Intended for healthcare professionals

Letters Evidence base of guidelines

Artificial intelligence could bring relevant guidelines into every consultation

BMJ 2019; 366 doi: (Published 07 August 2019) Cite this as: BMJ 2019;366:l4788
  1. Christopher L Manning, retired general practitioner
  1. Teddington TW11 9HG, UK
  1. chrisso95{at}

Elwenspoek and colleagues review monitoring strategies in UK guidelines for chronic diseases.1 Over a decade ago, I visited several publishers of clinical guideline resources for general practitioners and suggested that primary care IT providers develop an artificial intelligence (AI) engine that would interact with the data entered by the doctor during the consultation (free text, Read codes, etc) and be linked to the National Institute for Health and Care Excellence guidelines on specific conditions.

The AI would offer prompts and direct guidance, thereby effectively delivering, in interactive form, real time assistance and bringing relevant guidelines and evidence (such as those in figure 1 of the article) into every consultation.

GPs are busier than ever, and NICE guidelines are essentially protocols based on “expert opinion” that lack “robust evidence for optimal monitoring strategies and testing intervals," so perhaps experts and relevant primary care professional bodies should get their heads together and develop a system that will help busy doctors and concerned patients to tackle (not just be aware of) the “uncertainties when making shared decisions about chronic disease monitoring.” Worthy aspirations are no good; frontline practitioners need specific actions and resources that will enable every consultation to be safe, sound, and supportive for both patient and doctor.


  • Competing interests: None declared.


View Abstract

Log in

Log in through your institution


* For online subscription