Preparatory Lifestyle Program Before Functional Restoration for Chronic Low Back Pain

Background and rationale

Chronic low back pain is a leading cause of disability worldwide. Intensive multidisciplinary Functional Restoration Programs (FRP), introduced in France in the 1990s, deliver comprehensive physical, psychological, and cognitive-behavioral care over a short period and have demonstrated effectiveness on physical capacity, quality of life, and return to work. As these programs have opened to broader patient populations, marked disparities in therapeutic response have emerged. Several factors may explain heterogeneity in outcomes: differences in baseline deconditioning, severe kinesiophobia, professional and family constraints limiting adherence after discharge, and the difficulty of transferring acquired skills to daily life. The waiting period before program entry is often unstructured, with no specific intervention provided, potentially allowing sedentary behavior and physical deconditioning to progress.

A preparatory phase grounded in adapted physical activity, hereafter referred to as PFRP, was developed to address these gaps. The PFRP follows a lifestyle-based approach centered on patient autonomy and self-directed practice, aiming to (1) prepare deconditioned patients physically and psychologically for the intensive program, and (2) develop self-management skills earlier in the care pathway to facilitate transfer of acquired behaviors after program completion. To our knowledge, no previous study has examined a preparatory phase prior to a conservative functional restoration program; prehabilitation research has focused primarily on pre-surgical contexts.

Study design

This was a prospective, controlled, non-randomized, single-center study conducted in a rehabilitation center in northern France. Participants were assigned to one of two groups according to admission scheduling and organizational constraints of the residential program (groups of 2 to 6 patients). Randomization was not feasible due to the longitudinal nature of the study and admission logistics. Identical inclusion and exclusion criteria were applied to both groups to limit selection bias.

Interventions

PFRP group: 12-week preparatory phase delivered in group format at the Day Hospital, totaling 26 hours of therapeutic exposure (2-hour initial assessment, weekly 2-hour supervised sessions until FRP entry, 4 hours dedicated to exercise learning, 2 hours of therapeutic education). Approximately 18 hours of autonomous practice at home were prescribed (69% of total exposure). Supervised sessions, led by an Adapted Physical Activity Instructor, included resistance training at 60% of 3-repetition maximum, mobilization and stretching, and cardiorespiratory training on an ergocycle. Sessions were scheduled at the end of the day to accommodate working participants. Therapeutic education addressed misconceptions about pain and self-management strategies. After the 12-week phase, PFRP participants entered the 4-week intensive FRP.

FRP group: 4-week intensive program in a group setting at the Day Hospital, totaling 117 hours (34 hours of independent training, 76.5 hours of supervised group sessions including physiotherapy, occupational therapy, adapted physical activities, balneotherapy, and stretching). Four weekly therapeutic education sessions addressed pain perception, post-program physical activity planning, and exercise physiology. The FRP group received no structured intervention during the waiting period equivalent to the PFRP phase.

Assessments

Outcomes were assessed at four time points: T0 (start of preparatory phase or waiting period), T1 (FRP entry), T2 (FRP exit), and T3 (4-month follow-up).

Co-primary outcomes were selected a priori based on the theoretical framework of the intervention: (1) sedentary behavior measured by the ONAPS-SED subscale, as a direct measure of the behavioral change targeted by the intervention, and (2) impact of pain on Daily Activities measured by the DALLAS-DA subscale, reflecting functional autonomy in everyday life. The ONAPS questionnaire was not administered at T2 because intensive hospitalization does not reflect habitual physical activity behavior.

Secondary outcomes were organized by hypothesized mechanism: other DALLAS subscales (Work/Leisure Activities, Anxiety/Depression, Social interest), other ONAPS subscales (total physical activity, Moderate-to-Vigorous Physical Activity Intensity), functional disability (EIFEL), cognitive mediators (fear-avoidance beliefs about physical activity and work, FABQ-PA and FABQ-W), physical fitness (trunk muscular endurance via Sorensen, Ito-Shirado, lateral plank, and Killy tests; trunk and hamstring mobility; isokinetic trunk flexor/extensor strength at 30°/s and 90°/s), pain intensity (Visual Analog Scale), and psychological well-being (GHQ-12).

Statistical analyses

Sample size was estimated a priori using G*Power for a repeated-measures ANOVA within-between interaction (2 groups, 4 timepoints, medium effect size f=0.25, α=0.05, power=0.70), yielding a minimum of 22 participants; 24 were enrolled to account for attrition. Given the small sample size and non-normal distributions, non-parametric tests were used: Wilcoxon signed-rank for intra-group comparisons, Mann-Whitney U for inter-group comparisons and Group × Time interactions on change scores. Effect sizes were reported as rank-biserial correlations. The Benjamini-Hochberg procedure was applied to control the false discovery rate. Missing data (2.4%) were imputed using last observation carried forward following intention-to-treat principles.

Registration note

This study was retrospectively registered on ClinicalTrials.gov. Registration was not performed prior to enrolment because the study was classified as non-interventional research under French legislation (Loi Jardé) and received ethics approval under that framework. The requirement for prospective registration under ICMJE criteria, which apply a broader definition of clinical trial including non-randomised controlled studies with prospective allocation, was identified only at the manuscript preparation stage. The outcomes registered correspond exactly to those defined a priori in the study protocol and reported in the resulting manuscript.