Improving Musculoskeletal Care Quality With a Web-Based Clinical Decision Support System in UK Primary Care: Mixed Methods Cluster-Randomized Controlled Trial

Key numbers

• 95% CI 2.52-59.9
• 18% reduction in carbon emissions
• 10% annual cost savings if

Background Musculoskeletal (MSK) conditions account for one in seven general practice consultations in the United Kingdom. Despite national guidelines, suboptimal care remains common. Digital clinical decision support systems (CDSSs) offer a promising route to improve care quality but have not been robustly evaluated in MSK primary care. Objective This study aimed to evaluate the efficacy, clinical effectiveness and cost-effectiveness, carbon reduction, safety, and implementation of a CDSS (CrossCover) for MSK consultations in UK general practice. Methods We conducted a mixed methods, cluster-randomized controlled trial in 19 general practices between February and July 2024. Practices were randomized 1:1, stratified by list size. Intervention practices received training and access to the web-based CDSS for 3 months. We analyzed anonymized electronic health records to identify all adult MSK consultations. The primary outcome was the proportion of patient records containing one or more of 12 predefined non-evidence-based management options, assessed by blinded clinical review of 120 patient records (60 per arm). Reviewers assessed the complete 3-month record in both arms; in the intervention arm, suboptimal management was counted only where it occurred at a consultation in which the CDSS was used. Secondary outcomes included intention-to-treat treatment comparisons of self-management advice, prescriptions, referrals, and imaging. Clinician engagement, safety, environmental and economic modeling, and clinician experiences from semistructured interviews were evaluated. Outcome assessors were blinded; clinicians were not. Results Among 10,182 eligible patients with MSK conditions (4268 intervention; 5914 control), the CDSS was used in 365 (8.6%) patient cases. Intention-to-treat analysis identified that the intervention increased self-management alone. Per-protocol analysis demonstrated that when used, suboptimal care arose in 3% (2/60) of CDSS-guided consultations vs 25% (15/60) of control consultations (adjusted odds ratio 12.30, 95% CI 2.52-59.9; P=.002). It also increased self-management advice and community referrals, and it reduced prescribing and fit-note use. Safety concerns were logged in 6.6% (24/365) of uses with no high-risk issues identified. Environmental modeling estimated an 18% reduction in carbon emissions and 10% annual cost savings if the intervention were scaled. Interviews (n=18) highlighted positive perceptions of the CDSS's content and safety netting but noted challenges in integration. Conclusions When used, the CrossCover CDSS significantly reduced suboptimal primary care MSK management. However, while safety and clinician feedback were favorable, overall engagement was limited, underscoring the need for improved digital integration, workflow alignment, and implementation support.

Jonathan Hill, Hollie Birkinshaw, Muhammed Usman, Tammy Millward, James Bailey, Simon Wathall, Rebecca Naylor, Robert Malcolm, Melissa Pegg, Hayden Holmes, Kate Lee, Tanya Wright, Adrian Chudyk, Alex Braybrooke, Roanna Burgess, Martyn Lewis

Leverage2 / 5

Reusable within one subfield (a technique, dataset, or protocol a few groups will adopt).

10% of this desk’s rubric

Magnitude3 / 5

Large gain: roughly 2x, or a clear new state of the art on a hard, unsaturated problem.

20% of this desk’s rubric

Evidence5 / 5

Definitive: phase 3 randomized evidence on hard endpoints, multi-lab replication, or community verification at scale.

32% of this desk’s rubric

Novelty2 / 5

A new combination of known ideas.

8% of this desk’s rubric

Trajectory2 / 5

Some room to improve with obvious engineering.

5% of this desk’s rubric

Stakes2 / 5

Benefits a professional community (practitioners, clinicians, engineers).

25% of this desk’s rubric

• Citations1 (reference 20, via openalex, Oct 20, 2026, 06:17 UTC)

• Reviewed by heuristic-v7 on Oct 20, 2026, 06:17 UTC. Paper type: clinical.
• Categories: research-article, Randomized Controlled Trial, Journal Article, Humans, Musculoskeletal Diseases, Internet, Decision Support Systems, Clinical, Adult, Middle Aged, Cost-Benefit Analysis
• TOP, No.4 in the Medicine edition of October 20, 2026.