A preplanned effect modifier analysis of the Specific Treatment of Problems of the Spine randomized controlled trial.
To identify characteristics associated with larger or smaller treatment effects in people with low back disorders undergoing either individualized physical therapy
or guideline-based advice.
Summary of Background Data.
Identifying subgroups of people who attain a larger or smaller benefit from particular treatments has been identified as a high research priority for low back disorders.
The trial involved 300 participants with low back pain
and/or referred leg pain (≥6 wk, ≤6 mo duration), who satisfied criteria to be classified into five subgroups (with 228 participants classified into three subgroups relating to disc-related disorders, and 64 classified into the zygapophyseal joint dysfunction subgroup). Participants were randomly allocated to receive either two sessions of guideline based advice (n = 144), or 10 sessions of individualized physical therapy
targeting pathoanatomical, psychosocial, and neurophysiological factors (n = 156). Univariate and multivariate linear mixed models determined the interaction between treatment group and potential effect modifiers (defined a priori) for the primary outcomes of back pain, leg pain (0–10 Numeric Rating Scale) and activity limitation (Oswestry Disability Index) over a 52-week follow-up.
Participants with higher levels of back pain, higher Örebro scores (indicative of higher risk of persistent pain) or longer duration of symptoms derived the largest benefits from individualized physical therapy
relative to advice. Poorer coping also predicted larger benefits from individualized physical therapy
in the univariate analysis.
These findings suggest that people with low back disorders could be preferentially targeted for individualized physical therapy
rather than advice if they have higher back pain levels, longer duration of symptoms, or higher Örebro scores.
Level of Evidence: 2