Exploring the Future Potential of AI-Enabled Smartphone Processors
James Barlow
1. Mainstreaming remote care in the UK. Lessons
from the Preventative Technologies Grant and
Whole System Demonstrators programmes
James Barlow
NHS Connecting for Health Evaluation Programme 2009
Masterclass
17 September 2009
www.haciric.org
2. Overview
▬ Research background
▬ Remote care
▬ The future never quite arrives
▬ What we know from existing experience and
research
▬ Conclusions
3. Research background
▬ Initial EPSRC funded projects on telecare planning and
implementation, 2000-03 and 2003-2006
▬ Evaluation of Welsh telecare programme (continuing)
▬ EPSRC funded project on PTG implementation (2006-
2008)
▬ Part of DH funded consortium evaluating the WSD
programme and WSDAN (Imperial, King’s Fund, UCL,
Oxford, Manchester, LSE)
▬ James Barlow, Steffen Bayer, Jane Hendy, Theti
Chrysanthaki
4. Terminology
▬ ‘Telecare’
▬ ‘Telehealth’
▬ ‘Telemonitoring’
▬ ‘Telemedicine’
▬ ‘Assistive technology’
▬ ‘Smart homes’
▬All are used interchangeably to describe
the remote delivery of health and social
care
5. Remote care applications
Mitigating risk
Safety and security monitoring, e.g.
Bath overflowing, gas left on, door unlocked
Prevention Prevention
Information & The Individual monitoring:
communication, e.g. • Physiological signs
health advice, virtual individual in • Lifestyle / activities
self-help groups their home or
wider
environment
Improving
functionality Electronic assistive technology, e.g.
Prevention
environmental controls, doors
opening/closing, control of beds
6. Policy drivers
▬ The UK has taken a strong
lead. Over 20 government
reports since 1998 have
called for telecare
▬ New finance (£170m +) via
Preventative Technology
Grant, Whole System
Demonstrators and other
initiatives
8. Practice by Telephone
The Yankees are rapidly finding out the benefits of the telephone. A newly
made grandmamma, we are told, was recently awakened by the bell at midnight,
and told by her inexperienced daughter, "Baby has the croup. What shall I do
with it?" Grandmamma replied she would call the family doctor, and would be
there in a minute. Grandmamma woke the doctor, and told him the terrible
news. He in turn asked to be put in telephonic communication with the anxious
mamma. "Lift the child to the telephone, and let me hear it cough," he
commands. The child is lifted, and it coughs. "That's not the croup," he declares,
and declines to leave his house on such small matters. He advises grandmamma
also to stay in bed: and, all anxiety quieted, the trio settle down happy for the
night
The Lancet 29 Nov 1879, Page 819
With thanks to Nicholas Robinson
9. … its arrival has been heralded
throughout the last decade
"The innovations we will
encounter as we step beyond
feasibility are dazzling in their
potential"
R. Merrell, Yale University
School of Medicine, 1995
"Over the next decade, the telemedicine
industry will expand into new markets and
service areas. Furthermore, its rapid rise will
have a profound impact on the delivery and
quality of medical care worldwide. In the United
States alone, we expect telemedicine will
represent at least 15 percent of all health care
expenditures by 2010”
Telemedicine Industry Report 2000
“Telecare has arrived. This year’s annual
review reflects the transformation of our
sector from social alarms to Telecare, and
the repositioning of the Telecare service
model from the periphery of housing,
social care and health to centre stage”
Association of Social
Alarms providers, 2004
“2008: The year telecare
grows up?”
E-
Health Insider, 2007
With thanks to ?What If!
10. Diffusion of telecare in Surrey 1998-2005
COPD Project
Brockhurst Dementia unit
LAA: Safe Thames Ward, Molesey Hospital
At Home
Columba MEWS Hospital Discharge project
NEECH videophone pilot
Leatherhead Hospital
Mid Surrey Falls Project Dormers SMART House
Guildford Falls Project
Dray Court Telecare flat Mid Surrey Wristcare pilot
Tandridge Telecare Flat
COPD at Home Project
Community Alarm Teams,
Elmbridge, Guildford, Mole Valley
10
& Runnymede
11. ▬ … but despite thousands of
pilot or trial projects remote
care has not yet become a
mainstream part of care
delivery
▬ Pockets of excellence don’t
spread and pilot projects are
not sustained
12. ▬Lack of progress in UK (and elsewhere)
is largely due to:
• organisational problems (esp. integration
within and between care providers)
• a lack of obvious business models
• … and limited benefits evidence is also
playing a part
13. What’s needed to stimulate remote care?
Adoption Spread Mainstreaming
‘Business case’
Evidence
Evaluation
Awareness
uptake
Project mgt Leadership
Champions
Enthusiasts Pump priming
Grants
time
Source: Barlow, Hendy, Chrysanthaki
14. The existing evidence base
Focus of study Evidence on:
Individual outcomes, i.e. Systemic outcomes, i.e.
clinical or QOL economic impact or impact on
improvement processes
Specific application, Relatively good, growing Limited, problematic – poor
e.g. aimed at patients – numerous individual specification of assumptions,
with diabetes studies on which to build lack of robust data
systematic reviews
General application, Largely anecdotal, Virtually unresearched –
e.g. aimed at a growing – not yet peer based on simulation
general population reviewed modelling with limited data
(e.g.‘frail older
people’)
Barlow et al: (JTT 2007)
15. A lack of evidence isn’t always a barrier
to government policy initiatives …
16. … but it is now becoming important for remote care
implementation because … … for social care
organisations,
▬ Remote care now embracing new research, … we’re
stakeholders across the care system – very practical, … to
move from social to health care have (evidence), that
fits more with health
▬ Financial investment beyond the pilot
stage needs to be made
▬ More robust evidence needed to build
business cases for all parties –
provider, commissioner, technology
… you need the
supply chain evidence, … when you
▬ Evidence increases stakeholder get the evidence you
receptiveness get true buy-in
18. We don’t even know how many remote
care users there are
▬ Poor data due to
4000
inconsistent
3000
definitions
2000
▬ Example from 5 1000
0
leading LAs … Site Site Site Site Site
1 2 3 4 5
Recorded users 2006
Recorded users 2008
Source: Hendy & Barlow
New users claimed in 2008
19. ▬ And despite the reported benefits in
terms of admissions avoidance, speedier
discharges etc, this is largely based on
anecdote or poorly designed studies
20. Exploring the potential impact on
healthcare
▬ Simulation modelling experiments can help
us think about how remote care might
change services
▬ … the figures aren’t important in these examples
21. Frail elderly care
death r w Inst
from HC to
waiting Inst entry
3
fM
death rate Inst
600
waiting entry fM
Inst fM from waiting to
Inst fM
Inst entry 3
death rate death rate
550
HC fL HC fM
death rate from HC fL from hc fM to from HC fH
h to h fL to fM to waiting Inst from waiting to death rate Inst
from HC fH
healthy HC fL HC fM HC fH waiting Inst entry 4 entry fH
aging from healthy to from HC fL to
HC fL HC fM
from HC fM to
HC fH death rate
Inst fH Inst fH
500
Pop. in instit. care
HC fH
death r w
Inst fH
share to
TC
450 pessimistic
from healthy to effect of TC on frac
rate to inst care entry TC fH to optimistic
TC fL
fM
to waiting Inst
waiting Inst
400
from TC fL
to h
from TC fM effect of TC on frac
rate to inst care entry
fH
best guess
from TC fH to
death rate TC fL
from TC fM
to fL
TC fM
fM
TC fH
350 base run
TC fL from TC fL to from TC fM to death rate
TC fM TC fH TC fH
effect of TC on fty
progression
300
250
200
132 1
5 10 15 20
Time (years)
130
Admissions / month
128
Base case
126
Best guess
A 20% decline in
124 demand for care
122
Falling home places?
(initially)
120
0 2 4 6 8 10 12 14 16 18 20
hospital
Time (Years) admissions?
Source: Bayer & Barlow
22. Effect of telecare on care costs in year 20
3 - 5% reduction in costs
Change 2.0%
in costs
0.0%
-2.0%
-4.0%
-6.0% 20% reduction
-8.0%
20% less
Same
80% reduction Effect of telecare
on entry into
40% more
Cost of telecare institutional care
package compared
to a conventional
care package
23. Chronic heart failure
frac death r at risk
frac death r
asympt
frac death r sympt
Stabilisation in the
dying at risk
frac r dev HF dying
asymptomatic
dying
symptomatic
demand for hospital
admissions?
asymptomatic symptomatic time constant
high risk transfer to TC
becoming high unknown developing usual care
risk developing HF symptoms
investment in leaving high risk frac r dev
prevention
detection of symptoms
transfer to TC
detection fraction presymptomatic effectiveness of
cost of risk without screening HF managing
reduction per frac r dev sympt unsymptomatic
investment in known
person screening
TC places
asymptomatic
screening cost and known developing sympt
per person known disease symptomatic
dying known
TC
total hospital days
unsymptomatic
dying sympt TC
frac death r
<frac death r
asympt>
sympt TC 100,000
<frac death r 1 1
1
sympt> TC effect on frac
death r sympt 90,000 1 1 1
1 3 1 3 3 3 3
1 34 34 345
4 45 45 5
2 23
5 5 45 45 2
80,000 2
2
2
2 2
2
70,000
60,000
0 25 50 75 100
Time (Month)
Source: Bayer & Barlow total hospital days : base 1 1 1 1 1 hospital days/Month
total hospital days : TC 3M 2 2 2 2 2 hospital days/Month
total hospital days : Prevention 3M 3 3 3 3 hospital days/Month
total hospital days : Screening 3M 4 4 4 4 hospital days/Month
total hospital days : TSP 1M 5 5 5 5 5 hospital days/Month
24. Conclusions
▬ Remote care has potential in managing LTCs and
coping with aging population
▬ There is support for remote care at the individual level
– the hurdles are at a system level
▬ ‘Evidence’ for costs / benefits becoming more
important as pilots move towards mainstream
investment decisions
▬ … but stories still crucial in convincing sceptics
▬ WSD may help, but need more resources to support
gathering local data on consistent basis