Kieran Walsh: WHO recommendations on digital clinical decision support for health system strengthening

The WHO has recently published recommendations on the use of digital clinical decision support for health system strengthening. Clinical decision support is a recognised tool to improve the performance of healthcare professionals in diagnosis and management. Poor performance in these tasks can result in significantly impaired quality of care—so clinical decision support is important. 

And the WHO broadly recommends the use of health worker decision supportbut it does so with a number of caveats. One of these relates to clinical decision support being too prescriptive and resulting in healthcare workers blindly following diagnostic or treatment algorithms and losing (or never gaining) the clinical competence needed to practice independently. This is a reasonable criticism and there is no clear answer to it. Ultimately clinical decision support should be “support”it should support the actions of the healthcare professional, but not force them to follow certain protocols.

The WHO recommendations also highlight feasibility issues with clinical decision support. These issues relate to “network connectivity, access to electricity, usability of the device, sustaining training and support to health workers using the digital tools.” Networks can be unstablebut the advent of clinical decision support on an app means that healthcare professionals can use clinical decision support without a connection. And providers of clinical decision support must certainly do more to ensure that they are usable. Importantly the recommendations state that clinical decision support should only be used “in the context of tasks that are already defined as within the scope of practice for these health workers.” This is significant as it is a clear warning that clinical decision support must not be misused and that healthcare workers should not be abandoned to clinical decision support tools and left to get on with things without adequate education and preparation. The direction of travel of clinical decision support is towards more algorithms and decision treeshowever these algorithms will only result in improved care if they are based on sound evidence. A clinical decision support resource may be based on sound evidencewhere evidence is availablebut sometimes there might be no evidence. In this circumstance the clinical decision support will have to revert to expert opinionjust as doctors do now. Finally, the WHO recommendations suggest that healthcare professionals should show the clinical decision support resources to their patientsso that they don’t become a barrier in the consultation.

This paper gives a summary of the main points in the WHO recommendations as they relate to clinical decision support. Most of the recommendations are sound—however you could argue that they don’t go far enough. For example, far from being a potential barrier in the consultation, clinical decision support resources could help doctors and patients come to shared decisions—by looking at the resources together. In our own clinical decision support resource—BMJ Best Practice—many of the topics contain patient information sheets that doctors can print out and give to their patients. This is appreciated by both healthcare professionals and patients. Importantly the paper makes considerable mention of costs—which can be considerable in this area. [1] The ideal clinical decision support tool would be low cost and deliver high value to patients and populations alike. 

Kieran Walsh is clinical director of BMJ Learning and BMJ Best Practice. He is responsible for the editorial quality of both products. He has worked in the past as a hospital doctor—specialising in care of the elderly medicine and neurology.

Competing interests: Kieran Walsh works for BMJ which produces the clinical decision support tool—BMJ Best Practice.


  1. Walsh K, Reeves S, Maloney S. Exploring issues of cost and value in professional and interprofessional education. J Interprof Care. 2014 Nov;28(6):493-4.