In the context of healthcare, data along with artificial intelligence are not merely options any longer; they are turning into the main factors that drive decisions, patient experience and outcomes. The only real problem, however, is not just the collection of a huge amount of data or the implementation of numerous AI tools. The problem is to employ them in such a manner that the human aspect is the one that predominates. This is basically the message that can be found in an article in Digital First Magazine that presents the point of view of Christopher Hutchins, founder of Hutchins Data Strategy Consultants.

Hutchins represents the change as a move “from data collection to data meaning.” Some healthcare organisations have gathered enough data to build enormous databases comprising clinical, operational and patient-experience data. But if they do not work out how to extract actionable insights, those that help clinicians and engage patients, the promise will only be a fraction fulfilled.

In this regard, AI is not about taking over the job of human providers. It is about allowing them to do more. To illustrate, AI algorithms can identify patterns overlooked by humans. However, if the process becomes too technical or impersonal, patients at the same time may get alienated. Hutchins is very much keen on “human connection” and considers it a main point not to be thrown into the background.

The article points to three pragmatic focus areas:

Firstly, data must be clean, well-governed and accessible for AI tools which are dependable to be built on such data.

Secondly, AI tools should be part of clinical workflows in such a way as to facilitate decision-making without disturbing it.

Thirdly, perpetuating transparency and patient-centric design so that technology is a means to an end, that is, the provider-patient relationship, rather than a replacement for it.

Practically speaking, this means that if an AI-driven tool predicting readmissions is launched by a hospital but the staff are not able to comprehend the output or patients cannot make sense of its influences, the tool will be of no use. But, if clinicians communicate the findings to patients and get their feedback then results will be better.

This movement is supported by the study which suggests AI used in healthcare should be explainable and based on trust. When a black-box-type model is heavily relied upon, things can turn out to be the opposite of what one expects. Besides the ethical side, the impracticality of the situation has to be taken into account as well.

To sum up: Healthcare industry is at a turning point. Data and AI tools have a great potential, however, that potential would only be realised if the “human” aspect stays in the spotlight. Technology has to be a means of caring, not a way of supplanting it. Those organisations which will find the right balance between smart tools and the meaningful human connection will probably be the ones to set the ‍‌‍‍‌‍‌‍‍‌pace.