Advice

The technology

The technology

SYNE‑COV (Sensyne Health) is a cloud-based software. It uses a machine-learning algorithm to analyse clinical data in electronic health records to estimate the chance of health deteriorating in a person with COVID‑19. This is measured by 3 clinical outcomes:

  • having invasive mechanical ventilation

  • being admitted for intensive care

  • the risk of people with COVID‑19 dying in hospital.

SYNE‑COV uses clinical data including demographics, vital signs, blood biochemistry and respiratory physiology to provide a real-time risk estimation. This helps clinician manging patients' symptoms and to help make clinical decisions about treatment. Each risk estimation ranges from 0% to 100%. The lower the score the less likelihood an outcome occurs.

The company states that the algorithm is a supporting tool and its risk prediction should be used along with other relevant patient data for clinical decision making.

Innovations

SYNE‑COV is developed as a clinical algorithm for predicting clinical outcomes in people with COVID‑19 who are admitted to hospital.

In the UK, there have been methods developed for predicting COVID‑19 related outcomes such as the OpenSAFELY platform (Williamson et al. 2021). Soltan et al. (2021) used linear and non-linear machine-learning classifiers to predict hospital admission in people presenting to hospital, comparing prediction between people with and without COVID‑19. The company notes that there is no commercial, UK Conformity Assessed marked technology similar to SYNE‑COV.

Current care pathway

NICE has published a guideline on managing COVID‑19. There is no recommendation on predicting the risk of condition progression for people admitted to hospitals. The guideline recommends ensuring healthcare professionals have access to resources to support discussions about treatment plans. Hospitals should review management for people who are deteriorating, and use the track-and-trigger system (National Early Warning Score 2 [NEWS2]) for managing COVID‑19 in hospitals. The recommendation is in line with the guidance by the Royal College of Physicians (RCP), which emphasises using NEWS2 in managing COVID‑19. The RCP guidance states that using NEWS2 will ensure that people who are deteriorating, or at risk of deteriorating, have a timely initial assessment by a competent clinical decision maker. NEWS2 should supplement clinical judgement in assessing the patient's condition.

The NICE guideline on acutely ill adults in hospital: recognising and responding to deterioration gives best practice advice on the care of adult patients admitted to hospital. It recommends that adult patients in acute hospital settings should have physiological observations recorded at the time of their admission or initial assessment. After the admission, these measurements for physiological signs should be regularly monitored regularly with predetermined response criteria to changes to identify patients whose physiological status is deteriorating or is at risk of deterioration. NICE recommends using a tool (NEWS) based on an aggregate score made up of the measures of respiratory rate, oxygen saturations, blood pressure, pulse rate, levels of consciousness and temperature to improve the detection and response to patients' physiological deterioration.

The following publications have been identified as relevant to this care pathway:

Population, setting and intended user

SYNE‑COV is intended to help clinical management for people aged between 18 and 100 years with COVID-19 being assessed in hospitals.

The company states that SYNE‑COV should be used by senior clinicians, for instance healthcare professionals who are at consultant level in acute or secondary care settings.

Costs

Technology costs

The cost of SYNE‑COV ranges from £18,000 to £30,000 per year depending on the number of beds and patients in each NHS trust.

Costs of standard care

National track-and-trigger systems can be automated or paper based. Paper-based early warning score (EWS) charts are available for free to download from the Royal College of Physicians website.

A study reported that, on average, it took 3 minutes 35 seconds nursing time to do manual vital signs observations and EWS calculations (Wong et al. 2017). Current hourly pay of a band 5 nurse ranges from £12.74 to £15.66 depending on the length of service. Wong et al. (2017) also reported that the using automated EWS systems reduced nursing time to 2 minutes 30 seconds. The cost of the automated system ranges from £30,000 to £90,000 for system installation, configuration and set up, and costs a further £0.35 to £0.70 per acute bed, every day.

Resource consequences

The technology was developed in collaboration with 1 NHS hospital trust, and has not yet been widely used in the NHS.

SYNE‑COV is intended to be used as a supporting tool along with current EWS for monitoring deterioration in COVID‑19. As an add-on intervention, the resource impact of using SYNE‑COV would result in additional costs to standard care. However, it has the potential to be resource releasing if using the results to improve COVID‑19 management that could reduce hospital stay and mortality. There is currently no evidence to support these claims.

No changes to facilities or infrastructure are needed to adopt the technology because SYNE‑COV is compatible with existing NHS IT infrastructure. It can be integrated into existing hospital systems and dashboards with no need for a separate application to view the risk prediction. The company provides training on how to use and interpret the results given by the predictor, and the training cost is included in the cost of the technology.