• November 3, 2026 |
  • |

Fatigue-Sensitive Wearable Load Budget and Work-Conditioning for Heavy-Duty Workers

By:
SHARE
ABSTRACT
Persistent low back pain remains a leading cause of occupational disability, particularly among heavy-duty workers in logistics, warehousing, and construction. Conventional rehabilitation often fails to address the cumulative physical demands and fatigue-related symptom fluctuations experienced in these occupations. This article proposes a two-stage clinical investigation evaluating an individualized, fatigue-sensitive "back-load budget" integrated with physical therapist–led work conditioning. The initial feasibility stage will assess wearable sensor performance, diary adherence, individualized workload calibration, and the practicality and safety of therapist-guided workload adjustments. Following feasibility evaluation, a prospective, two-arm randomized controlled trial will compare conventional work conditioning supplemented by wearable- and diary-informed workload budgeting against conventional work conditioning alone. The proposed intervention combines trunk and pelvis inertial sensors, structured smartphone diaries, and a provisional traffic-light decision system to guide occupational exposure and rehabilitation progression. The primary randomized trial outcome will be the rate of new low-back-pain flare-up episodes during the eight-week intervention, with the total number of days spent in flare-up episodes, flare-up severity, functional capacity, disability, and sustainable work participation evaluated as secondary outcomes. This protocol provides a framework for assessing the feasibility and potential clinical effectiveness of individualized, fatigue-responsive occupational load management.

Introduction

Persistent low back pain imposes a profound occupational burden on workers engaged in physically demanding tasks, particularly within heavy-material handling logistics, warehousing, and construction sectors. These environments are characterized by high-volume repetitive lifting, sustained trunk flexion, heavy carrying, and forceful pushing or pulling maneuvers. Such biomechanical exposures are strongly associated with transient and sustained risks of musculoskeletal injury. Evidence demonstrates that time spent lifting, bending, twisting, and squatting is significantly correlated with transient risks of low back pain flares within 24 hours of the activity1. Furthermore, while leisure-time physical activity often confers protective musculoskeletal benefits, high-demand occupational physical activity—such as lifting heavy weights while inclined or adopting awkward postures—substantially increases the prevalence of recurrent low back pain2.

The relationship between cumulative workload, physical fatigue, insufficient recovery, and symptom exacerbation is central to occupational disability. In heavy-duty occupational settings, cumulative physical demands, fatigue, and insufficient recovery may contribute to fluctuations in low back pain symptoms, although individual tolerance varies and biomechanical exposure alone cannot reliably predict pain flare-ups. Interventions focusing exclusively on generic ergonomic feedback, isolated movement correction, or standard strengthening often lack the dynamic responsiveness required to manage real-world, cumulative exposure. They typically fail to account for the diurnal variations in worker fatigue and symptom recovery. Consequently, personalized occupational workload progression, guided by objective exposure metrics and subjective recovery reports, represents a critical evolution in occupational rehabilitation. The proposed study aims to investigate whether an individualized, fatigue-sensitive lifting-load budget, integrated with physical therapist–led work conditioning, can effectively mitigate next-day pain flare-ups and sustain occupational function in this vulnerable population.

Literature review and research gap

The literature highlights a distinct dichotomy between the therapeutic effects of controlled exercise and the deleterious effects of unmanaged occupational loading. Physical activities involving high or sustained biomechanical demands may be associated with delayed low back pain exacerbations, highlighting the importance of monitoring symptoms after occupational exposure3. To quantify this exposure, wearable inertial sensors have emerged as a viable tool. Combining trunk motion data with under-the-foot force data yields low back disorder risk estimates that strongly correlate with ground truth biomechanical analyses4. Specifically, integrating trunk inertial measurement units (IMUs) with pressure-sensing insoles enables automated estimation of time-series lumbar extension moments during material handling5. Furthermore, inertial motion capture systems combined with inverse dynamics have been investigated for estimating joint loading, although measurement accuracy depends on the modeling assumptions, sensor configuration, and validation conditions.6

Table 1: Limitations of Existing Interventions vs. Proposed Integrated Approach

Intervention Type

Primary Mechanism

Identified Limitations

Proposed Solution

Standalone Wearable Feedback

Real-time posture alerts

Lacks tissue conditioning; poor compliance due to privacy concerns7.

Hybrid tracking linked to progressive physical therapy.

Conventional Work Conditioning

Generic strengthening and task simulation

Fails to account for daily cumulative fatigue and delayed flare-ups.

Daily “back-load budget” adjusting therapy intensity.

Ergonomic Modifications

Workstation redesign, lifting aids

Often impractical in dynamic construction or warehousing environments.

Individualized pacing and microbreaks based on tolerance.

Despite these technological advancements, wearable feedback as a standalone intervention possesses significant limitations. Occupational safety professionals cite employee privacy and confidentiality as primary barriers to widespread adoption7. Moreover, sensors cannot independently diagnose pathology or prescribe rehabilitation. Conversely, physical therapy–led work conditioning demonstrates moderate efficacy. Intense physical conditioning probably reduces sickness absence duration slightly compared to usual care at 12-month follow-up for workers with chronic back pain8. Multi-domain workplace return-to-work interventions that include health-focused and work-modification domains show strong evidence for reducing disability duration9. However, a critical research gap persists: the integration of individualized, fatigue-sensitive load budgets (derived from both objective IMU data and subjective diaries) with therapist-guided work conditioning remains insufficiently investigated. This proposed protocol seeks to bridge this gap by dynamically linking exposure monitoring with clinical rehabilitation.

Research objectives and hypotheses

The primary objective of this proposed clinical investigation is to determine whether an individualized fatigue-sensitive lifting-load budget integrated with physical therapy–led work conditioning reduces the rate of new low-back-pain flare-up episodes during the eight-week intervention compared with a conventional work-conditioning program without personalized cumulative load budgeting. Improvements in occupational function will be evaluated as secondary outcomes.

Secondary objectives include evaluating changes in low back disability, work-task lifting and carrying tolerance, cumulative occupational exposure, perceived fatigue, muscular endurance, fear-avoidance beliefs, exercise adherence, work ability, and sustainable work participation. The central hypothesis is that workers receiving the fatigue-sensitive load budget will experience a lower rate of new low-back-pain flare-up episodes than those receiving conventional work conditioning. Reductions in flare-up severity and duration will be evaluated as secondary outcomes, without assuming that wearable-derived exposure estimates directly measure individual tissue tolerance. A secondary hypothesis posits that this integrated approach will yield superior long-term sustainable work participation and functional capacity, as the avoidance of severe exacerbations will facilitate more consistent adherence to progressive work conditioning.

Methodology and proposed study design

This protocol proposes a two-stage clinical investigation. The initial feasibility stage will assess wearable sensor performance, participant adherence, diary completion, workload-budget calibration, and the practicality of therapist-guided adjustments. Findings from this stage will inform refinements to the monitoring procedures and traffic-light decision rules before proceeding to a prospective, two-arm randomized controlled trial evaluating clinical effectiveness.

The study setting will encompass heavy-material handling logistics, warehousing facilities, and heavy construction environments, ensuring the cohort faces high-volume repetitive lifting and trunk flexion. A retrospective cohort study examining work conditioning among injured workers in physically demanding occupations reported favorable functional-capacity outcomes associated with this rehabilitation approach.10 However, because the study was observational, its findings do not independently establish that work conditioning is superior to standard physical therapy.

Participants will be recruited based on a confirmed history of persistent post-traumatic or work-related low back pain lasting longer than 12 weeks and current employment in physically demanding occupational roles involving regular lifting, carrying, pushing, pulling, or sustained trunk movements. Eligible participants must be actively performing occupational duties, including modified-duty assignments with measurable physical exposure, during the study period. Workers who are completely off work at enrollment will be excluded from the initial feasibility stage and randomized trial because the proposed workload-budget algorithm requires ongoing occupational exposure monitoring. Future research may evaluate a separately adapted protocol for workers transitioning from medical leave to active employment.

Exclusion criteria will include systemic inflammatory conditions, recent spinal surgery (within 6 months), red-flag pathologies (e.g., cauda equina syndrome, active infection), and contraindications to moderate-to-vigorous physical activity. Following the initial feasibility stage, participants in the randomized controlled trial will be allocated in a 1:1 ratio to either the experimental group (fatigue-sensitive load budget plus conventional work conditioning) or the active control group (conventional work conditioning alone), using concealed allocation through a centralized digital randomization system.

Both groups will receive the same planned duration, frequency, and core components of the eight-week work-conditioning program. The experimental group will additionally receive wearable- and diary-informed workload monitoring, individualized traffic-light classifications, and therapist-guided adjustments to occupational exposure and rehabilitation intensity. The control group will receive standard clinical monitoring and symptom-guided safety adjustments without the individualized workload-budget algorithm. Participants in both groups will wear the same monitoring sensors and complete the same daily symptom and recovery diaries to support comparable outcome assessment. Wearable-derived workload feedback and individualized traffic-light recommendations will be available only to the experimental group. Both groups will receive equivalent scheduled therapist contact, with any additional safety-related consultations documented. Assessor blinding will be maintained for objective functional capacity evaluations, although participants and treating therapists cannot be blinded to treatment allocation.

The proposed randomized treatment duration is eight weeks, with assessments at baseline, four weeks, eight weeks (end of intervention), and six months post-intervention. Evidence from gradual return-to-work interventions provides relevant support for evaluating sustainable occupational recovery.11 Before the randomized trial, the initial feasibility stage will assess participant recruitment and retention, wearable sensor data completeness, daily diary adherence, successful workload-budget calibration, therapist compliance with traffic-light decision rules, and intervention-related adverse events. Provisional progression criteria will include at least 80% diary completion, at least 80% usable wearable monitoring data, and successful workload-budget calibration for at least 90% of feasibility participants, with no unresolved intervention-related serious safety concerns. These thresholds will be prespecified and reviewed before progression to the randomized trial.

The randomized trial sample size will be finalized after the feasibility stage using estimates of new low-back-pain flare-up episode incidence, the clinically meaningful treatment effect, count-data dispersion, anticipated attrition, expected diary completeness, and variation in occupational work exposure. The calculation will be based on the prespecified primary endpoint and planned negative binomial regression analysis. The study will require approval from an appropriate research ethics committee before participant recruitment, and the randomized controlled trial will be prospectively registered in a recognized clinical trial registry. The ethics approval and registration identifiers will be reported once available. Written informed consent will be obtained before participation, with approved procedures governing wearable monitoring, diary data, participant safety, and confidentiality.

Occupational workload assessment and monitoring

A hybrid methodology will be employed for occupational workload assessment, utilizing wearable inertial measurement units (IMUs) positioned over the lower trunk and pelvis, together with a structured smartphone app-based diary. The proposed sensor configuration will record three-axis acceleration and angular velocity to estimate trunk flexion, rotation, movement duration, and repetitive movement frequency. Sensor placement, attachment methods, sampling frequency, and calibration procedures will be standardized during the initial feasibility stage to assess measurement consistency during occupational tasks. The proposed processing pipeline will include sensor calibration, synchronization, signal filtering, segmentation of occupational movement periods, and extraction of movement-duration, frequency, and trunk-orientation features. Candidate processing parameters and task-classification performance will be evaluated during the feasibility stage before the workload-index algorithm is finalized.

Previous research examining device-measured occupational and leisure-time physical activity among blue-collar workers has investigated associations with subsequent use of prescribed medication for musculoskeletal pain.12 These findings provide relevant context for occupational exposure monitoring but do not establish specific mechanical-load thresholds or demonstrate that individualized workload budgeting reduces low back pain. Lifting, carrying, and pushing/pulling activities will be identified through a combination of sensor-derived movement patterns, task classification, and structured occupational diary records. Because IMUs alone cannot directly determine the weight of handled objects or internal spinal loading, these variables will be treated as estimated occupational exposure measures requiring appropriate validation. However, translating inertial motion capture measurements into real-world ergonomic risk assessments presents methodological challenges, particularly when estimating biomechanical loading from wearable-derived movement data.13 Therefore, raw kinematic data may be supplemented by machine learning methods for classifying lifting-posture patterns using time- and frequency-domain features.14 These classifications will describe movement characteristics rather than establish universally safe or unsafe postures or individual injury thresholds.

Lindex=Σ(wi×Ei)+αF

As expressed in Equation 1, the proposed cumulative workload index (Lindex) combines normalized occupational exposure variables (Ei), including lifting frequency, task duration, and trunk movement characteristics, using weighting coefficients (wi). A normalized subjective fatigue score (F), weighted by parameter (α), is incorporated to account for perceived exertion. During the initial feasibility stage, exposure variables will be normalized against participant-specific baseline measurements. Preliminary weighting coefficients will be selected using clinical input and observed occupational exposure patterns, evaluated through sensitivity analyses, and finalized before the randomized trial begins. The resulting index will provide an exploratory estimate of cumulative occupational exposure and fatigue rather than a direct measurement of spinal loading or tissue injury risk. Its measurement reliability, sensitivity to changes in workload, and association with next-morning symptom responses will be evaluated before the algorithm is applied in the randomized controlled trial.

Fatigue-sensitive lifting-load budget

The personalized “back-load budget” operationalizes the concept that occupational load tolerance may fluctuate according to fatigue, symptoms, and recovery. During the initial feasibility stage, each participant will undergo a standardized baseline assessment incorporating job demands, functional lifting and carrying capacity, trunk movement characteristics, pain during occupational activities, post-shift fatigue, and next-morning symptom recovery. Initial wearable measurements and structured diary records will be used to establish an individualized reference range for occupational exposure. Physical therapists will review these measurements alongside clinical findings to determine provisional workload limits, rather than treating sensor-derived values as validated thresholds for tissue injury. The feasibility stage will evaluate the consistency of these measurements and the practicality of adjusting workload limits before the approach is tested in the randomized controlled trial.

The budget operates on a traffic-light decision system, triangulating multimodal fatigue data. Previous research has investigated multimodal fatigue detection using kinematic signals and machine learning, including wearable-derived movement features.15,16 However, the accuracy and applicability of these methods depend on the study population, sensor configuration, and validation procedures; their performance in heavy-duty occupational settings will require evaluation during the feasibility stage.

The criteria are outlined below:

Table 2: Traffic-Light Decision System Criteria

Status

Objective IMU Exposure

Subjective Diary Response

Required Action

Green

Within 80% of baseline tolerance

Full next-morning recovery; RPE < 5/10

Maintain current exposure; proceed with standard work conditioning.

Yellow

80-100% of baseline tolerance

Incomplete recovery; moderate post-shift fatigue

Implement pacing; modify workload; increase microbreaks; adjust PT intensity.

Red

Exceeds 100% of individualized baseline workload; unusual movement patterns trigger sensor-data review.

Concerning symptoms; severe flare-up; no recovery

Temporarily pause or modify the relevant occupational tasks as a precaution; arrange therapist reassessment and escalate concerning symptoms for medical evaluation.

The proposed 80% and 100% workload boundaries are provisional operational thresholds rather than clinically validated injury limits. The 100% boundary represents the individualized reference workload established during baseline assessment, while the 80% boundary provides a preliminary buffer intended to prompt workload adjustments before that reference level is reached. These values are proposed for feasibility testing, not as established physiological safety thresholds, and will be reviewed against observed symptoms, adherence, and clinical assessments before the randomized trial.

Classification will incorporate both objective occupational exposure and subjective fatigue and symptom recovery. When these indicators conflict, the more conservative classification will apply. For example, severe next-morning pain or incomplete recovery may trigger a Red or Yellow classification even when measured exposure remains below the individualized workload threshold. Red-flag symptoms will require immediate clinical assessment regardless of wearable measurements. Physical therapists will review classifications, document the reasons for adjustments, and recalibrate individual thresholds according to clinical findings and recovery patterns. The initial feasibility stage will assess the consistency, safety, and practicality of these decision rules before their effectiveness is evaluated in the randomized controlled trial.

A flowchart of a traffic-light system using green, yellow, and red categories to guide occupational load for workers.

Figure 1: Proposed fatigue-sensitive traffic-light decision system for occupational load management

Physical therapy–led work-conditioning intervention

The physical therapy intervention will be tightly coupled to the fatigue-sensitive load budget, ensuring that rehabilitation intensity complements, rather than exacerbates, occupational exposure. Progressive rehabilitation will incorporate targeted strength and endurance training focusing on trunk and hip strength, trunk-muscle endurance, and functional lower-limb strengthening. Motor control training will be incorporated to support coordinated trunk-muscle activation and movement control. Previous research has examined motor control exercises for chronic low back pain, although improvements in pain and disability should not be attributed solely to changes in transversus abdominis activation.17 Furthermore, research comparing externally loaded resistance training with unloaded exercise for chronic low back pain suggests that both approaches may improve pain and functional outcomes, without consistently demonstrating the superiority of external loading.18 Within the proposed intervention, progressive weighted occupational simulations are therefore included to develop task-specific functional capacity rather than on the assumption that loaded exercise produces greater pain relief.

Work-task simulation will replicate floor-to-waist lifting, waist-to-shoulder lifting, carrying, pushing/pulling, and job-specific transfers. Movement variability strategies will be taught to help workers avoid unnecessarily prolonged or repetitive static loading. Aerobic conditioning will build cardiovascular endurance appropriate to the heavy-duty demands of logistics and construction. Graded occupational exposure will systematically introduce previously feared or poorly tolerated activities, managing fear-avoidance behaviors.

Crucially, the program features fatigue-aware progression: therapists will adjust work-task volume and intensity according to occupational exposure, reported fatigue, next-morning symptom recovery, and clinical assessment rather than applying a fixed mathematical relationship between fatigue and rehabilitation intensity. If a worker presents in a “Yellow” state due to a heavy shift, the therapist will prioritize pacing, microbreaks, and mobility over high-intensity loaded simulations. Ergonomic and organizational modifications—such as adjusting lift distances, work heights, and handling procedures—will be collaboratively designed. Weekly intervention progression is therefore dynamically informed by both the individualized load budgets and ongoing therapist assessment.

Worker-facing feedback and occupational self-management

To facilitate occupational self-management, participants will utilize a worker-facing feedback dashboard accessible via smartphone. The dashboard will summarize cumulative occupational activity, the frequency and duration of monitored trunk positions, perceived work-shift fatigue, symptom trends, next-morning recovery scores, and the currently assigned workload budget (Green, Yellow, or Red). It will provide actionable recommendations for pacing or task modification based on the integrated data.

This feedback mechanism is designed to support collaborative decision-making between workers and their treating physical therapists. Employers may participate in implementing agreed-upon workplace accommodations but will not receive individual wearable measurements, pain diaries, fatigue scores, or clinical classifications without the worker’s separate, explicit authorization. However, ethical and privacy considerations are paramount. Qualitative research in occupational settings reveals that while real-time biometric tracking is supported by leadership for safety, rank-and-file workers often oppose it due to concerns over autonomy, privacy, and potential job security impacts19.

Therefore, the dashboard will strictly contextualize data as a tool for recovery and workload pacing, avoiding any suggestion that sensor-derived measures establish direct injury thresholds or independently diagnose low back pathology. Individual monitoring records will be stored securely using encrypted transmission and storage, with access restricted to authorized research and clinical personnel. Identifying information will be separated from research datasets, and data retention and deletion procedures will be specified in the approved study protocol.

Participants will be informed about data collection, permitted uses, access rights, and withdrawal procedures during consent. Reports of severe or worsening symptoms will trigger therapist review and, when clinically indicated, referral for medical assessment. Suspected serious neurological symptoms or other clinical red flags will require urgent medical evaluation independently of the wearable-derived workload classification. By utilizing standard hip or trunk placement typical of objective physical activity measurement20, the system maintains clinical relevance while prioritizing worker autonomy.

Outcome measures

The selection of outcome measures is critical for evaluating the proposed intervention. The primary endpoint will be the rate of new low-back-pain flare-up episodes per valid diary observation day during the eight-week randomized intervention. A new flare-up episode will be defined as a participant-reported increase of at least 2 points on the 0–10 Numeric Pain Rating Scale relative to the participant’s prespecified baseline pain level, accompanied by a reported worsening beyond usual day-to-day symptom variation. An episode will remain ongoing until pain returns to within 1 point of the baseline level for at least two consecutive days.

The proposed two-point increase and two-day recovery criteria are provisional operational definitions that will be reviewed for clinical appropriateness and measurement feasibility before the randomized trial. Any refinements will be finalized and prespecified before participant enrollment in the randomized stage. Consecutive days meeting the flare-up criteria will be counted as one episode rather than multiple new episodes. The total number of days spent in a flare-up will be evaluated separately as a secondary outcome.

Participants will complete a structured smartphone diary each morning to record pain intensity, perceived recovery, and relevant symptoms. Missing diary entries will not automatically be classified as flare-free days. Valid observation days will be defined according to prespecified diary-completeness criteria, and incomplete episode boundaries will be documented and addressed through sensitivity analyses. This approach distinguishes flare-up incidence from flare-up duration while accounting for differences in diary adherence.

The selection of outcome measures is critical for evaluating the proposed intervention. The primary endpoint will be the rate of next-day low-back-pain flare-up days per valid diary observation day during the eight-week randomized intervention. A flare-up will be operationally defined as a participant-reported increase of at least 2 points on the 0–10 Numeric Pain Rating Scale compared with the preceding day’s recorded pain level. Participants will complete a structured smartphone diary each morning to record pain intensity, perceived recovery, and relevant symptoms. Days without a valid current-day or preceding-day pain rating will not be classified as flare-up or non-flare-up days. Diary completion rates and reasons for missing entries will be documented, with sensitivity analyses assessing the potential influence of incomplete reporting. Flare-up severity, measured by the magnitude of pain increase, will remain a secondary outcome. This approach directly evaluates whether individualized workload budgeting reduces next-day symptom exacerbations while accounting for differences in diary adherence.

Secondary outcomes are designed to evaluate whether acute flare-up reductions translate into sustained occupational capacity over time. These include objective functional and disability metrics such as the Oswestry Disability Index (ODI) and standard Functional Capacity Evaluation (FCE) measures. Additional secondary outcomes encompass the Numeric Pain Rating Scale, Work Ability Index, work-task lifting and carrying tolerance, fatigue ratings during and after work, objective occupational exposure measures (via IMU), trunk/hip strength and endurance, Fear-Avoidance Beliefs Questionnaire, pain self-efficacy, missed workdays, modified-duty status, and sustainable return-to-work outcomes. Treatment attendance, home-exercise adherence, and participant satisfaction will serve as exploratory measures of intervention acceptability.

Proposed statistical analysis

The prospective statistical analysis framework will adhere to intention-to-treat principles. Descriptive participant characteristics will be summarized to evaluate baseline comparability between the experimental and control groups. Between-group comparisons for continuous secondary outcomes (e.g., ODI, Work Ability Index) will utilize longitudinal mixed-effects modeling to account for repeated measures across the baseline, 4-week, 8-week, and 6-month time points, incorporating random intercepts for participants to handle within-subject correlation.

For the primary endpoint, defined as the rate of new low-back-pain flare-up episodes per valid diary observation day during the eight-week randomized intervention, negative binomial regression will be used to compare episode incidence rates between the experimental and control groups. The model will include the logarithm of valid diary observation days as an offset, with observation time adjusted where missing diary entries prevent reliable identification of new episodes. Treatment allocation will be the primary predictor, with baseline pain severity and prespecified occupational characteristics included as covariates. Results will be reported as adjusted incidence rate ratios with 95% confidence intervals. Model assumptions, including overdispersion and the suitability of the negative binomial distribution, will be evaluated. Secondary analyses will examine the total number of days spent in flare-up episodes, while sensitivity analyses will assess the influence of missing diary entries, incomplete episode boundaries, and alternative flare-up definitions.

Work participation outcomes will be evaluated using measures of sustained employment, modified-duty status, and work absence during follow-up. Survival analysis (e.g., Cox proportional hazards models) will be considered only for clearly defined time-to-event outcomes, such as time to first sickness absence, where sufficient events and follow-up data are available. All results will report effect sizes and 95% confidence intervals. Missing-data procedures may utilize multiple imputation by chained equations when the missing-at-random assumption is considered plausible. The extent and patterns of missingness will be examined, and sensitivity analyses will assess the potential impact of departures from this assumption. Sensitivity analyses will test the robustness of the findings against different imputation models. Baseline pain severity, job demands, and baseline occupational exposure will be considered as prespecified covariates in the primary analysis, with the final covariate set established before the randomized trial begins.

Discussion and anticipated clinical implications

The proposed investigation anticipates that combining wearable- and diary-informed occupational exposure monitoring with PT-led work conditioning will substantially improve the management of persistent low back pain in heavy-duty environments. The potential contribution lies in shifting the rehabilitation paradigm from static, generic exercise prescriptions to dynamic, fatigue-responsive load management. Rehabilitation informed by individual fatigue responses and recovery patterns warrants further investigation because it may help account for delayed symptom exacerbations following demanding occupational activities, without assuming that wearable-derived exposure estimates directly measure biomechanical overload or predict injury.

If the hypotheses are supported, the clinical implications for occupational physical therapy are significant. Therapists could use hybrid monitoring to inform individualized adjustments to graded occupational exposure, potentially reducing cycles of excessive activity followed by symptom exacerbation and reduced work participation. Implementation feasibility relies on balancing resource requirements: while high-fidelity IMUs offer objective precision, lower-cost diary-based monitoring may serve as a scalable alternative for broad industrial application. It is crucial to distinguish established evidence—such as the efficacy of multi-domain return-to-work programs—from the proposed hypotheses regarding real-time workload budgeting, which this trial aims to empirically test.

Limitations and future research

Several methodological and practical limitations must be acknowledged. First, differences in occupational demands and injury histories across warehousing, logistics, and construction sectors may introduce heterogeneity that complicates the standardization of the load budget. Measurement limitations of wearable technology remain; IMUs cannot directly measure internal spinal forces, necessitating reliance on kinematic proxies and subjective diaries. Consequently, potential inaccuracies and recall bias in self-reported work diaries must be factored into the analysis. Furthermore, the lack of universally validated lifting-load thresholds requires the trial to rely on individualized baselines, which may be complex to calibrate initially.

Additionally, the proposed workload-budget algorithm and traffic-light thresholds have not yet undergone feasibility testing or clinical validation. The initial restriction to workers who are actively employed, including those on modified duties, also limits the applicability of findings to individuals who are completely off work or transitioning back from medical leave.

Methodologically, the inability to blind participants and treating therapists completely poses a risk of performance bias. Workplace compliance and the willingness of employers to accommodate temporary “Yellow” or “Red” budget modifications represent significant external variables that could impact participant adherence and follow-up challenges. Generalizability across other occupational groups (e.g., healthcare workers, agriculture) remains uncertain. Future clinical validation and implementation research should focus on refining the algorithms that integrate IMU and diary data, exploring cost-effective sensor alternatives, and evaluating employer-level economic outcomes.

Conclusion

This proposed research protocol outlines a rigorous methodological framework to evaluate an individualized, fatigue-sensitive lifting-load budget integrated with physical therapist–led work conditioning for heavy-duty workers with persistent low back pain. By operationalizing a traffic-light decision system based on hybrid wearable and diary data, the intervention seeks to dynamically align daily occupational exposure with physiological recovery. This approach addresses a gap in occupational rehabilitation by investigating whether individualized management of cumulative occupational exposure, fatigue, and recovery can reduce subsequent low back pain exacerbations.

The proposed two-stage clinical investigation will first evaluate the feasibility, reliability, and safety of wearable monitoring, smartphone diary reporting, individualized workload calibration, and therapist-guided adjustments. Subject to satisfactory feasibility findings, the subsequent randomized controlled trial will evaluate whether integrating a fatigue-sensitive back-load budget with conventional work conditioning reduces the rate of new low-back-pain flare-up episodes during the eight-week intervention and improves functional and occupational outcomes compared with conventional work conditioning alone. Although clinical effectiveness remains to be established, this framework offers a testable approach to integrating occupational exposure monitoring with individualized rehabilitation and sustainable return-to-work planning.

REFERENCES AND NOTES

  1. Suri, P., Timmons, A. K. I., Korpak, A. M., Tanus, A. D., Brubeck, H. F., Daniels, C. J., Scott, H., Morelli, D., Costa, N., Hodges, P. W., Day, M. A., Friedly, J. L., Heagerty, P. J., & Jensen, M. P. (2025). Transient and long-term risks of common physical activities in people with low back pain. JAMA Network Open, 8(12), Article e2547915. https://doi.org/10.1001/jamanetworkopen.2025.47915
  2. Amorim, A. B., Simic, M., Pappas, E., Zadro, J. R., Carrillo, E., Ordoñana, J. R., & Ferreira, P. H. (2019). Is occupational or leisure physical activity associated with low back pain? Insights from a cross-sectional study of 1059 participants. Brazilian Journal of Physical Therapy, 23(3), 257–265. https://doi.org/10.1016/j.bjpt.2018.06.004
  3. Hasenbring, M. I., Andrews, N. E., & Ebenbichler, G. (2021). Approach to activity, biomechanical loads, and flare-ups of back pain. International Association for the Study of Pain. https://www.iasp-pain.org/resources/fact-sheets/approach-to-activity-biomechanical-loads-and-flare-ups-of-back-pain/
  4. Nurse, C. A., Elstub, L. J., Volgyesi, P., & Zelik, K. E. (2023). How accurately can wearable sensors assess low back disorder risks during material handling? Exploring the fundamental capabilities and limitations of different sensor signals. Sensors, 23(4), Article 2064. https://doi.org/10.3390/s23042064
  5. Matijevich, E. S., Volgyesi, P., & Zelik, K. E. (2021). A promising wearable solution for the practical and accurate monitoring of low back loading in manual material handling. Sensors, 21(2), Article 340. https://doi.org/10.3390/s21020340
  6. Diraneyya, M. M., Ryu, J., Abdel-Rahman, E., & Haas, C. T. (2021). Inertial motion capture-based whole-body inverse dynamics. Sensors, 21(21), Article 7353. https://doi.org/10.3390/s21217353
  7. Schall, M. C., Jr., Sesek, R. F., & Cavuoto, L. A. (2018). Barriers to the adoption of wearable sensors in the workplace: A survey of occupational safety and health professionals. Human Factors, 60(3), 351–362. https://doi.org/10.1177/0018720817753907
  8. Schaafsma, F. G., Whelan, K., van der Beek, A. J., van der Es-Lambeek, L. C., Ojajärvi, A., & Verbeek, J. H. (2013). Physical conditioning as part of a return to work strategy to reduce sickness absence for workers with back pain. Cochrane Database of Systematic Reviews, 2013(8), Article CD001822. https://doi.org/10.1002/14651858.CD001822.pub3
  9. Cullen, K. L., Irvin, E., Collie, A., Clay, F., Gensby, U., Jennings, P. A., Hogg-Johnson, S., Kristman, V. L., Laberge, M., McKenzie, D., Newnam, S., Palagyi, A., Ruseckaite, R., Sheppard, D. M., Shourie, S., Steenstra, I. A., Van Eerd, D., & Amick, B. C., III. (2018). Effectiveness of workplace interventions in return-to-work for musculoskeletal, pain-related and mental health conditions: An update of the evidence and messages for practitioners. Journal of Occupational Rehabilitation, 28(1), 1–15. https://doi.org/10.1007/s10926-016-9690-x
  10. Lutz, A. D., Rodriguez, S., DeBuck, K. F., Denninger, T. R., Myers, L. M., & Myers, N. L. (2026). Work conditioning: The appropriate stepped physical therapy care for injured workers in physically demanding roles: A retrospective cohort study. Journal of Occupational and Environmental Medicine, 68(5), 385–390. https://doi.org/10.1097/JOM.0000000000003611
  11. Maas, E. T., Koehoorn, M., & McLeod, C. B. (2021). Does gradually returning to work improve time to sustainable work after a work-acquired musculoskeletal disorder in British Columbia, Canada? A matched cohort effectiveness study. Occupational and Environmental Medicine, 78(10), 715–723. https://doi.org/10.1136/oemed-2020-107014
  12. Kyriakidis, S., Mose, S., Søgaard, K., Holtermann, A., Rasmussen, C. D. N., & Gupta, N. (2025). Occupational and leisure-time physical activity and prospective musculoskeletal pain-relief prescribed medication among blue-collar workers: 24-h device-measured physical activity and Danish registers. The Journal of Pain, 31, Article 105387. https://doi.org/10.1016/j.jpain.2025.105387
  13. Nail-Ulloa, I., Zabala, M., Henriquez-Schott, M., & Gallagher, S. (2025). From lab to field: Translating inertial motion capture into applied ergonomic risk assessment and intervention. Human Factors and Wearable Technologies, 175, 30–40. https://doi.org/10.54941/ahfe1006356
  14. Prisco, G., Romano, M., Esposito, F., Cesarelli, M., Santone, A., Donisi, L., & Amato, F. (2024). Capability of machine learning algorithms to classify safe and unsafe postures during weight lifting tasks using inertial sensors. Diagnostics, 14(6), Article 576. https://doi.org/10.3390/diagnostics14060576
  15. Hwang, S., Kwon, N., Lee, D., Kim, J., Yang, S., Youn, I., Moon, H.-J., Sung, J.-K., & Han, S. (2025). A multimodal fatigue detection system using sEMG and IMU signals with a hybrid CNN-LSTM-attention model. Sensors, 25(11), Article 3309. https://doi.org/10.3390/s25113309
  16. Kakhi, K., Jagatheesaperumal, S. K., Khosravi, A., Alizadehsani, R., & Acharya, U. R. (2025). Fatigue monitoring using wearables and AI: Trends, challenges, and future opportunities. Computers in Biology and Medicine, 195, Article 110461. https://doi.org/10.1016/j.compbiomed.2025.110461
  17. Shanbehzadeh, S., ShahAli, S., Hides, J., Ebrahimi-Takamjani, I., & Rasouli, O. (2022). Effect of motor control training on trunk muscle morphometry, pain, and disability in people with chronic low back pain: A systematic review and meta-analysis. Journal of Manipulative and Physiological Therapeutics, 45(3), 202–215. https://doi.org/10.1016/j.jmpt.2022.06.003
  18. Ranzani, M., Pozzi, A., Fornasari, D., Ristori, D., & Testa, M. (2025). Is resistance training with external loads superior to unloaded exercise in the management of chronic low back pain? A systematic review and meta-analysis. Archives of Physiotherapy, 15, 297–313. https://doi.org/10.33393/aop.2025.3533
  19. Topazian, R. J., Wec, A., Ali, J., Frattaroli, S., Locke, P., Taylor, J. A., & Crifasi, C. K. (2025). Ethical use of wearable device data in occupational settings: Perspectives from the fire service. Journal of Occupational and Environmental Medicine, 67(9), e621–e629. https://doi.org/10.1097/JOM.0000000000003446
  20. Berger, M., Bertrand, A. M., Robert, T., & Chèze, L. (2023). Measuring objective physical activity in people with chronic low back pain using accelerometers: A scoping review. Frontiers in Sports and Active Living, 5, Article 1236143. https://doi.org/10.3389/fspor.2023.1236143
Home » Fatigue-Sensitive Wearable Load Budget and Work-Conditioning for Heavy-Duty Workers
© Hampton Global 2026.
Join our newsletter
Stay up to date on latest stories