UMIN-CTR Clinical Trial

Public title

Effects of AI-based real-time feedback on low back load during patient transfer assistance: a randomized controlled trial

Acronym

AI-Transfer Feedback Trial

Scientific Title

Immediate effects of AI-based real-time feedback of lumbar moment on biomechanical low back load during patient transfer: a randomized controlled trial

Scientific Title:Acronym

AI-Transfer RCT

Region

Japan

Condition

Low back load during patient transfer assistance in healthcare workers (healthy volunteers)

Classification by specialty

Not applicable

Classification by malignancy

Others

Genomic information

NO

Narrative objectives1

To examine the immediate effects of AI-based real-time feedback of lumbar moment waveforms, derived from markerless motion analysis (VP-Ergono), in addition to standard transfer instruction and review of participants' own video, on biomechanical low back load (peak lumbar moment and cumulative lumbar load) during patient transfer among healthcare workers who routinely perform transfer assistance. The association between changes in mechanical load and acute subjective burden (pre-to-post changes in pain and fatigue) is also examined. This is a single-session, immediate intervention; effects on chronic low back pain are not evaluated. The direction of change in lumbar load is not prespecified.

Basic objectives2

Efficacy

Basic objectives -Others

Trial characteristics_1

Trial characteristics_2

Developmental phase

Primary outcomes

Co-primary outcomes (both derived from markerless motion analysis of video; between-group comparison of pre-to-post change): (1) peak lumbar moment (Nm) during patient transfer, and (2) cumulative lumbar load (Nm x s, natural-log transformed). Multiplicity is adjusted by the Bonferroni method (two-sided alpha = 0.025 for each). The direction of change is not prespecified; two-sided tests are used.

Key secondary outcomes

Secondary: (a) pain VAS (0-100 mm) and (b) fatigue VAS (0-100 mm), change from immediately before to after the transfer task; (c) median %RVE of right (primary) and left (comparator) multifidus by surface EMG, normalised to a 6-kg trunk-flexion reference contraction. Exploratory: integrated EMG (adequate electrode contact only), rate of change in lumbar moment, right-left asymmetry.
Additional secondary/exploratory outcomes:
(1) Baseline occupational exposure: transfers per working day, patients in charge, night shifts (yes/no, per month), years of clinical experience (self-reported on the record sheet).
(2) Low back pain: presence of LBP of VAS 30 mm or more within the preceding 3 months (the stratification factor) and presence of LBP within the preceding 7 days.
(3) Estimated daily cumulative lumbar load = cumulative lumbar moment per transfer x self-reported transfers per working day.
(4) Reliability and convergent validity: trial-to-trial reliability of lumbar moment measures from repeated baseline trials (ICC, SEM, MDC) and correspondence with simultaneously recorded multifidus %RVE.
(5) Implementation cost: equipment cost and time required for measurement and feedback per participant.
(6) Effect modification by history of LBP (VAS 30 mm or more within 3 months, as stratified), pre-specified exploratory; by age and years of experience, exploratory.

Study type

Interventional

Basic design

Parallel

Randomization

Randomized

Randomization unit

Individual

Blinding

Open -but assessor(s) are blinded

Control

Active

Stratification

YES

Dynamic allocation

NO

Institution consideration

Institution is not considered as adjustment factor.

Blocking

YES

Concealment

Numbered container method

No. of arms

2

Purpose of intervention

Educational,Counseling,Training

Type of intervention

Device,equipment

Behavior,custom

Interventions/Control_1

In addition to standard transfer instruction, participants review video of their own transfer performance and receive real-time feedback of lumbar moment waveforms derived from markerless motion analysis (VP-Ergono) (single session).

Interventions/Control_2

In addition to standard transfer instruction, participants review video of their own transfer performance without presentation of lumbar moment waveforms (single session). Both groups receive standard instruction and video review, so the between-group difference is limited to the presence or absence of AI feedback (waveform presentation).

Interventions/Control_3

Interventions/Control_4

Interventions/Control_5

Interventions/Control_6

Interventions/Control_7

Interventions/Control_8

Interventions/Control_9

Interventions/Control_10

Age-lower limit

Not applicable

Age-upper limit

Not applicable

Gender

Male and Female

Key inclusion criteria

Physical therapists, occupational therapists, speech-language-hearing therapists, nurses, and care workers employed by the medical corporation who provide written informed consent.

Key exclusion criteria

(1) Severe acute low back pain or neurological symptoms that make it difficult to perform the measurement task; (2) history of spinal surgery; (3) marked limitation of lower-limb joint range of motion; (4) any other condition judged by the principal investigator to make participation unsuitable.

Target sample size

60

Name of lead principal investigator

1st name	Daichi
Middle name
Last name	Yoshida

Organization

Fukuoka Sakurajuji Medical Corporation Hanabatake Hospital / Fukuoka Sakurajuji Medical Corporation (Sakurajuji Advanced Rehabilitation Center SACRA)

Division name

Rehabilitation Department

Zip code

830-0038

Address

914 Nishimachi, Kurume City, Fukuoka Prefecture

TEL

0942-32-4565

Email

mugonjikkou@yahoo.co.jp

Name of contact person

1st name	Daichi
Middle name
Last name	Yoshida

Organization

Fukuoka Sakurajuji Medical Corporation Hanabatake Hospital

Division name

Rehabilitation Department

Zip code

830-0038

Address

914 Nishimachi, Kurume City, Fukuoka Prefecture

TEL

09092057660

Homepage URL

Email

mugonjikkou@yahoo.co.jp

Institute

Fukuoka Sakurajuji Medical Corporation (Sakurajuji Advanced Rehabilitation Center SACRA)

Institute

Department

Personal name

Daichi Yoshida

Organization

Japan Society for the Promotion of Science

Organization

Division

Category of Funding Organization

Japanese Governmental office

Nationality of Funding Organization

Japan

Co-sponsor

Faculty of Rehabilitation, Nishi Kyushu University

Name of secondary funder(s)

Organization

Institutional Review Board of Sakurajyuji Fukuoka Hospital

Address

5-11 Watanabe-dori 3-chome, Chuo Ward, Fukuoka City

Tel

092-791-1100

Email

mugonjikkou@yahoo.co.jp

Secondary IDs

YES

Study ID_1

25K13431

Org. issuing International ID_1

Japan Society for the Promotion of Science (JSPS)

Study ID_2

Org. issuing International ID_2

IND to MHLW

Institutions

医療法人福岡桜十字　花畑病院（福岡県）

Date of disclosure of the study information

2026

Year

08

Month

10

Day

URL releasing protocol

Publication of results

Unpublished

URL related to results and publications

Number of participants that the trial has enrolled

Results

Results date posted

Results Delayed

Results Delay Reason

Date of the first journal publication of results

Baseline Characteristics

Participant flow

Adverse events

Outcome measures

Plan to share IPD

IPD sharing Plan description

Recruitment status

Open public recruiting

Date of protocol fixation

2026

Year

02

Month

17

Day

Date of IRB

2026

Year

02

Month

17

Day

Anticipated trial start date

2026

Year

02

Month

18

Day

Last follow-up date

2028

Year

03

Month

31

Day

Date of closure to data entry

Date trial data considered complete

Date analysis concluded

Other related information

Design details fixed before resumption of measurement (Oct 2026) and before any between-group analysis. Randomisation: stratified permuted blocks (block size 4) within 4 strata defined by professional category (rehabilitation professionals: PT/OT/ST vs nursing/care staff) and history of low back pain (VAS 30 mm or more within the preceding 3 months). Concealment: sequentially numbered, opaque, sealed envelopes prepared by a co-investigator not involved in measurement, one opened per participant after baseline. Three external pilot participants are excluded; 10 newly employed staff randomised in Apr-May 2026 under this procedure are included (ITT). Recruitment plan: nursing/care staff (target 27, minimum 20) and all eligible rehabilitation staff are recruited in parallel. A blinded sample size re-estimation based on the pooled (group-unlabelled) variance of the primary outcome is pre-specified when 30 participants have completed both measurements; the final sample size will be set between 60 and 72, with no interim efficacy analysis. Primary analysis: ANCOVA adjusted for baseline, stratification factors, age, experience and BMI. Registered retrospectively (first participant Feb 2026). Outcome computation (SAP v3.4 s.4.1): lumbar moment despiked (5-frame median); each transfer is the contiguous interval above 50 Nm around the detected peak (gaps <1 s bridged; limit to be confirmed on the first 10 participants, blind to group); peak = maximum within that interval; cumulative load = time integral above 50 Nm within it; mean of 3 trials. Fixed window: sensitivity analysis. The waveform is shown only during the review immediately after the baseline transfers, not during the post-intervention transfers. Harms are recorded. Group labels are merged only after an external review and the final pre-unblinding SAP (v4.0). SAP v3.4 (25 Oct 2026) and code are deposited on Zenodo before unblinding (DOI to be added). Funded by JSPS KAKENHI 25K13431.

Registered date

2026

Year

08

Month

10

Day

Last modified on

2026

Year

10

Month

25

Day

Link to view the page

Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000069425