Evaluation of a Musculoskeletal Digital Assessment Routing Tool (DART): Crossover Noninferiority Randomized Pilot Trial
Bibliographic record
Abstract
BACKGROUND: Musculoskeletal conditions account for 16% of global disability, resulting in a negative effect on patients and increasing demand for health care use. Triage directing patients to appropriate level intervention improving health outcomes and efficiency has been prioritized. We developed a musculoskeletal digital assessment routing tool (DART) mobile health (mHealth) system, which requires evaluation prior to implementation. Such innovations are rarely rigorously tested in clinical trials-considered the gold standard for evaluating safety and efficacy. This pilot study is a precursor to a trial assessing DART performance with a physiotherapist-led triage assessment. OBJECTIVE: The study aims to evaluate trial design, assess procedures, and collect exploratory data to establish the feasibility of delivering an adequately powered, definitive randomized trial, assessing DART safety and efficacy in an NHS primary care setting. METHODS: A crossover, noninferiority pilot trial using an integrated knowledge translation approach within a National Health Service England primary care setting. Participants were patients seeking assessment for a musculoskeletal condition, completing a DART assessment and the history-taking element of a face-to-face physiotherapist-led triage in a randomized order. The primary outcome was agreement between DART and physiotherapist triage recommendation. Data allowed analysis of participant recruitment and retention, randomization, blinding, study burden, and potential barriers to intervention delivery. Participant satisfaction was measured using the System Usability Scale. RESULTS: Over 8 weeks, 129 patients were invited to participate. Of these, 92% (119/129) proceeded to eligibility assessment, with 60% (78/129) meeting the inclusion criteria and being randomized into each intervention arm (39/39). There were no dropouts and data were analyzed for all 78 participants. Agreement between physiotherapist and DART across all participants and all primary triage outcomes was 41% (32/78; 95% CI 22-45), intraclass correlation coefficient 0.37 (95% CI 0.16-0.55), indicating that the reliability of DART was poor to moderate. Feedback from the clinical service team led to an adjusted analysis yielding of 78% (61/78; 95% CI 47-78) and an intraclass correlation coefficient of 0.57 (95% CI 0.40-0.70). Participant satisfaction was measured quantitively using amalgamated System Usability Scale scores (n=78; mean score 84.0; 90% CI +2.94 to -2.94), equating to an "excellent" system. There were no study incidents, and the trial burden was acceptable. CONCLUSIONS: Physiotherapist-DART agreement of 78%, with no adverse triage decisions and high patient satisfaction, was sufficient to conclude DART had the potential to improve the musculoskeletal pathway. Study validity was enhanced by the recruitment of real-world patients and using an integrated knowledge translation approach. Completion of a context-specific consensus process is recommended to provide definitive definitions of safety criteria, range of appropriateness, noninferiority margin, and sample size. This pilot demonstrated an adequately powered definitive trial is feasible, which would provide evidence of DART safety and efficacy, ultimately informing potential for DART implementation. TRIAL REGISTRATION: ClinicalTrials.gov NCT04904029; http://clinicaltrials.gov/ct2/show/NCT04904029. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/31541.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.032 | 0.030 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.004 | 0.003 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.005 | 0.003 |
| Insufficient payload (model declined to judge) | 0.009 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".