Track infection spread with a smartphone

Until recently it has been almost impossible to accurately predict the spread of an airborne infection. We at Edinburgh Napier University have been working with Imperial College London using smart phone location tracking technology to address this challenge. The resulting model will help provide important clues about how quickly a pandemic might occur by recording the nature and frequency of interactions between individuals.

Tracking infected persons
With location-tracking technology in handheld devices, such as smartphones, it is now possible to track the path of an infected person and their contact with others with one metre accuracy. Wireless technologies such as Wi-Fi and Bluetooth, mobile phone network assisted GPS and radio frequency identification tags (RFID tags) as well as ultra-wideband are key factors in allowing us to develop this infection spread model.

The technology helps overcome the challenge of tracing the original infection source from within a large geographical area. For instance, patient turnover in hospitals is so high that it is often difficult to establish where and when patients become infected, allowing a contaminated area to continue to spread the infection.

A global issue
Healthcare Associated Infections (HAI) is a global problem as highlighted by The Medical Ward of the 21st Century from the University of Calgary. The annual cost of HAI is estimated to be around $30 billion across the whole of the US alone.

There is increasing pressure from governments to adopt RFID technology to improve patient safety. The e-Health agenda added to the increasing pressure from Connecting for Health and the National Patient Safety Agency (NPSA), plus the rise of infections such as MRSA and Clostridium difficile (C.Diff) shows that there is also an increase in pressure in the UK to improve efficiency and patient safety.

Prioritising
We aim to create a system using this tracking technology to identify, track and audit factors that could cause infections such as MRSA in a healthcare environment. The model we are proposing will potentially allow emergency health providers to prioritise who may have come into contact with an individual exposed to a serious airborne illness, such as influenza during an outbreak.

At present most companies in the infection control sector focus on disinfectant solutions. This is important but does not give the full auditable solution compared to RFID tagging. It may also lead to resistant strains. RFID technology has also been used to monitor employees and patients washing their hands but again this does not give a complete picture of the whole of the healthcare infrastructure.

Identifying risks
In order to prevent infection and reduce mortality it is vital to identify avoidable risks at an early stage by tracking down contaminated areas within the hospital and implement changes to the existing practices wherever possible. Various epidemic models have been used in the past for risk analysis related to MRSA, but their application is not easy, results must not be expected in a timely manner and data collection is one of the biggest challenges, as staff are already burdened with a huge variety of administrative tasks and documentation.

Often responses to outbreaks of dangerous infections in hospitals, though, are not based on scientific analysis but are more undirected and based on worst case scenarios. This usually leads to ward closures and results in a loss of functionality of the hospital for days or sometimes even weeks.

It’s worth noting, however, that some risk factors are related to an individual’s condition such as immune competence, general physical condition and past medical history. In fact the typical infection pattern for hospital acquired C.Diff, for example, is that patients pick up the bacteria from the hospital environment, in particular from surfaces, which then manifest an attack within the patient’s body. This is why outbreaks of certain infections do not occur everywhere in the hospital, but are usually related to a particular unit or procedure.

The key Healthcare Associated Infections (HAI) studied in our research were Urinary Track Infections Surgical Site Infections, which account for 16 per cent of all HAI infection for surgical patients, as well as Pneumonia and Bloodstream Infections.

Location data
The new software will record and playback location data with high-precision. It uses Susceptible Infectious Recovered (SIR) modelling and the epidemiological technique of contact tracing to predict the spread of a disease through a network of people, taking account of parameters such as transmission and recovery rates.

Our experiments show that the technology gives location readings that are sufficiently accurate to monitor the movement and of individuals and their contact with others. This will provide important clues about how quickly a pandemic might occur.

The tracking has been used with a lab environment, allowing for a deep understanding on how health care entities might interact and thus how infections could spread.

For more information:
Web: www.iidi.napier.ac.uk