Heart failure & cardiomyopathies
Triple cardiovascular disease detection with an artificial intelligence-enabled stethoscope (TRICORDER) in the UK: a cluster-randomised controlled implementation trial
Primary result · New HF diagnosis (per 1000 patient-years)Result — go to the resultIncidence rate ratio 0.94 95% CI 0.86–1.02Authors' conclusion
Implementation of an AI stethoscope in routine primary care did not significantly increase detection of heart failure or increase community-based diagnosis after 12 months of implementation.
Study context
- Population
- UK primary care practices
- Participants
- 205 practices
- Comparison
- Artificial intelligence (AI)-stethoscope implementation vs routine care
- Interventionn = 96 practices [701,933 registered patients]
- Controln = 109 practices [851,242 registered patients]reference
- Primary endpoint
- Incidence of any newly coded HF diagnosis (all subtypes) per 1000 patient-years (IRR); coprimary: HF detection by place of diagnosis (community vs hospital admission)
Methodological detailspragmatic, open-label (masking not feasible)
- Pragmatic
- Open-label (masking not feasible)
Primary result
New HF diagnosis (per 1000 patient-years)
Incidence rate ratio
0.94
95% CI 0.86–1.02
Between groups
IRR 0.94 (95% CI 0.86–1.02)
Per 1000 patient-years, as printed
As reported in the abstract
- No difference in community-based or hospital-based diagnoses
- P > 0.05
Secondary outcomes
| Outcome | Intervention | Control | |
|---|---|---|---|
| AI-stethoscope use | 12 725 patient examinations | ||
| Clinical users | 972 | Not reported in the abstract | Control: not reported in the abstract |
Original publication
Triple cardiovascular disease detection with an artificial intelligence-enabled stethoscope (TRICORDER) in the UK: a cluster-randomised controlled implementation trial. Lancet. Published 28 January 2026. doi:10.
Summarised from the published abstract; every number is checked against it