RESEARCH IN PRACTICE

Controls Control groups in a study are important so that the investigator has a baseline to compare the treatment to. In the literature reviewed, controls took two forms: ■ internal controls where all subjects received treatment on one side, the other side of the body being used as the control (1,6)

■ or multiple groups within the study, one being the control group (2,3,4,5,7).

Lightfoot et al (3) actually had a control group as well as using an internal control within the ‘massage’ group. The type of control implemented varied between studies. Hilbert et al (2) used a place- bo lotion applied by the same person implementing the treatment and rest. Smith et al (5) used a similar control, applying moisturis-

as a treatment it is obviously difficult to achieve a totally blinded study. None of the studies reviewed were blinded in any way.

Placebo treatment Another element that is difficult to achieve in soft tissue therapy research is a placebo treatment. A placebo treatment is one that has no effect on the variables being investigated but leads the sub- jects to believe that they have received some form of intervention. This avoids any bias in the subjects reporting on the outcomes.

Lightfoot et al (3) made some efforts to address problems with reporting by creating a third trial group which undertook a stretch- ing intervention. In an attempt to prevent an expectation effect from subjects in their reporting, the participants were told that the study was looking at the effects of stretching on DOMS. Stretching was used because previous studies had shown that pre-exercise stretching will not alleviate or prevent DOMS. Hilbert et al (2) described their control treatment as a placebo but, as they applied a lotion to the control group and as the physiological effects of ‘massage’ are not well understood, it can not be definitively stated that this application did not have some effect on the subjects. This is why a true placebo treatment in soft tissue therapy research would be difficult to achieve.

OUTCOMES This section highlights the inconsistencies of these studies and hence exemplifies the difficulty in drawing conclusions and the need to perform further well-defined studies.

er to the control group and then instructing subjects to rest. Four of the studies (1,3,6,7) used no treatment for the control group. No description of the control was given by Rodenburg et al (4).

Random trials Random trials are important to avoid any bias within or between groups in a study. This involves subjects being randomly assigned to groups often decided by a computer-generated randomisation. Only one of the studies reviewed used such a programme to assign subjects to groups (5). Four of these studies stated simply that subjects were randomly assigned to groups (1,3,4,7) with no description of how the randomisation was done. Furthermore, Rodenberg et al (4) described the group allocation as random even though 27 out of 50 subjects were assigned to the control group - no explanation was given as to why the numbers in control vs treat- ment groups was not even. The randomisation in Tiidus et al (6) was ambiguous, stating that one leg was randomly selected as the treatment leg but not clarifying if this choice was made by the subject or the investigator. Hilbert et al (2) did not state that group allocation in their study was random.

Sample size A reasonable sample size is also necessary to make any results sta- tistically significant. Sample sizes across the papers reviewed here were widely varied, ranging from 8 subjects (1) to 50 subjects (4).

Blind studies Another important component of ‘good research’ is to have, where possible, a blinded study. This means that investigators, subjects and those administering treatment are not aware of which group they are dealing with. However, in the case of soft tissue therapy

www.sportex.net

Within this literature review varying outcomes were found: ■ three papers found ‘massage’ decreased the soreness and/or tenderness associated with DOMS (1,2,5) ■ three papers found ‘massage’ had no effect on alleviating or attenuating soreness associated with DOMS (3,6,7) ■ and one paper (4) found that the treatment administered was effective in decreasing the effects of the eccentrically induced mus- cle injury, but involved multiple interventions, and so no conclu- sions can be drawn from this study on the effectiveness of ‘massage’ as a treatment.

Although four out of the seven papers reviewed found the treatment effective in decreasing the symptoms of DOMS it is difficult to draw any broad conclusions from these results due to a number of incon- sistencies and flaws within them. Primarily, the problem lies in the methodology or, more specifically, in the description of the soft tissue therapy treatment used. In general, there was found to be a lack of definition, adequate direction, explanation and philosophy behind the techniques used. Of the three studies finding ‘massage’ to be beneficial, only one (5) gave a well defined ‘massage’ protocol. Smith et al (5) described the treatment as a ‘sports mas- sage’, clearly outlining the duration of the treatment (30 minutes) and breaking the treatment down into individual techniques/strokes, giving a description of the application and duration of each stroke. The treatment was guided by audio cues, thus giving some consistency between subjects’ treatments. The treatment was administered two hours post-DOMS inducement. This was the most clearly described soft tissue therapy treatment and would be the most reproducible methodology. However, no reason- ing for the techniques or time allocated to each technique used was given. Hilbert et al (2) described the treatment used as ‘muscle

9