3.2 Characteristics of the studies reported on PEs
Four trials out of the seven included studies were secondary analyses of RCTs. The number of participants ranged between 40 (7) and 477 (24). Participants’ ages did not vary considerably among the included studies: 18-60 years of age in two trials (25, 26), 20-70 years in two trials (5, 24), 18-64 years (27), 18-65 years (6) and 21-65 years of age (7). Interventions varied significantly between the included trials with some having more than one type of the same intervention. All of the included seven trials addressed the outcome expectations. For measurement of PEs, five trials used different types of scales with different numbers of data points. Four trials used an eleven-point scale (0 to 10) (5, 7, 24, 27), while one trial used a two-point scale (25). The remaining two (6, 26) used questionnaires to measure PEs. In addition, all trials assessed PEs at the baseline, while some of them provided a specific time point (5, 6, 7, and 24), however, others did not (25-27).
Regarding the outcomes of interest namely pain, functional level and HRQOL, four trials addressed only one of the outcomes of interest (5-7, 25), two trials (6, 26) addressed two and only one trial addressed three of the outcomes of interest (27). The outcome of levels of function (disability) was addressed in six trials, pain intensity in four trials and HRQOL in only one trial. In addition, a variety of measurement tools were used to measure these outcomes. The Roland Disability Questionnaire (RMDQ) was used in four trials to measure disability and levels of function. The ODI was used in one trial to measure disability (25). Additionally, the Low Back Rating Scale was used in one trial (26) to measure both disability and pain intensity. Numerical rating scales were used in two trials to assess pain intensity, while the visual analogue scale (VAS) was used in only one trial. The 36-Item Short Form Survey (SF-36) was used only by one trial to assess the outcome of HRQOL (27).
Finally, the follow up points of outcomes varied between short (≤ 6 weeks), medium (> 6 weeks - ≤ 6 months) and long (> 6 months) with some studies having two follow-up points (6, 24, 25). One trial assessed the outcomes immediately after the treatment (7). One trial had a follow-up point after an 8 week period (24), two were at 10 weeks (5, 6), one at 12 weeks (27) and one at 14 weeks (26). The longest follow-up point was 52 weeks (24) and as well as 1 and 5 years (25). Table 1 provides the characteristics of these included trials.