This pilot study utilized a randomized, cross-over design. Subjects 18 years of age and older were recruited from the Family Medicine Clinic of the Texas College of Osteopathic Medicine in Fort Worth, TX during February through May 2005. Subjects were asked to participate in the study while waiting for their regularly scheduled physician appointment. They were approached if their participation would not interfere with clinic operations.
Participants were asked to watch a video with 2 minute demonstrations representing a High-Velocity Low-Amplitude (HVLA), placebo light touch (LT), and placebo sub-therapeutic ultrasound (ULTRA) treatment for low back pain. The HVLA treatment consisted of lateral recumbent lumbar soft tissue treatment followed by a 'lumbar roll' manipulation applied to both the left and right side of the simulated patient. The LT treatment consisted of a series of static bilateral hand placements on the posterior thoraco lumbar junction and upper pelvis followed by anterior lower ribs and pelvis of the simulated patient. The ULTRA treatment consisted of a close up image of the ultrasound device followed by sequential application of the non-functional ultrasound applicator (without gel) in circular motions with mild pressure to the same anterior and posterior bilateral body areas on the simulated patient to which the light touch was applied. All treatments were videotaped on the same day using the same professionally dressed osteopathic physician and same casually dressed simulated female patient. All treatments were applied through the clothing. Treatments were applied for 2 minutes with 1 minute allocated for each side. All treatments required the simulated patient to lie down, change sides, and return to a sitting position. The simulated patient began and ended each treatment in the sitting position facing the camera with the physician standing behind her.
The order in which subjects viewed the demonstrations was randomized to control for order effect bias. Six video tapes were available to account for the different combinations of treatments. The combinations were HVLA, LT, ULTRA; HVLA, ULTRA, LT; ULTRA, HVLA, LT; ULTRA, LT, HVLA; LT, HVLA, ULTRA; and LT, ULTRA, HVLA. A verbal description of each treatment narrated by the same person was included on the video, and a written description was included on each survey. Subjects indicated the extent of their agreement (using a 4 point Likert scale) with 4 statements that were presented after each treatment was viewed: 1. "I believe this treatment would allow me to get better quicker"; 2. "I believe this treatment would decrease my low back pain"; 3. "I believe this treatment would make me more able to do the things I want to do"; 4. "This seems like a logical way to treat low back pain" (Table 1). Demographic information and previous experience with osteopathic manipulative treatment, chiropractic treatment, massage therapy, and ultrasound were collected; however, a distinction was not made between previous experience with diagnostic and therapeutic ultrasound.
Repeated measures analysis of variance (ANOVA) was performed on each of the 4 questions to determine differences in responses for treatments. Post hoc analyses were performed using Least Significant Difference (LSD). LSD was used because of the pilot nature of the research study. It does not adjust for the number of comparisons in the post hoc analyses. A partial Eta squared (η2p) was calculated for each of the 4 statements, representing the proportion of the total variability accounted for by the test. A Cohen's d (d) was calculated for the different possible combinations of the four treatment presentations. Cohen's d is an effect size calculated as (mean of treatment group 1 – mean of treatment group 2)/the pooled standard deviation. Throughout this manuscript, treatment group 1 and treatment group 2 are ordered as HVLA and ULTRA; ULTRA and LT; or HVLA and LT. Cohen's d values are considered to be a small effect size at 0.2, a moderate effect size at 0.5, and a large effect size at 0.8 .
Participants responding "yes" to the question "Have you ever had osteopathic manipulative treatment" or "Have you ever had chiropractic treatment" were combined since small numbers of respondents prevented meaningful separation for sub group analysis. Independent samples t tests were performed for each question relating to the ULTRA and LT groups to determine differences between those having ever had manipulative treatment (osteopathic or chiropractic) and those not having ever had manipulative treatment.
Results were considered significant at the alpha 0.05 level. All analyses were performed using SPSS 12.0 (SPSS Inc, Chicago, IL). Study procedures were approved by the University of North Texas Health Science Center's Institutional Review Board.