Ahmed Alboksmaty, Paul Aylin, Thomas Beaney, Jonathan Benger, Jonathan Clarke, Ara Darzi, Sarah Elkin, Kelsey Flott, Aidan Fowler, Ana Luisa Neves
Centre for Mathematics of Precision Healthcare, Institute of Global Health Innovation - Imperial College London
The Wellcome Trust, National Institute for Health Research
At the start of the Covid-19 pandemic, clinicians noticed that patients with Covid-19 infection could have dangerously low levels of oxygen in their blood without feeling particularly breathless. This phenomenon was called ‘silent hypoxia’ and prompted health systems across the world to use small finger-worn medical devices known as pulse oximeters to detect the blood oxygen levels of patients to monitor the severity of their disease.
While these devices have been used for decades in hospitals and by clinicians in the community, their use by patients to monitor their own condition was relatively new.
From the first months of the Covid-19 pandemic in the UK, several local health services began to offer this form of remote monitoring to patients, in particular those at increased risk of severe disease due to their age or their other medical conditions.
The Institute of Global Health Innovation examined data from four of these services in 2020 and found these services to be generally safe. However, we also found extensive variation in how the services were implemented and who was offered the service. (Clarke et al. 2021)
In November 2020, the National Health Service in England launched the Covid Oximetry @home service (CO@h) aiming to provide home monitoring for patients with Covid-19 infection using a pulse oximeter.
We were asked to evaluate the program to identify whether it was associated with changes to the risk of death from Covid-19 or the risk of being admitted to hospital. We were also asked to investigate whether there were particular groups of patients who were more likely to receive care through this pathway.
The most important part of this project was collecting the right data from a range of sources to give us the most complete picture possible for our analysis. Working closely with NHS Digital we developed a dataset containing data from primary care, hospitals and the Office for National Statistics. This allowed us to understand when people had a positive Covid-19 test, what their other clinical conditions were and if they went on to got to hospital or to die within 28 days of their positive Covid-19 test.
Our evaluation was built around three different studies:
A study looking at how mortality and hospital attendance changed after the evaluation started for the whole population.
A study looking at how mortality and hospital attendance changed after the evaluation started for people who attended A&E around the time of their positive Covid-19 test.
A study looking at who was enrolled onto the program after having a positive Covid-19 test.
For more details on how we carried out each study, please see the Outputs.
We wanted to understand the impact of the CO@h programme in as broad a way as possible. A single study wouldn’t have given us the scope to look at the range of things we felt were important.
[TO BE FINISHED]
This work was mostly conducted to help the NHS understand how their CO@h service had been implemented and to guide its future use.
Over the course of the project we learned many important things about how new digital care pathways are implemented in the NHS. Lots of these findings may be applied to other pathways using pulse oximeters (for example in the care of patients with chronic respiratory diseases like COPD), or could be applied to other medical devices such a blood pressure monitors or personal wearable devices like fitness trackers and smart watches.
We also learned that digital pathways may be more likely to be offered to some patients than others. While in we can’t comment on whether this was a clinically appropriate decision or not, we feel it is important that new digital care pathways do not widen existing health inequalities or introduce new ones. It’s therefore crucial that researchers investigate how future digital care pathways can be implemented in a way that is fair to all patients.
This work has been published as three preprint articles that are currently undergoing peer-review:
Population level impact of a pulse oximetry remote monitoring programme on mortality and healthcare utilisation in the people with covid-19 in England: a national analysis using a stepped wedge design. (2021) T Beaney, J Clarke, A Alboksmaty, K Flott, A Fowler, JR Benger, P Aylin, S Elkin, AL Neves, A Darzi. medRxiv
Trends and associated factors for Covid-19 hospitalisation and fatality risk in 2.3 million adults in England. (2021) T Beaney, AL Neves, A Alboksmaty, K Flott, A Fowler, JR Benger, P Aylin, S Elkin, A Darzi, J Clarke. medRxiv
Evaluating the impact of a pulse oximetry remote monitoring programme on mortality and healthcare utilisation in patients with covid-19 assessed in Accident and Emergency departments in England: a retrospective matched cohort study. (2021) T Beaney, J Clarke, A Alboksmaty, K Flott, A Fowler, JR Benger, P Aylin, S Elkin, A Darzi, AL Neves. medRxiv