RELATIONSHIP OF REDUCED HAMSTRING FLEXIBILITY WITH LUMBOPELVIC RHYTHM AND MUSCLE ACTIVATION

This cross-sectional biomechanical study investigates how reduced hamstring flexibility alters sagittal lumbopelvic rhythm and trunk-hip muscle activation during forward bending. All procedures are completed in a single laboratory visit and do not involve any clinical intervention or longitudinal follow-up. Study setting and oversight Testing is conducted in a university motion-analysis laboratory equipped with force-isolated flooring, adjustable height fixtures, and electromagnetic shielding. The protocol was approved by the Institutional Review Board of Sahmyook University (IRB #SYU-2025-03-033-002). Written informed consent is obtained before any study-related activity, and the identity of participants is coded to protect confidentiality. Participant allocation and sample size A priori power analysis (one-way ANOVA, effect size f = 0.55, α = 0.05, power = 0.80) indicated that at least 36 participants were required; 38 healthy adults were enrolled to offset potential attrition, and 37 completed the protocol. Participants are not assigned to treatment arms but are retrospectively stratified into three flexibility categories based on their dominant-limb straight-leg-raise (SLR) angle: (1) ≤ 60° (short group), (2) 61-79° (moderate group), (3) ≥ 80° (long group). Eligibility criteria-including absence of low-back pain in the preceding six months-are recorded in a separate module. Instrumentation and calibration * Inertial-measurement units (IMUs; Xsens MTw Awinda, 60 Hz) are affixed to T12, S2, and the posterior thigh with double-sided tape and elastic wraps after a two-point static calibration that aligns sensor axes with anatomical reference frames. * Surface electromyography (EMG; Delsys Trigno, 1 000 Hz) records bilateral erector spinae at L3, multifidus at L5, biceps femoris long head, and gluteus maximus. Skin is shaved, abraded, and cleaned with 70 % alcohol; electrode pairs are placed parallel to muscle fibers following SENIAM guidelines. * Maximal voluntary contractions (MVCs) for each muscle are collected for normalisation. All raw data are synchronised via a hardware trigger and stored in a de-identified repository. Experimental tasks After warm-up, participants perform two tasks in random order, three trials each, barefoot and with knees extended: 1. a full-range forward bend from upright to maximum comfortable flexion and return, and 2. a partial bend from upright to the point where fingertips touch the patella and return. A metronome (80 beats · min-¹) standardises angular velocity. Verbal instructions avoid coaching specific pelvic or lumbar strategies. Primary technical outcomes Sagittal lumbar and pelvic angles are derived from IMU quaternions and filtered with a fourth-order Butterworth filter (6 Hz). The Relative Contribution Index (RCI) is calculated as the percentage of total trunk excursion attributable to the lumbar spine, sampled at 10 % increments of the flexion and extension phases. Higher RCI indicates greater lumbar-dominant motion. Secondary outcomes EMG envelopes are obtained via 20-450 Hz band-pass filtering, full-wave rectification, and 50 ms moving RMS smoothing, then normalised to MVC. Variables include phase-specific RMS amplitude, median-frequency shift (fatigue index), and onset latency relative to movement initiation (5° pelvic tilt threshold). Quality assurance and data integrity A detailed standard-operating-procedure manual governs sensor placement, calibration, and data extraction. Two investigators independently inspect raw signals for artefacts such as sensor dropout, cable motion (EMG), or excessive noise (\> 2 SD from baseline). Automated scripts perform range and consistency checks (e.g., lumbar angle cannot exceed 90°; EMG RMS must be below 100 % MVC in unloaded tasks). Raw and processed files are archived on an encrypted server with automated daily backup. Source-data verification is performed on 15 % of records by an auditor who cross-checks digital files against paper consent and data-capture sheets. A project-level data dictionary (CSV format) defines every variable, its unit, permissible range, and coding scheme. Version-controlled MATLAB and Python scripts are maintained in a private Git repository, and changes trigger a checksum comparison to detect unintended alterations. Plan for missing or unusable data Trials with sensor artefact, loss of synchrony, or premature task termination are repeated once immediately; if unusable, they are marked "missing." If \< 10 % of RCI values are missing, single-imputation using predictive mean matching is applied; otherwise, the participant is excluded from that analysis. EMG values with motion artefact are processed by wavelet-based denoising; segments that remain unusable are set to missing and handled as above. Statistical analysis The primary hypothesis (hamstring flexibility influences RCI) is tested with one-way ANOVA, followed by Bonferroni-corrected pairwise comparisons. Phase-specific differences are explored with mixed-effects models (fixed = group × phase, random = subject). Pearson or Spearman correlations (based on Shapiro-Wilk normality) quantify relations between SLR angle and continuous RCI or EMG variables. Significance is set at p \< 0.05; effect sizes are reported as partial η² or r. Statistical analyses are performed in R (v4.3.2) with the "lme4" and "effectsize" packages. Safety monitoring and adverse-event reporting The biomechanical tasks are low-risk; however, any adverse sensation (e.g., back discomfort) prompts immediate cessation and optional referral to on-campus physical-therapy services. Adverse events are documented in the electronic case-report form and reviewed weekly by the principal investigator; serious adverse events would be reported to the IRB within 24 hours, although none are anticipated. Patient-registry considerations This investigation is not a patient registry; therefore, external certification, on-site audits, and participant-year calculations do not apply. Should future registry expansion occur, the quality-assurance framework described above provides the foundation for broader surveillance.