Figure 1.png (1)

RespiraSense: improving patient outcomes by #MakingEveryBreathCount

Tuesday, April 3, 2018

RespiraSense is a patient safety tool that improves quality of care by enhancing ability of medical teams to identify the deteriorating patient up to 12 hours earlier than the standard of care. It is supported by the NHS Innovation Accelerator (NIA).


We speak to Myles Murray, inventor of RespiraSense and founder and CEO of PMD Solutions, about what he coins the ‘Diagnostic Vital Sign’.

Hi Myles, welcome to Respiratory Futures. Can you start by telling us why this innovation is important?

In 2017, NHE England and NHS Improvement launched National Early Warning Score 2 (NEWS 2) following the first NEWS in 2012 by the RCP. The aim was to standardise practice, with the updated version (NEWS 2) being introduced as the standard for patient monitoring. It is to be rolled out across the NHS by 2019 and expected to be used in future CQC assessments of a hospitals quality of care [1].

NEWS 2 is based on a simple scoring system of six physiological parameters; a score is allocated to each parameter and an overall score is then aggregated. The purpose of NEWS2 is to detect the warning signs of a deteriorating patient at a much early stage, considerably improving the clinical outcome by allowing prompt and earlier intervention in the critically ill patient.

The central element of NEWS 2 is the more accurate measurement and timely monitoring of Respiratory Rate [3]. All the available evidence has shown that respiratory rate (RR) to be the single earliest sign of patient deterioration – most clinicians are well aware that changes in respiratory rate almost always precede changes in oxygen saturation and blood pressure – the usual prelude to acute sepsis, respiratory failure or cardiac arrest [4]. For example, one simple study looking at attitudes towards vital signs monitoring in the detection of clinical deterioration showed well that in the early stages of deterioration patients’ SpO2 may be in the normal range, but the RR will increase in response to inadequate tissue oxygenation [7]. As is well known in the respiratory community, increased RR compensates increased demand for oxygen until the divergence point is reached where the level of compensation can no longer meet the demand, as shown in Figure 1.

Heimlich valve
Figure 1

Abundant good evidence has shown that significant changes in RR occur earlier than changes in SpO2, heart rate, blood pressure and heart rate And yet up until now respiratory rate (in contrast to heart rate or SpO2) has proved a difficult variable to measure reliably outside of an ICU setting. At present, the delayed intervention when a patient deteriorates all too often occurs after the patient fatigues and is no longer able to maintain an elevated RR - the peri-arrest situation with a critically ill patient.

RespiraSense, for the first time, provides clinicians with an easy, low cost way of measuring respiratory rate continuously and accurately in the general ward setting. It means that medical teams could deploy timelier interventions by having perhaps as much as 6-12 hours earlier indication of deterioration than the present standard of care, with a 95% specificity. This would not only have an impact on improving outcomes (reduced escalation of care) but it would impact length of stay and unit cost of care – all the while improving both the quality and safety of in-hospital patient care.

So why measure respiratory rate (RR) continuously?

RespiraSense is a discrete wearable sensor that remotely monitors the patient from admittance to discharge throughout the hospital. By monitoring both the thoracic and abdominal movements, in addition to activity tracking, RespiraSense offers superior motion-tolerance for sensitive and specific alerting to abnormal breathing, thereby reducing burden of alarm fatigue and maximising the resources on at-risk patients. 

Where else is RespiraSense being trialled?

RespiraSense has been clinically validated in terms of accuracy and robustness for over 6 years. Thus, positioning it to become the next standard of care for respiratory monitoring.

  • Ventilator Bench testing, RespiraSense is +/-1 breath per minute (PMD)
  • Real-world Accuracy, RespiraSense is +/- 3 breath per minute (Kellett2014, Kellett2016, Lee2014, Subbe2017)
  • Motion-tolerance (Subbe2017)
  • Identifies patient deterioration <12hrs earlier than standard of care (TREND - Brabrand2017)
  • Reducing Length of Stay (Bradford Case Study – Smith2017)

Presently, PMD Solutions is seeking to collaborate with Respiratory and Acute Medical teams to address the NHS Priority of identifying the deteriorating patient and reducing preventable escalations in care. Leading to reduced average length of stays, freeing resource capacities, and improving patient flow.

How is it building on what we already know?

The standard of care has been manual counting for over 170 years, so you might wonder why this should change. Hutchinson in 1846 [figure 3] and 170 years later with Badawy et al in 2017, showed that respiratory rate is only manually measured in even numbers, with over 95% of measurements being the same 4 or 5 numbers (16, 18, 22, 24, 30) [10]. Semler et al (2013) found a significant difference between actual RR and the one that was recorded; out of 368 recordings, nearly 72% showed 18 or 20 breaths/min despite the fact that only 13% were actually within that range [11]. This bias can lead to missed early signs of patient deterioration and delayed activation of Early Warning Scoring protocols. This remains so even though all other vitals, e.g. heart rate, are objectively monitored and are normally distributed.

Heimlich valve
Figure 2

After all, respiratory rate is the earliest sign that something is amiss and teams need to diagnose the underlying cause – hence the term 'diagnostic vital sign'. 

Heimlich valve
Figure 3

Combine this with a simple discrete wearable device to continuously monitor respiratory rate in a motion-tolerant and robust way, and medical teams will have a powerful tool for identifying the very first stages of deterioration. With alerts easily verified with a blood chemistry analysis to inform, in a timely manner, the best course of treatment to ensure the best possible outcome for each patient.

Heimlich valve
Figure 4

What have clinicians said about RespiraSense?

"RespiraSense provides a unique way of continuously monitoring respiratory rate that will increase early diagnosis, improve patient safety and reduce pressure on NHS emergency and clinical care." Professor Anoop Chauhan, Portsmouth Hospital NHS Trust.

"By using the RespiraSense device and trusting the RR, we were able to move the patient to a general ward room one day sooner than we would have without the use of the device. This is due to the accuracy of the device and the confidence we had in it to be a more clear and accurate predictor of deterioration." Dr Peter Smith, Bradford Royal Infirmary.

What’s next?

NHS England has approved the National Early Warning Scoring System 2 for nationwide use by April 2019. The respiratory rate remains a vital component of this. In addition, the focus on improving patient flow and reducing healthcare costs is driven by NHS priorities such as 'The Deteriorating Patient'.

In 2018, PMD will undertake a Stepped Wedge Randomised Control Trial to evaluate the level of clinical and economic impact a continuous and objective RR monitoring will have in the general ward.

In addition, RespiraSense is anticipated to make a NICE Medtech Innovation Brief submission in early 2018 to support its claims.

Thanks Myles. 

Find out more about RespiraSense on the NHS Innovation Accelerator website: 


  4. Jonsson T et al (2011) Nursing documentation prior to emergency admissions to the intensive care unit. Nursing in Critical Care; 16: 4, 164-169.
  5. Hodgkin JE, Soeprono FF, Chan DM. Incidence of metabolic alkalemia in hospitalized patients. Crit Care Med. 1980;8:725-728.
  6. Mazzara JT, Ayres SM, Grace WJ. Extreme hypocapnia in the critically ill patient. Am J Med. 1974;56:450-456.
  7. Mok W et al (2015) Attitudes towards vital signs monitoring in the detection of clinical deterioration: scale development and survey of ward nurses. International Journal of Quality in Health Care; 27: 3, 207-213.
  8. Michelle A Cretikos, Rinaldo Bellomo, Ken Hillman, Jack Chen, Simon Finfer and Arthas Flabouris. Med J Aust 2008; 188 (11): 657-659. 
  9. Badawy et al (2017) Is everyone really breathing 20 times a minute? Assessing epidemiology and variation in recorded respiratory rate in hospitalised adults. 2017 Oct;26(10):832-836. doi: 10.1136/bmjqs-2017-006671. Epub 2017 Jun 26.
  10. Semler M et al (2013) Flash mob research: a single-day, multicenter, resident-directed study of respiratory rate. Chest; 143: 6, 1740-4.