Side effects 1 trial reported temporary worsening of low back pain in 17% of the traction group and 15% of the exercise group. In this review we followed the guidelines of the Cochrane Back Review Group (Furlan 2009) and the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We used the Cochrane Back Review Group's 'Risk of bias' tool to assess the risk of bias of the included RCTs (Furlan 2009).
Studies included in the previous version of the review were not assessed using this tool. We defined a study at low risk of bias as one that met six or more criteria and had no fatal flaws. Further research is unlikely to change either our assessment or our confidence in the results.
Twenty-three of the studies included a relatively homogeneous population of people with LBP and sciatica (Bihaug 1978; Coxhead 1981; Fritz 2007; The results of risk of bias analysis for individual studies are summarized in Figure 1. Most of the included studies did not properly report their random and concealed treatment assignment.
In three studies, loss to follow-up exceeded 20% of the study population (Coxhead 1981; Harte 2007), or significantly more subjects were lost to follow-up in one treatment group compared to the number of subjects lost to follow-up in the other group (Gudavalli 2006). In nine of the included trials it was not clear how many subjects were lost to follow-up (Larsson 1980; High risk. Although total loss to follow-up was only 16.6%, significantly more subjects in the active trunk exercise program group dropped out of the study (T) 13, C) 25).
Low risk There was no significant difference between the groups on any baseline characteristic. High risk Outcome measures were assessed by a provider of care and a blinded outcome assessor. High risk Care providers were not blinded, but it is unlikely that they were.
Low risk Participants were unaware of the amount of withdrawal and were therefore blinded. Low risk Outcome assessors were unaware of the amount of withdrawal and were therefore blinded. Unclear risk It is not clear whether co-interventions were allowed during the treatment period or whether they were part of the treatment protocol.
Low risk. The demographics of the included participants were similar for both groups in terms of age, sex, type of work, sick leave, weight, height, and previous back pain treatment.
Comparison 1 Low-back pain with/without radiation, traction versus placebo, sham or no treatment, Outcome 1
Low back pain with/without irradiation, two types of traction Outcome or title of subgroup no.
Comparison 4 Low-back pain with/without radiation, traction versus other treatment, Outcome 1 Pain intensity
Comparison 4 Low-back pain with/without radiation, traction versus other treatment, Outcome 2 Functional
Low back pain with radiation, traction versus placebo, sham or no treatment Outcome or subgroup ten-.
Comparison 7 Low-back pain with radiation, traction versus other treatment, Outcome 2 Functional status
Comparison 7 Low-back pain with radiation, traction versus other treatment, Outcome 3 Global improvement
Comparison 8 Low-back pain with radiation, two types of traction, Outcome 1 Pain intensity
Comparison 8 Low-back pain with radiation, two types of traction, Outcome 2 Global improvement
Comparison 9 Low-back pain without radiation, traction versus sham, Outcome 1 Pain intensity
There is a low risk of selection bias if the researchers describe a random component in the sequence generation process, such as: referencing a table of random numbers, using a computer random number generator, tossing coins, shuffling cards or envelopes, rolling dice , to draw. There is a low risk of selection bias if the participants and investigators enrolling participants could not foresee allocation because one of the following, or an equivalent method, was used to hide allocation: central allocation (including telephone, web-based, and pharmacy-directed randomization); sequentially numbered drug containers of identical appearance; or sequentially numbered, opaque, sealed envelopes. There is a high risk of bias if participants or researchers enrolling participants may be able to provide assignments and thus introduce selection bias, such as assignment based on: using an open random assignment scheme (for example, a list of random numbers); job envelopes were used without appropriate safeguards (e.g. if envelopes were unsealed or non-transparent or not consecutively numbered);
There is a low risk of performance bias if blinding of participants was ensured and it was unlikely that the blinding could have been broken;. There is a low risk of performance bias if blinding of staff was ensured and it was unlikely that the blinding could have been broken;. There is a low risk of attrition bias if there were no missing outcome data; reasons for missing outcome data would likely be unrelated to the true outcome (for survival data, censoring would be unlikely to bias); missing outcome data were balanced in numbers, with similar reasons for missing data across groups; for dichotomous outcome data, the proportion of missing outcomes compared to the observed event risk was not enough to have a clinically relevant impact on the intervention effect estimate; for continuous outcome data, the plausible effect size (difference in means or standardized difference in means) between missing outcomes was not sufficient to have a clinically relevant impact on observed effect size, or missing data were imputed using appropriate methods (if dropouts were very large is , imputation with even 'acceptable' methods may still present a high risk of bias) (Van Tulder 2003).
There is a low risk of reporting bias if the study protocol is available and all prespecified study outcomes (primary and secondary) of interest in the review are reported in the prespecified manner, or if the study protocol is not available , but he is clear that published reports include all expected results, including those that were predetermined (persuasive text of this nature may be unusual). There is a high risk of reporting bias if all prespecified primary study outcomes are not reported; one or more primary outcomes are reported using measurements, analysis methods, or subsets of data (eg, subscales) that were not prespecified; one or more reported primary outcomes were not prespecified (unless a clear justification was given for their reporting, such as an unexpected adverse effect); one or more outcomes of interest in the review are reported incompletely so that they cannot be included in a meta-analysis;. There is a low risk of bias if the groups are similar on the basis of demographic factors, value of key outcome measures and important prognostic factors (examples in the field of back and neck pain are duration and severity of complaints, occupational status, percentage of participants with neurological symptoms) (Van Tulder 2003).
There is a low risk of bias if there were no co-interventions or they were similar between the index and control groups (Van Tulder 2003). There is a low risk of bias if adherence to the interventions was acceptable, based on the reported intensity/dose, duration, number and frequency of both index and control intervention(s). There is a low risk of bias if all randomized participants were reported/analysed in the group to which they were allocated at randomisation.
There is low risk of bias if all important outcome measures for all intervention groups are measured at the same time (Van Tulder 2003). There is a low risk of bias if the study appears to be free from other sources of bias not addressed elsewhere (eg study funding). The review was conducted using the latest methods regarding risk of bias assessment and reporting, as stated in the Handbook.
