A person testing their blood glucose with a handheld meter

MyWay Digital Health Project

Decision support for inpatient diabetes care

A prototype app from MyWay Digital Health that brings together inpatient, outpatient and point-of-care data to help ward staff spot people with diabetes at risk of harm and make better treatment decisions.

Context

Almost 10% of the global population has diabetes and rising. In the UK, 15-17% of all hospital (inpatient) beds are occupied by people with diabetes at any time. Diabetes inpatient care consumes around £2.5 billion per year.

Problem

There are significant numbers suffering a deterioration in their diabetes care, such as high or low glucose events or foot ulcers, because of poor management. Non-Diabetes hospital staff are poorly trained in diabetes management and there is limited capacity with specialist diabetes teams. Consequently, staff training and tools to support clinical decision making are vital to improve patient outcomes.

Solution

To develop a prototype digital tool app to help clinicians make correct decisions and more easily identify individuals at risk of poor outcomes during an inpatient stay. This includes supporting triage of patients coming into hospital, flagging those needing specialist review, prompting staff to assess and protect feet, advising around glucose target and treatment changes, such as insulin prescription and supporting the right decision to admit or discharge a patient. This tool uses information (data) from medical records and from connected hospital glucose meters.

A screen from the MyDiabetes Clinical app. Question 6 asks whether any of the following apply: any concern about critical condition; unlikely or unable to follow advice; unable to self care or no support; unlikely to attend the follow up; tomorrow is Saturday, Sunday or a public holiday. Yes is selected and the app advises Admit.

Evaluation

A prototype of the decision support app for an inpatient setting was developed and tested in a simulated environment provided by MWDH.

A series of 4 workshops with 11 primary and secondary health care practitioners were organised using prototype screens to discuss protocols, research and resources identified in the first phase of development and a post-workshop questionnaire.

Also, 4 testers were asked to evaluate a set of clinical scenarios and comment on the system outputs.

Results

The project developed prediction models by looking for patterns in historical data to develop algorithms that can predict who is most likely to come to harm. Also developed, were rules and pathways coded into the app using clinical consensus linking to agreed protocols and guidelines to deliver advice and support to clinicians throughout the hospital at the point of need.

Project at a glance

REGION

NHS Lothian

STATUS

Completed

CLINICAL LEAD

Dr Stuart Ritchie

CLINICAL SPECIALITY

Diabetes

Create a free website with Framer, the website builder loved by startups, designers and agencies.