Nightingale PCP #727534 – ‘Common Challenge’
European Commission Horizon 2020 Pre-commercial
Procurement: Project number 727534
www.nightingale-h2020.eu
Five Academic Hospitals seek innovative partners to deliver cutting
edge health care solutions for wireless monitoring and
identification of high-risk patients, both in hospital and at home.
Executive Summary
Clinical need
Patients die because signs of deterioration are missed. There is a huge unfulfilled
need for better monitoring of vital signs to identify high-risk patients who are on
general hospital wards or at home. Patient deterioration is often overlooked or not
detected at all. One of the reasons is the intensity in nursing and frequency of vital
signs monitoring which decreases from the Intensive Care via ward towards home.
Early detection of vital instability is crucial to prevent death and disability.
Pre-commercial Procurement (PCP)
Five leading European academic hospitals (Utrecht, the Netherlands; Stockholm,
Sweden; London, United Kingdom; Leuven, Belgium and Aachen, Germany) uses
the European commission’s Pre-commercial Procurement (PCP) funding scheme to
challenge and stimulate European industry to develop a system to connect patients
and carers to wirelessly monitor patients’ vital signs and identify high risk individuals.
The available budget for the development of this innovative solution is 5.3 million
euro, consisting of 3.75 million euro in the form of innovation subsidies to the
industry.
Finding a solution
The solution should consist of one or more unobtrusive wireless sensors that do not
interfere with the patient’s daily activities or rehabilitation. Intelligent analysis
software is a key feature of the requested system, as the false alarm rate must be
extremely low to be acceptable to the users. It should be well-integrated into the
hospital’s electronic medical record system. The sensors should involve a
mechanism that allows patients and informal carers to communicate with their
healthcare team, and allow entry of qualitative data (e.g., pain and well-being
scores).
We expect that successful implementation of such a system will empower patients
and carers. It will have the potential to transform healthcare by reducing death and
disability from undetected deterioration, and can provide a ‘safety net’ to high-risk
patients after discharge home.
We invite the industry to solve this huge unfilled need for better patient monitoring!
1
Nightingale PCP #727534 – ‘Common Challenge’
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No727534
Nightingale - Connecting Patients and Carers using wireless technology:
The Common Challenge
Contents
2.
The Clinical Need
3.
The challenge – need for open market consultation
4.
The Business Case
5.
Functional specifications for a system to wirelessly
connect patients, carers and the healthcare team
9.
Levels of sophistication with clinical decision support
based on intelligent alarms
10. Functional communication specifications
11. A brief introduction to Pre-commercial Procurement (PCP)
15. Public Procurement of Innovation (PPI)
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No727534
2
Nightingale PCP #727534 – ‘Common Challenge’
Nightingale - Connecting Patients and Carers using wireless technology:
The Common Challenge
The Clinical Need
Despite improvements in diagnostic and therapeutic options for in the last decades,
there is an increasingly important ‘Achilles heel’ in the care of patients who become
acutely ill, who recently underwent surgery, those who return to a general ward in the
hospital after discharge from a Critical Care environment, or those receiving
continuing care at home.
Early detection of deterioration is crucial to prevent death and disability from
rapidly fatal conditions such as cardiac arrest, respiratory arrest, upper airway
obstruction, internal bleeding and sepsis. Such monitoring is primarily a nursing
task. However, nurses on hospital wards must care for increasingly unwell patients;
on many wards it is only feasible to perform one set of vital sign observations once
per nursing shift (8 hr). Increasing the intensity of patient observation and vital sign
monitoring by more frequent nurse rounds is severely limited by the availability of
sufficient numbers of trained nurses and budget constraints.
Once discharged home, patients’ vital signs are no longer monitored at all. While this
may seem acceptable for patients who are stable and nearly completely recovered
from their illness, the reality today is different. A sizeable proportion of mortality
occurs in the first week after discharge; on the other hand, patients are now
discharged home earlier than ever before.
While enhanced early recovery after surgery is associated with better outcomes,
early discharge also means that some major complications will become first
manifested in the home setting.
A system that connects the high-risk patient at home with the healthcare team, and
generates alerts when vital signs deteriorate, will allow early recognition and early
treatment, with rapid rehospitalisation where needed.
Market search (as of November 15th 2016)
The consortium partners have performed an extensive search to discover possible
solutions that are currently available. We identified several ‘wearable’ devices that
are designed to measure one or more vital signs and transmit these data wirelessly
to another device such as a smartphone or network-attached ‘bridge’ (and from there
to a server and the patient record).
Currently the majority of wearable sensor devices appear to be geared towards
lifestyle and sports. We identified only a handful of companies that produce a
‘medical grade’ wearable sensor that can measure and wirelessly transmit vital signs
such as heart rate, respiratory rate and temperature. None of these companies
currently has a user front end for the care giver that intelligently interprets the various
signals to avoid repeated triggering of false alarms. Most systems that present data
on a central terminal use traditional user-selectable alarm ‘threshold’ functionality.
None of the currently available systems is sufficiently developed to allow remote
monitoring of patients’ vital signs in the home situation.
Nightingale - Connecting Patients and Carers using wireless technology:
The Common Challenge
3
Nightingale PCP #727534 – ‘Common Challenge’
The challenge – need for open market consultation
Today, even the last generation of conventional ‘wired’ vital signs monitors used in
hospitals are primitive, static, bulky and have alarm systems that generate false
alarms over 90% of the time. The consortium partners believe that it is
technologically feasible to design a system that meets our advanced specifications.
We actively seek a dialog with industry to refine these preliminary
specifications.
This dialog within an open market consultation will help us to produce an ambitious
but realistic set of tender specifications in the second half of 2017.
We realize that a successful solution design may need intense collaboration between
different companies and might require integrating various types of knowledge
currently dispersed over wide areas (eg. aerospace, climate science, weather
prediction, self-driving cars, etc.). While advanced electronics companies and the
medical device industry appear to have found each other, integration with companies
that master advanced software tools such as machine learning and artificial
intelligence is still developing. Finally, such companies need input from medical
experts and access to patients to test and fine-tune their solutions.
4
Nightingale PCP #727534 – ‘Common Challenge’
Nightingale - Connecting Patients and Carers using wireless technology:
The Common Challenge
The Business Case
The business case for our requested solution is very strong. Avoidable adverse
events (including death) are extremely costly; for example, most major surgical
complications result in excessive Length of Stay and added costs for additional
drugs and interventions (total cost up to more than twice the cost of admission
without complications).
Moreover, patients with complications have a higher likelihood of being readmitted to
hospital after discharge home. For hospitals and third-party payers, unplanned
readmissions are an increasing cause for concern. Even complications that cannot
be avoided by improved patient monitoring will benefit from rapid recognition and
timely treatment.
The partners have initiated formal Health Technology Assessment (HTA) studies to
help define the business case and cost effectiveness ratio of the envisioned solution.
5
Nightingale PCP #727534 – ‘Common Challenge’
Nightingale - Connecting Patients and Carers using wireless technology:
The Common Challenge
Functional specifications for a system to wirelessly
connect patients, carers and the healthcare team
The system will consist of four building blocks, creating all together the solution for
safe patient monitoring.
Figure 1. Functional components of Nightingale
Sensing system

The ideal multi-parameter sensor is small, light-weight and can be worn on an
area of the body without the patient being aware of its presence

The sensor can be continuously used for uninterrupted vital signs monitoring
for at least 5 days
Please note: for practical reasons, we define ‘continuous’ here as: at least one
full set of vital signs measurements every 2 min)
o this requires that it either uses very little battery power for its sensing
functions and data transmission,
or:
o that it can extract its required energy from the patient

robustness: the sensor ideally can withstand a daily shower. Preferably the
sensor can also remain attached during imaging studies (with MRI studies as
a possible unavoidable exception)
6
Nightingale PCP #727534 – ‘Common Challenge’

the sensor measures at a minimum the following basic parameters, most of
which are already available in some medical grade wireless sensors:
o heart rate
o respiratory rate
o temperature
o motion and (3-axis) position

a sensor that can also measure oxygen saturation is highly desirable
o the consortium partners are aware of the fairly high energy
requirements of the current generation pulse oximeters
o intermittent readings of oxygen saturation to preserve battery life (at
fixed intervals or triggered by abnormalities in any continuous
parameter) might be acceptable

a sensor that gives an approximation of the patient’s current global
hemodynamic status, in particular blood pressure or pulse pressure is highly
desirable. This does not need to be a conventional arm or finger cuff-based
‘exact’ blood pressure measurement, but must be able to reliably track
changes over time in perfusion pressure.

Other sensor measuring stress, distress, relevant blood parameters (lactate,
pH, other) are potentially be very valuable.
Data transmission protocol and interference rejection

the consortium expects safe and reliable data transmission with a minimum
of dropouts, but does not wish to prescribe any specific protocols for data
transmission
o the transmission protocol can therefore be either an industry standard
(e.g., WiFi, Bluetooth LE, etc.) or a proprietary protocol
o however, ‘open standards’ are highly desirable from an interoperability
perspective
o any sensor and integrated wireless signal transmission system must be
compliant with all applicable safety regulations, both regarding patient
harm (electrical safety, EM radiation) and potentially unsafe RF
interference causing malfunction of other medical devices

Data transmission needs to be compliant with HL7 international standards
for transfer of clinical data.

The sensor (whether or not via a relay device) is able to cover an entire
home; e.g., the sensor worn by the patient can transmit from any room in
the patient’s home without the necessity to be in the vicinity of a relay
device.

Wireless data security requires the use of robust encryption, in particular
while wireless vital signs data are in transit

patient acceptance of the system will be highly dependent on the
convenience of using the entire system.
7
Nightingale PCP #727534 – ‘Common Challenge’
o for example: a system using a number of fixed ‘wired’ receivers
(bridges) in many rooms to connect the patient to the health care worker
over the network will have dead zones in some areas of the hospital.
o using such a system for home monitoring means that the patient
effectively has ‘house arrest’.

when there are two wireless transmission protocols active at the same time
to sequentially and wirelessly relay the vital signs data streams (first from
sensor to mobile device and then from mobile device to a hospital server),
this might theoretically increase the likelihood of data loss caused by
electromagnetic interference or other network issues.
Artefact rejection system

a high quality artefact rejection system is necessary to prevent false alarms,
‘alarm fatigue’ in caregivers (from the need to check the patient repeatedly on
spurious readings) and unnecessary anxiety in patients

many artefacts in vital signs monitoring are caused by motion artefact. In
contrast to anaesthetised patients and patients on intensive care units, most
patients in the targeted groups are mobile and may be actively exercising
several times a day.

The consortium partners suggest active monitoring of patient motion. Not only
will this greatly facilitate flagging of motion artefacts, we also believe that
motion itself (or lack thereof) might prove to be an important vital sign that is
worthy of tracking alongside the more traditional vital signs
Analysis system
A sophisticated analysis and alarm generation system is paramount. Unlike
operating rooms and intensive care units where ‘one on one’ nursing is the rule, on
general hospital wards many patients will be simultaneously monitored. In case of
abnormal signals, the nurse can immediately check the patient to see if there is any
real deterioration of vital status or that the alarm was caused by technical problems
with the sensor or data transmission. In contrast, when telemonitoring is applied in
the home care situation there is no easy way to perform such ‘cross-checks’ for
sensor dislodgment, strange movements, etc.
Therefore we request that the solution design contains:

intelligent analysis of the incoming vital signs data streams

analysis of trends and rates of change

pattern analysis in multiple simultaneous data streams
8
Nightingale PCP #727534 – ‘Common Challenge’

generation of probabilistic alarms
The consortium partners realise that this might require the use of machine learning
and artificial intelligence. There are several possible levels of sophistication with
clinical decision support based on such intelligent alarms.
9
Nightingale PCP #727534 – ‘Common Challenge’
Nightingale - Connecting Patients and Carers using wireless technology:
The Common Challenge
Levels of sophistication with clinical decision support
based on intelligent alarms
Different levels of sophistication can be distinguished. Starting with the early warning
systems as we have today ending with a level based on artificial intelligence that is
self-learning and self-optimising.
Nightingale is aiming for the best solution possible, with at least level 3.
Figure 2 Levels of sophistication
10
Nightingale PCP #727534 – ‘Common Challenge’
Level 0:
No artificial intelligence, simple threshold alarms only
(the user must set the lower limit and the upper limit for each measured variable).
For example: <alarm> IF (heart rate > 100/min OR heart rate < 40/min).
A slightly more sophisticated alternative simple alarm scenario: < ‘orange’ alarm> IF
(heart rate > 90/min OR heart rate < 50/min); <‘red’ alarm> IF (heart rate > 110/min
OR heart rate < 40/min). Or a ‘no data’ alert, when the patient moves to far from a
relay device.
Such systems may also attempt to avoid triggering of false alarms by applying
median filtering (or some other method of averaging the signal over longer periods of
time) to remove brief movement-induced artefact transients or by simply delaying the
alarm for a predefined time period, as currently used in some monitor systems.
Level 1:
As described above for ‘level 0’, but with simple integration of multiple
data sources: for example an ‘automated early warning score’. Early warning scores
are simple sum scores calculated by adding points for each vital sign that is ‘out of
range’. For example: the national EWS (NEWS, United Kingdom)
Figure 3 Physiological parameters
Level 2:
An advanced system to track vital signs that does not only track
changes in single parameters, but also considers the relationship between changes
in various vital signs. For example, a combination of progressive tachycardia (high
heart rate, increasing over time), progressive tachypnea (high respiratory rate,
increasing over time) combined with fever, suggests early sepsis which may
necessitate measurement of blood lactate and intensive care expert consultation.
Alternatively: a pattern of reduced motion, decreasing respiratory rate followed by
progressive hypoxemia (low blood oxygen saturation) in a patient who recently had
surgery and is receiving opioid pain medication suggests the possibility of opioid
overdose. The latter constellation requires the nurse to check on the patient
immediately and discuss the need for an opioid antagonist such a naloxone with the
attending doctor.
11
Nightingale PCP #727534 – ‘Common Challenge’
Level 3:
A clinical decision support system that incorporates more than only the
intelligent analysis of vital signs data. Such a system may also take existing or new
abnormal laboratory results into account (e.g., a new observation of hyperkalaemia
(high blood potassium level) that becomes available in the laboratory tab of the
EMR). A level 3 system may also take environmental information and important
diagnostic information into account, including structured nursing observations such
as restlessness, sleepiness/sedation, dyspnea, sweating, etc. In the home setting,
such information can be entered directly by the patient or informal carer.
Level 4:
A fully-integrated hospital-wide clinical decision support system that
monitors the patient’s status in real-time and constantly updates the risk of death and
severe complications when a new data element becomes available. It may be even
optimized to detect several specific rare but lethal conditions. For example, a lifethreatening disturbance of ventricular rhythm (‘torsade des pointes’ caused by ‘long
QT-syndrome’) as a result of inadvertent drug-drug interactions. Such a highly
sophisticated system takes as its inputs clinical data (current diagnoses, recent
procedures, prior history, nurse observations (including a sense of ‘worry’ regarding
the condition of the patient), doctors observations, data entered by the patient or
informal carer, on-line continuous vital signs data, laboratory data, new imaging
results, and updates its predictions every time a new data element reaches the
system.
Level 5: an artificial intelligence integrated self-optimising system. The most obvious
application of artificial intelligence in healthcare is data management. Nightingale is
collecting many data, storing it, normalizing it, tracing its lineage – it is the first step
in revolutionizing the existing healthcare system. With artificial integrated in
Nightingale it should be possible to optimise the dedicated patient specific decision
support.
The consortium partners require a solution with at least ‘level 3’
sophistication.
12
Nightingale PCP #727534 – ‘Common Challenge’
Nightingale - Connecting Patients and Carers using wireless technology:
The Common Challenge
Functional communication specifications
Caregiver notification system
A critical component of the system is a robust and acceptable caregiver notification
system. We envision that a device to communicate caregiver notifications:





is portable, lightweight, easy to clean/disinfect (waterproof)
has the option to display ‘actionable’ recommendations
can display interactive patient video
that requirements may be met either with dedicated hardware or by apps on
standard mobile platforms
the consortium is open to suggestions for alternative or parallel modes of
alerting caregivers, in particular ‘hands-free’ solutions. Similarly, in the home
setting, informal carers may need to be alerted once potentially dangerous
deterioration occurs.
Bi-directional communication system
In the hospital a nurse can personally check signs of deterioration. This is, however,
not an option when the system is employed in the home after discharge from the
hospital. In such cases, it is crucial that as soon as an alarm is generated the
caregiver can initiate communication with the patient to verify the presence of any
symptoms or signs that - together with the abnormal vital signs - might give reason
for concern and to consider the need for rapid medical attention. Alternatively, if the
alarm condition was the result of an artefact, the caregiver can reassure the patient
and/or informal carer, and assist them to get reliable vital signs monitoring back
online, for example by suggesting to reposition or replace a sensor.
Such a system for bi-directional communications should be:

simple and cheap
o this may require using existing video calling technology, using the frontfacing camera available in the latest generation smartphones

reliable

well-integrated with the vital signs relaying system

be connected both with the patient AND the informal carer(s) who live in the
patient’s home or is nearby. This is important in acute situations with severe
deterioration in case the patient is unable to respond to the caregiver’s (video)
call request.
13
Nightingale PCP #727534 – ‘Common Challenge’
Nightingale - Connecting Patients and Carers using wireless technology:
The Common Challenge
A brief introduction to Pre-commercial Procurement (PCP)
Several years ago the European Commission designed a new scheme to stimulate
European Innovation, called ‘Pre-Commercial Procurement’ (of Innovative
Solutions), for situations where there is currently no existing solution in the market.
The European Commission funded Nightingale project is an implementation of PreCommercial Procurement (PCP). Rather than directly subsidising European industry,
the prospective users of the new technology assist in guiding the development
process using a ‘funnel-shaped’ procurement model. This innovative process
presents an opportunity for suppliers to develop new solutions and to introduce new
technologies and products, in direct cooperation with the healthcare institutes and
their employees. The PCP process is exempted from the EU procurement directives,
is flexible and allows an early dialogue between potential suppliers and the
healthcare institutes in order to optimize the solution. Intellectual Property Rights
(IPR) of the developed solution will be shared between the Nightingale procurers and
the supplier. The development work will be co-financed by Nightingale project
funding, within the Horizon 2020 programme of the European Union.
The PCP process will start with an extensive preparation phase (‘Phase zero’; below
leftmost panel) and is followed by three development phases, as visualised below:
14
Nightingale PCP #727534 – ‘Common Challenge’
Phase 0 - The preparation of the Nightingale PCP starts with the publication of a
Public Information Notice (PIN) and then enters an Open Market Consultation, which
will consist of the following sequential components: I) Market Sounding, II) Market
Sounding Review, and III) Market Consultation. In the Market Sounding, through
different channels, the Nightingale procurers will widely announce the project and
raise the interest of possible developers/suppliers. All interested companies in the
EU are invited to fill in an online questionnaire that will help the procurers to gain
more insight into the market and the scope.
In the Market Sounding Review, a summary/conclusion of the questionnaires will be
made.
In the Market Consultation, the Nightingale procurers will invite all those interested
companies for a face to face workshop & dialogue, with the aim of further explaining
the Nightingale and obtaining more insight in the feasibility and technical
developments & possibilities. Partly based on the results of this Market Consultation,
the Nightingale procurers will determine the scope and criteria that will be used in the
PCP. The Market Consultation will be concluded with a formal Call for Tender.
Interested companies (or consortia) can then submit their proposal on paper. These
submitted proposals will be evaluated and the six companies with the most
promising proposals will be invited to participate in the first phase Nightingale PCP
process (Phase 1). A contract between the Nightingale consortium and every
individual company will be signed.
From there, the Nightingale PCP process will include three phases of solution
development.
In Phase 1, which will last 3 months, the six selected companies will each develop
their proposal, including feasibility studies. The Nightingale consortium partners will
evaluate the tenders with respect to their technical, economical and organizational
viability. Four companies (minimum: 3) will be invited to proceed to the second
phase, in which actual prototype systems will be developed and tested.
In Phase 2, one or more interim evaluations are foreseen to assist in continuously
improving the prototypes under development. The consortium may decide to seek
the advice of an independent advisory board. After critical evaluation of these early
prototypes in the four hospitals at the end of the second phase, two companies will
be invited to develop and test a pre-production model within the hospitals during the
last, third phase of the PCP process.
Phase 3: in this phase the prototype system will be refined to a pre-production
model. Again, one or more interim evaluations and feedback sessions are foreseen,
based on field tests with actual patients and care givers using the system. During the
entire PCP process, selection of companies/consortia to enter the next phase will be
based on transparent and objective evaluation criteria. After this phased process, the
PCP process ends.
15
Executive Summary
Clinical
need – ‘Common Challenge’
Nightingale
PCP #727534
Patients die because signs of deterioration are missed. There is a huge unfulfilled
need for better monitoring of vital signs to identify high-risk patients who are on
general hospital
wards or atPatients
home. Patient
deterioration
often overlooked
or not
Nightingale
- Connecting
and Carers
usingiswireless
technology:
detected
at all.Challenge
One of the reasons is the intensity in nursing and frequency of vital
The
Common
signs monitoring which decreases from the Intensive Care via ward towards home.
Early detection of vital instability is crucial to prevent death and disability.
Public Procurement of Innovation (PPI)
Pre-commercial Procurement (PCP)
In order to help the companies with such new highly innovative technology to
Five leading European academic hospitals (Utrecht, the Netherlands; Stockholm,
overcome market inertia, the European Commission also facilitates the uptake of
Sweden; London, United Kingdom; Leuven, Belgium and Aachen, Germany) uses
innovative
solutions by the intended end-users. The EU can subsidize the purchase
the European commission’s Pre-commercial Procurement (PCP) funding scheme to
of large
quantities
of the new
devices
or solutions
using
the Public
Procurement
challenge
and stimulate
European
industry
to develop
a system
to connect
patients of
Innovation
(PPI)
model.
The
Nightingale
Consortium
intends
to
conduct
a
PPI after
and carers to wirelessly monitor patients’ vital signs and identify high risk individuals.
theThe
completion
the PCP
if the
PCP is successful
(i.e., atsolution
least one
‘market-ready’
available of
budget
for the
development
of this innovative
is 5.3
million
solution
has
been
generated).
euro, consisting of 3.75 million euro in the form of innovation subsidies to the
industry.
For more information about PCP, please visit the EU website by clicking on the links
below:
Finding a solution
https://ec.europa.eu/digital-single-market/en/innovation-procurement
The solution should consist of one or more unobtrusive wireless sensors that do not
interfere with the patient’s daily activities or rehabilitation. Intelligent analysis
software
See
also: is a key feature of the requested system, as the false alarm rate must be
extremely low to be acceptable to the users. It should be well-integrated into the
https://ec.europa.eu/digital-single-market/en/news/frequently-asked-questions-abouthospital’s electronic medical record system. The sensors should involve a
pcp-and-ppi
mechanism that allows patients and informal carers to communicate with their
healthcare team, and allow entry of qualitative data (e.g., pain and well-being
scores).
We expect that successful implementation of such a system will empower patients
and carers. It will have the potential to transform healthcare by reducing death and
disability from undetected deterioration, and can provide a ‘safety net’ to high-risk
patients after discharge home.
We invite the industry to solve this huge unfilled need for better patient monitoring!
1*
*
This*project*has*received*funding*from*the*European*Union’s*Horizon*2020*research*and*innovation*programme*under*grant*agreement*No727534*
16