard A (Protouch keyboard, Key Tronic Cor-Significantly more improvement in 1999 (28) control group who reported to the site poration), keys gave greater feeling of looseness hand pain in (1) than in (2)  differences or standard deviations or confidence intervals were available. Otherwise (for most continuous outcomes), the differences between the groups have been presented as significant or nonsignificant (P < 0.05). It is indicated whether the intervention group was compared with a nontreatment group or to another active intervention group.
Physiotherapy. Four studies evaluating the effectiveness of physiotherapy were identified: two high-quality randomized controlled trials (15,17), one low-quality randomized controlled trial (14), and one controlled clinical trial (16). Two studies compared exercise therapy (optionally combined with additional physiotherapeutic interventions) with any other type of physiotherapy [eg, massage, ergonomic instruction (14,16)]. Neither study found significantly more improvement in symptoms in the exercise therapy groups than in the "other physiotherapy" group. There was moderate evidence (level 2) that exercise therapy and other forms of physiotherapeutic intervention are equally effective in providing symptom relief for patients with repetitive strain injury.
Two studies compared group exercises with individually conducted physiotherapeutic exercises (15,17). One found significantly more improvement in the selfassessment of pain status with individual exercise therapy (17), but the other study (15) found no significant differences between the groups. Therefore, the evidence (level 3b) is conflicting as to whether individually conducted exercise therapy is more effective than group exercise therapy.
Exercises. The effect of exercise (gymnastics) was assessed in one high-quality randomized controlled trial (18) and in one low-quality randomized controlled trial (19). The first study reported significantly more improvement in activities of daily living in the exercise versus the control group at the time of the posttreatment measurement. The other study (19) reported significantly more improvement in "pain at present" and in "pain at worst" when all the exercise groups together were compared with the control group receiving stress management. Both studies reported that exercise had a positive effect. One of these studies was of low methodological quality, and the other was of high quality. Therefore the evidence is limited that exercise relieves pain and also that exercise improves the daily functioning of patients with repetitive strain injury (level 3a).
One study (19) also reported on comparisons between exercise groups and found significantly more improvement in activities of daily living (both during employed work and during household duties) in the strength exercise group, as compared with the endurance group (limited evidence, level 3a).
Behavioral therapy. Four studies that implemented behavioral therapy were identified: three low-quality randomized controlled trials (20-23) and one controlled clinical trial (24).
Cognitive-behavioral treatment was one of the therapies tried. It is based on a multidimensional model of pain that includes physical, affective, cognitive, and behavioral components (30).
One study (21,22) assessed the effect of cognitivebehavioral group therapy and individually conducted cognitive-behavioral therapy versus a waiting-list control group. No statistically significant posttreatment or follow-up differences were found for any outcome measure between the cognitive-behavioral therapy groups and the waiting-list controls. In addition, no significant differences were found between the cognitivebehavioral therapy groups and the individually conducted intervention group.
At posttreatment measurement, significantly more improvement in pain was reported in the individually conducted intervention group than in the waiting-list control group. However, of the two outcome measures of pain (McGill and pain-rating scales) reported in this study, only one (McGill Pain Questionnaire) showed statistically significant differences between the groups. For activities of daily living, this study also reported two outcome measures, interference in daily living caused by pain and the Sickness Impact Profile. Again, for only one measure (interference in daily living) was significantly more improvement reported for the individually conducted intervention group than for the waiting-list control group. No statistically significant differences were found in the 6-month follow-up measurements. Another study (24) evaluated an occupational overuse intervention program. The program combined cognitivebehavioral intervention strategies with specific movement-oriented relaxation training, muscle and task rotation techniques, and training in efficient movement. However, no differences in activities were found between the groups after the treatment or at the time of the 6-month follow-up. Therefore, the evidence (level 3b) is conflicting for the effectiveness of cognitive-behavioral therapy for repetitive strain injury.
Respondent treatment is intended to modify the physiological response system directly, for example, by reducing muscle tension, and it includes providing the patient with a model of the relationship between tension and pain and teaching the patient to replace muscular tension by a tension-incompatible reaction. Electromyographic (EMG) biofeedback, progressive relaxation, and applied relaxation are frequently used (31). One study (23) examined the effectiveness of EMG biofeedback, applied relaxation training, and a combined procedure versus a waiting-list control condition. No statistically significant differences between one of the treatment groups and the waiting-list control group were found after the treatment or at the time of the 6-month follow-up with regard to pain (West Haven-Yale Multidimensional Pain Inventory, Pain Beliefs Questionnaire, self-monitored pain index) or activities of daily living (interference in daily living caused by pain). Another study compared a group receiving hypnotically induced vasodilatation treatment with a waiting-list control group (20) and reported significantly more reduction in pain in the intervention group after 6 weeks of treatment.
There is conflicting evidence (level 3b) that respondent treatment is more effective than a waiting-list control condition in relieving the pain symptoms of patients with repetitive strain injury. The evidence (level 3a) is limited that EMG biofeedback, applied relaxation training, a combined procedure, and a waiting-list control are equally effective in improving activities of daily living among patients with repetitive strain injury.
Energized wrist splint. In one low-quality randomized controlled trial the effect of an energized wrist splint was evaluated (27). Differences in outcomes between the intervention group and the placebo group were reported, but not statistically evaluated.
No evidence (level 4) therefore is available indicating that energized wrist splints are effective in treating repetitive strain injury.
Multidisciplinary rehabilitation. One low-quality controlled clinical trial (26) evaluated the effect of a multicomponent rehabilitation program that included physical conditioning, work conditioning, work-related pain and stress management, ergonomic consultation, and vocational counseling. The intervention group had significantly higher return-to-work rates than the "usual care" group. For patients who returned to work, a significantly higher percentage of the treatment group than of the "usual care" group returned to full-time employment (instead of part-time employment).
Therefore, evidence (level 3a) is limited showing that, for patients with repetitive strain injury, a multi-component rehabilitation program is more effective in return to work than "usual care" is.
Chiropractic treatment. One low-quality randomized controlled trial evaluated the effect of spinal manipulative therapy, combined with soft-tissue therapy in comparison with spinal manipulative therapy only (25). Significantly more improvement in the frequency of symptoms and improvement in the stages of repetitive strain injury was found in the combined therapy group, and, therefore, the evidence was limited (level 3a) that spinal manipulative therapy combined with soft tissue therapy is more effective than spinal manipulation therapy only in providing short-term symptom relief for patients with repetitive strain injury.
Ergonomic approach. Two high-quality randomized controlled trials that evaluated different ergonomic intervention strategies were identified (28,29). The first (28) described two keyboards with different force-displacement characteristics. Significantly more improvement in pain was found in the group using keyboard A versus the group using keyboard B. Keyboard A was, among other things, associated with a greater feeling of looseness of the keys when the subjects' fingers rested on the keycaps. In the activities of daily living (hand functional status), no significant differences between the groups were found. This result indicates limited evidence (level 3a) that force-displacement characteristics of keyboard keys can provide symptom relief for patients with repetitive strain injury. The other study (29) evaluated the effects of three alternative geometry keyboards in comparison with a regular keyboard. Only one type of keyboard (Natural Keyboard) showed significantly more improvement than the regular keyboard in both activities of daily living and pain. There is therefore limited evidence (level 3a) that the use of the Natural Keyboard is more effective than a regular keyboard in improving symptoms or activities of daily living.

Sensitivity analysis
Changing the cutoff point of the quality score to ≥ 5 and to ≥ 7 positive internal validity criteria resulted in 5 and 3 high-quality studies, respectively (table 1). However, there were no changes in the strength of the evidence.

Study selection
To enhance the validity of this systematic review, an attempt was made to cover the whole range of conservative treatment options for repetitive strain injury as comprehensively as possible. Due to the fact that there is no universal unambiguous definition of repetitive strain injury, some difficulties were expected in identifying all the studies on work-related disorders of the upper extremities and neck. An attempt was made to overcome this difficulty by using all the keywords used in relevant articles concerning repetitive strain injury and by searching various databases (Medline, Embase, Cochrane Controlled Trials Register). The references of all the identified studies were also screened. This procedure resulted in approximately 1000 references to studies, of which only 15 met the inclusion criteria. Despite the extensive search strategy used to identify all relevant studies on the effectiveness of nonsurgical therapy for repetitive strain injury, some studies may have been missed because they were indexed in other databases or published in nonindexed journals.
Although the search included studies published in English, German, Dutch, French, and Finnish, only studies published in English were identified and included in the current review. This approach may have led to publication bias, since studies with significant results are more likely to be published in English (32).

Methodological quality
According to the criteria applied, the methodological quality of the included studies on nonsurgical therapy for repetitive strain injury appeared to be disappointingly low. That was one of the reasons why the evidence of the efficacy of the various treatments was mostly limited. Only 5 studies were considered to be high-quality trials. Studies with lower methodological quality are more likely to be associated with biased findings. The quality of most studies could have been improved if a more-specific description had been given of compliance (g), avoidance or comparability of co-interventions (f), and the (reporting of) concealment of treatment allocation (b2). Blinding of patients (h), care-providers (e), and outcome assessors (i) also formed criteria that were often not met. Since the blinding of care-providers and patients is often very difficult and may have been impossible in some studies in this review (eg, in trials evaluating the effect of physiotherapy, exercise, or behavioral therapy options), it is even more essential to have blinded outcome assessments in trials evaluating these types of treatment.

Efficacy
To date, 12 randomized controlled trials and 3 nonrandomized clinical trials have been performed to evaluate the efficacy of nonsurgical therapies for repetitive strain injury. Because of the generally low methodological quality of the studies and the heterogeneity of the study populations, outcome measures, and interventions, it was decided not to perform a formal meta-analysis with statistical pooling of the data. Instead, a "best-evidence synthesis" was conducted, taking into account the number, the quality, and the outcome of the studies. No consensus has yet been reached on how to assess the strength of the available evidence, and therefore the levels of evidence used in this review are, to some extent, arbitrary. It was decided to apply this rating system because it has good face validity, it is simple and explicit, and it had already been applied in several reviews on the effectiveness of conservative treatment options for low-back pain (33,34).
In seven studies the intervention groups were compared with nontreatment control groups (18,19), waiting-list control groups (20)(21)(22)(23)(24), or a kind of placebo intervention group (27,29). Seven of the 15 tested interventions showed a significant difference in efficacy on at least one outcome measure in a comparison with a control group. In 8 studies, 15 active treatments were compared (14-17, 25, 28, 29). Four intervention trials showed relative efficacy, compared with that of other interventions. However, it remains questionable whether the other interventions were effective indeed.
When the strength of the available evidence for the efficacy of the various nonsurgical treatment options in providing symptom relief and improvement in activities of daily living is summarized, there is no strong evidence for the efficacy of any of the forms of intervention used. This result seems disappointing since many of the interventions are commonly used in daily practice. However, the conclusion that there is no strong evidence of effectiveness does not imply that there is strong evidence that there is no effect. There is definitely a need for higher quality trials with sufficiently large sample sizes that meet the internal validity criteria outlined in this review. Furthermore, future trials should evaluate the effectiveness of interventions for repetitive strain injury using valid, reliable and responsive outcome measures. In this review, it was striking that many studies included a substantial number of different outcome measures. Conceivably, this approach would increase the chance that one of the outcome measures would have a positive outcome by chance. As a consequence, authors should consider including fewer outcome measures. Furthermore, the hierarchy of the outcome measures should be carefully (definition of primary outcomes) considered, and a way must be found to combine multiple end points before data are analyzed (35).
Many studies reported on within-group comparisons. To exclude improvement due to placebo effects or to the natural course of events, it is essential to perform and report between-group comparisons, preferably comparisons between an experimental and a control group.

Presentation of statistical analyses
In some studies the outcomes were not presented consistently in the results section. There was a tendency towards the nonpresentation of insignificant results. Another prevalent shortcoming was that many studies did not present summary data and statistics in sufficient detail. The CONSORT Statement might be of great value in improving the quality of reports on randomized controlled trials (36).

Concluding remarks
In conclusion, little is known about the effectiveness of conservative treatment options for repetitive strain injury. As most studies were of poor methodological quality and did not meet the current standards for the conducting and reporting of randomized controlled trials (36), there certainly is an urgent need for high-quality trials. The development and application of a "gold standard" for the clinical diagnosis of repetitive strain injury could provide more comparable and generalizable results.