# Prognostic factors associated with improvement in patients with an episode of non-specific low back pain without radicular syndrome: a prospective observational exploratory study

## TL;DR

This study explores factors linked to improvement in patients with non-specific low back pain after chiropractic treatment, aiming to help personalize care.

## Contribution

The study identifies less-studied prognostic factors, such as treatment expectations and clinician prognosis, for short-term improvement in non-specific low back pain.

## Key findings

- Shorter episode duration and lower disability scores were linked to improvement at 7 days.
- At 4 weeks, clinician's prognosis and being a new patient were significant predictors of improvement.
- Negative treatment expectations and perceived stiffness showed significant associations with improvement.

## Abstract

Low back pain is a leading cause of disability worldwide, with most cases classified as non-specific(NSLBP). While manual therapy appears effective for treating NSLBP, further research is needed to identify candidate baseline factors associated with improvement to help tailor personalized treatment strategies. This prospective observational exploratory study, therefore, aims to identify candidate prognostic factors collected at baseline that are associated with short-term improvement in people with NSLBP.

This study was conducted in chiropractic clinics across France between March 1, 2022, and February 28, 2023. Adults with a new episode of NSLBP were included. Baseline data, including individual, clinical, and therapist-related candidate factors, were collected before and during the initial consultation. Participants were considered improved if they: (i) reported “all better” or “better” on perceived global change, (ii) achieved a 20-point improvement on both Visual Analog Scales (VAS for intensity and unpleasantness) or scored 0 on reassessment, and (iii) showed a 30% improvement on the Oswestry Disability Index (ODI) at 7 days and 4 weeks post-consultation. Missing data were handled using multiple imputation with chained equations (MICE). Logistic regression analyses (univariate and multivariable with spline terms when superior fit was demonstrated) identified candidate prognostic factors associated with clinical improvement.

Out of 1,394 patients contacted, 241 met the inclusion criteria, and 207 completed at least one follow-up assessment. After imputation and multivariable analysis, duration of episode (spline 1: 0.94[0.89-1.00]), Number of painful sites (0.75[0.62–0.92]), negative treatment expectations (0.48 [0.25–0.94]), disability score (spline 1: 0.94[0.89-1.00], spline 2: 0.77[0.62–0.96]), and pain intensity (1.05 [1.02–1.07]) were associated with improvement at 7 days. At 4 weeks, disability score (spline 1: 1.24[1.07–1.45], spline 2: 0.77[0.63–0.95]), pain intensity (1.02 [1.00–1.04]), episode duration (spline 1: 0.95[0.91-1.00]), new patient (0.50 [0.28–0.91]), and clinican’s prognosis (3.89 [1.49–10.10]) were associated with improvement.

Less-studied factors, such as negative treatment expectations, clinician’s prognosis, number of therapists, and perceived stiffness, highlighted significant associations with improvement in this exploratory phase. These findings suggest that incorporating these factors may be used when updating existing models.

The online version contains supplementary material available at 10.1186/s12998-025-00580-5.