An article on artificial intelligence could now be published in every issue of JRSM – perhaps it will. This month, we turn our attention to the effect artificial intelligence is having on general practice consultations. The question is that while the emphasis, certainly from policy makers, is on the efficiency savings and productivity improvements that artificial intelligence can deliver, what is less discussed is the impact its use will have on the quality of the consultation and patient outcomes. 1
General practice is under pressure, possibly in crisis. It is stretched for resources, technologically challenged, and overburdened. Care is more complex than ever; new challenges, emerging models of care. 2 The rhetoric from the current government is adversarial – even confrontational. Any vindication that general practitioners were feeling over their concerns about physician associates in general practice has evaporated following the announcement of the NHS 10-year plan – a plan that, importantly, pays insufficient attention to inequalities. 3
Yet evidence continues to support the importance of continuity of care, something that the 10-year plan will erode further. The relationship between doctor and patient will be diminished, to the detriment of patient care and doctors’ job satisfaction. Add to this the soulless, impersonal nature of consultations powered by artificial intelligence, and you see a picture of a service that disconnects user from provider; a model of care that ignores the tried and tested adage that, “not everything that counts can be counted”.
What artificial intelligence continues to lack is compelling evidence of its benefits – although there is plenty of evidence that sounds like it might be compelling. Making evidence sound compelling comes in many forms. One of these is to perform repeated interim analyses in clinical trials but, as the latest contribution from the James Lind Library explains, repeated interim analyses can exaggerate treatment benefit. 4 It’s a description that might equally apply to the value of artificial intelligence in general practice.
References
- Alboksmaty A, Hayhoe BWJ, Majeed A, Neves A-L. Reclaiming the primary care consultation for patients and clinicians: is AI-enabled ambient voice technology the answer? J R Soc Med 2025; 118: 277–279. DOI: 10.1177/01410768251360853 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Franklin J, Thakur A, Patel V. How doctors think about their role in transgender care: a qualitative study of UK general practitioners and endocrinologists. J R Soc Med 2025; 118: 283–292. DOI: 10.1177/01410768251363407 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Powell RA, Bhui K, Singh N, Rao M, Sathyamoorthy G. A decade of deprioritisation? Ethnicity and health in the 10-year NHS plan. J R Soc Med 2025; 118: 280–282. DOI: 10.1177/01410768251366877 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gluud C, Thorlund K. Long overdue recognition of Klim McPherson’s 1974 article on sequential analysis of trial data JLL Bulletin: Commentaries on the history of treatment evaluation. J R Soc Med 2025; 118: 304–307. DOI: 10.1177/01410768251380376 [DOI] [PMC free article] [PubMed] [Google Scholar]