Lessons out of Chaos - Human-Computer Interaction and Innovation in Handheld, Mobile and Wearable Technologies - page 125

are often reluctant to take the technology into this
real world where they lose control and a sense of
security and where the chaos of real life rears its
head and confounds test protocols, schedules and
projected costs. “Why can't people behave the
way they are supposed to and use the technology
the way it was intended”, is a phrase often heard
when technology is taken from the lab to the real
world. Thus, developers spend inordinate amounts
of time trying to eliminate the noise of real life,
rather than accepting that the noise is the key to
making the technology work.
We have spent the last nine years dealing
with this noise as we moved a new behavioral
monitoring technology from the laboratory to
the real world. Initially, we viewed this noise as
a hindrance to this successful transition, but over
time we became convinced that without embrac-
ing the chaos and listening very carefully to the
noise, our monitoring system could not be suc-
cessfully moved from the laboratory to the real
world. This process has been painful at times, but
we believe that the lessons we have learned can
benefit other developers and perhaps, help them
embrace the chaos also (see Alwan et al., 2006;
Alwan et al., 2007; Biswas, Mokhtari, Dong &
Yap, 2010; Blackburn, Brownsell, & Hawley,
2006; Celler, Lovell & Brasilakis, 2003; Cleland,
Rigby, Janssens & Balk, 2005; Helal, Cook &
Schmaltz, 2009 for other examples of behavioral
monitoring approaches).

was comprised of five motion sensors and a base
station connected, via the Internet, to a website
that processes the sensor data and converts them
to information that is then displayed with graphics
and text for caregivers. The PIN secure website
provides a daily summary for each person be-
ing monitored for six activities: waking time;
bathroom falls; the taking of medication; meal
preparation; overall level of activity; and nighttime
bathroom use, as well as ambient temperature. In
addition, the ELMS provides emergency alerts
for bathroom falls, non-wake-up, and high or low
temperature and has the ability to produce monthly
summary charts for all monitored activities. See
Figure 1 for an example of a QuietCare ® daily
summary page.
We developed a five stage real world testing
design that would allow us to install the system
into increasingly more complex environments. In
this way, it was hoped that we would be able, at
each stage, to refine the system before moving to
the next more challenging environment. We have
progressed through four of these stages and our
hopes and fears have been more than met as the
messiness of the real world has constantly intro-
duced noise and chaos into our well thought out
testing protocols. The five environments into
which we have tested the ELMS are: (1) an activ-
ity of daily living suite at a large regional hospi-
tal; (2) the homes of volunteers as part of indi-
vidual sequential testing; (3) the residences of
five elderly individuals in a high rise apartment
building with limited care provision; (4) in the
independent residences of 34 chronically ill indi-
viduals with nursing services provided by a home
health care company; and (5) in four different
care environments ranging from assisted living
to independent households named the Caring
Home Initiative (CHI). Although we have learned
many things from each of the completed stages,
we will limit ourselves to the discussion of the
two most illustrative examples from each of the
stages.

STAGES OF REAL WORLD TESTING

Over six years of laboratory research and devel-
opment on behavioral monitoring technology
resulted in the first real world ready iteration of
the Everyday Living Monitoring System (ELMS)
which became QuietCare ® when the system was
commercialized by Living Independently Group.
(In 2009 the patents and associated intellectual
property were sold to General Electric and the
new product branded GEQuietCare ® .) The ELMS