VA Palo Alto Health Care System VA Palo Alto

ABSTRACT
The
The results of a 10
month clinical study of a novel FDA h results
l off a 10‐month
h clinical
l
l study
d off a novell FDA
approved biofeedback device (the Paro
pp
(
p
approved
biofeedback device (the
Paro Robot) is presented. Robot)) is presented
Staff
Robot
Staff were trained to expose patients to use of the Paro
Robot St ff were trained
t i d to
t expose patients
ti t to
t use off the
th Paro
P
R b t
i specific
ifi clinical
li i l guidelines
id li
V i bl representing
i
using
using specific clinical guidelines. Variables representing Variables
positive and negative behavioral and mood observations were positive and negative
p
g
behavioral and mood observations were
reported
patients
order
pre
reported for individual patients in order to characterize pre‐
epo ted for
o individual
d dua pat
e ts in o
de to ccharacterize
a acte e p
e
i use and
d post‐use
b
i
i
use, active use, and post
use observations of patients. use active
observations
off patients
Secondaryy analyses
Secondary analyses regarding medication use and staff y regarding
g
g medication use and staff
subjective
ratings
were also
j
g of change
g as a result of Paro use
subjective ratings of change as a result of Paro
use were also reported
reported. Significance of findings is briefly discussed.
Significance
off findings
is
d Si
ifi
fi di
i briefly
b i fl discussed
di
d
D lil h O Noronha
N
h , PsyD
P yD, Kathy
K thy Craig
C ig, OT, Christina
Ch i ti Yee
Y , CTRS, RTC, CDP,
Delilah O. Noronha, PsyD, Kathy Craig, OT, Christina Yee, CTRS, RTC, CDP, Delilah
G ff
W
d i GNP
Lane PhD
Hendrix
Geoffrey
Geoffrey W. Lane, PhD, Levanne Hendrix, GNP, MSN, PhD, W. LLane,
PhD, LLevanne H
Hendrix,
GNP, MSN
MSN, PhD
PhD,
Jennifer
Jennifer Lam, RN, Cherina Tinio, RN, Ann Narciso, RN
RN,
J
if Lam,
L
Lam
RN Cherina
Ch i Tinio,
Ti i RN,
Tinio
RN Ann
A Narciso,
N i RN
Narciso
VA Palo Alto Health Care System
VA
Palo Alto Health Care System
y
INTRODUCTION
S pl Description
Sample
Sample Description
D
ip i
The Paro
The
h Paro robot is a robot
b is a “mental
“mental commit robot
l commit robot”
b ” designed to elicit the positive designed
d
d to elicit
l the
h positive
biopsychosocial variables that so
variables that so‐called
called "pet
pet therapy
therapy" is believed to elicit is believed to elicit
(Wada Shibata Musha & Kimura 2008)
(Wada, Shibata, Musha, & Kimura, 2008).
There is limited evidence that the Paro Robot in both group and There is limited evidence that the Paro
Robot in both group and
individual settings
individual settings has the effect of reducing depression (Shibata et al., g has the effect of reducingg depression
p
((Shibata et al ,
2004)), i
2004), increasing positive social interactions (Wada & Shibata, 2007; i g p iti
i li t
ti
((W d & Shib t , 2007;
Kidd et al.,
Kidd et al., 2006), and improving cognitive functioning in demented l 2006),
2006) and
d improving
i
i cognitive
i i functioning
f
i i in
i demented
d
d
older
older adults (Wada et al., 2006). Very limited case
ld adults
d l ((Wada
d et al.,
l 2006).
) Very llimited
d case‐study
study evidence d evidence
d
indicates that use of the Paro Robot reduces behavioral and indicates that use of the Paro
Robot reduces behavioral and
psychological symptoms in dementia (Marti Bacigalupo Giusti
psychological symptoms in dementia (Marti, Bacigalupo, Giusti, Mennecozzi & Shibata 2006; cited in Broekens et al., 2009). Mennecozzi, & Shibata, 2006; cited in Broekens
et al 2009)
This quality
This quality improvement pilot project was conducted to evaluate the q
y improvement
p
pilot
p
p
project
j was conducted to evaluate the
effectiveness of the Paro
ff ti
f th P
R b t i i
Robot in increasing positive outcomes and i g p iti
t
d
reducing
reducing negative outcomes among VA long
d i negative
ti outcomes
t
among VA long‐term
l
t
term care residents in a care residents
id t iin a
l k d dementia
locked
locked dementia unit. d
unit.
Sample consists of 14 residents at a secured dementia unit at the VA Palo Alto Sample
l consists off 14 residents
d
at a secured
d dementia
d
unit at the
h VA Palo
l Alto
l
Healthcare Systems Menlo Park division (e g the Community Living Center)
Healthcare Systems Menlo Park division (e.g., the Community Living Center). Average age of residents in the sample was 84 All residents had a diagnosis of
Average age of residents in the sample was 84. All residents had a diagnosis of dementia in their medical charts are male and are either listed as of Caucasian
dementia in their medical charts, are male, and are either listed as of Caucasian ethnicityy or “unknown
ethnicity or unknown.” Five of the nine residents tracked were respite care (short Five of the nine residents tracked were respite
p care (short
(
stay) residents, the remainder are long term care residents.
t y)) id t , th
i d
l gt
id t
METHOD
Primary analyses (change measures) are shown in Table 1, below. Secondary P
Primary
i
analyses
l
( h
(change
measures)) are shown
h
in
i Table
T bl 1,
1 below.
b l
Secondary
S
d
analysis
analysis (e.g., use / avoidance of PRNs, staff subjective ratings of change as result of l
(e
( g use / avoidance
d
off PRNs staff
ff subjective
b
ratings off change
h
as result
l off
Paro use) are in Tables 2 and 3. use) are in Tables 2 and 3
The Paro
The
Paro robot has been utilized on the unit as part of usual clinical care. robot has been utilized on the unit as p
part of usual clinical care
This is a q alitati e, clinical q ality impro
This is a qualitative, clinical quality improvement project evaluating it’s p ement p
project
j e al atingg it’s
i p t
impact on patient progress and provider satisfaction within a secured p ti t p g
d p id
ti f ti
ithi
d
d
dementia
dementia unit. Several phases of this project were completed over a 10
i unit.
i SSeverall phases
h
off this
hi project
j were completed
l d over a 10‐
10
month period
month period. A multidisciplinary staff at a Veterans Health Affairs (VHA) Community A
multidisciplinary staff at a Veterans Health Affairs (VHA) Community
Living Center (CLC; nursing home) developed a detailed protocol for use of
Living Center (CLC; nursing home) developed a detailed protocol for use of the Paro, and then trained nursingg staff in its indicated use Trackingg sheets
the Paro, and then trained nursing staff in its indicated use. Tracking sheets were developed which required unit staff to note several targeted behaviors d l p d hi h q i d it t ff t
t
lt g t db h i
and
and provider actions (i.e. decreased wandering, use / nonuse of PRN, etc). d provider
id actions
i
(i.e.
(i decreased
d
d wandering,
d i use / nonuse off PRN,
PRN etc).
)
Tracking forms were entered into an SPSS data file Negative and positive
Tracking forms were entered into an SPSS data file. Negative and positive behavioral and mood observations were collapsed together to form
behavioral and mood observations were collapsed together to form composite variables to simplify the primary analysis Primary analysis took
composite variables to simplify the primary analysis. Primary analysis took the form of a comparison of mean composite variables across patients
the form of a comparison of mean composite variables across patients, across the three studyy p
across the three study periods (e.g., pretreatment, active treatment, and periods ((e g , p
pretreatment, active treatment, and
posttreatment, or Time 1, Time 2, and Time 3). p
tt t
t, Ti
1, Ti
2, d Ti
3))
TTable 2. Frequency of Staff Subjective Rating of Change Table
bl 2.
2 FFrequency off St
Staff
ff SSubjective
bj ti R
Rating
ti off Ch
Change
in Patient Behavior due to Paro
in Patient Behavior due to Paro
RESULTS
A total of 47 observations were generated across a 10 month period. Each patient A
totall off 47 observations
b
i
were generated
d across a 10 month
h period.
i d Each
E h patient
i
had an average of three recorded contacts with the Paro robot across the study had an average of three recorded contacts with the Paro
robot across the study
period however contacts were very unevenly distributed across residents; the top
period, however, contacts were very unevenly distributed across residents; the top three residents accounted for nearly 2/3 (63 6%) of observational data generated
three residents accounted for nearly 2/3 (63.6%) of observational data generated. The average duration of a patient’ss use of the Paro
The average duration of a patient
use of the Paro Robot was approximately 37 Robot was approximately 37
minutes (st
minutes (st. dev. = 19).
( dev 19))
TTable 3. Frequency of Staff Reported Administration / Table
bl 3.
3 FFrequency off St
Staff
ff R
Reported
t d Ad
Administration
i i t ti /
Avoidance of PRNs due to Paro
Avoidance of PRNs due to Paro
Table 1. Means of Composite Mood / Behavioral Variables Table
ab e 1. Means
ea s of
o Composite
Co pos te Mood
ood / Behavioral
e a o a Variables
a ab es
(change measures) Across Time
(change measures) Across Time
0.4681
DISCUSSION
0.3723
0 2872
0.2872
0 1206
0.1206
0 0071
0.0071
Ti
1
Time 1
Ti
2
Time 2
Pos. Composite
Pos. Composite Variable
P
C
it Variable
V i bl
0.0182
Ti
3
Time 3
Neg.
Neg. Composite Variable
N Composite
C
it Variable
V i bl
Across 47 observations collected, data from this pilot project suggest the A
47 b
ti
ll t d, d t f
thi pil t p j t gg t th
P
Paro
R b t is
Robot
Robot is an effective intervention for increasing positive behavioral i an effective
ff ti iintervention
t
ti ffor iincreasing
i positive
iti b
behavioral
h i l
and
and mood observations in VA demented nursing home residents, and d mood
d observations
b
in VA demented
d
d nursing home
h
residents,
d
and
d
decreasing negative observations of the same Moreover staff appeared
decreasing negative observations of the same. Moreover, staff appeared to believe that use of the Paro avoided a measurable number of to believe that use of the Paro
avoided a measurable number of
medication administrations (e g pain medication anxiolytic or medication administrations (e.g., pain medication, anxiolytic
or
antipsychotic
antipsychotic medication), and resulted in all patients using the Paro
py
medication)) and resulted in all patients
p
usingg the Paro
either improving or returning to baseline after it’s use (e.g., no patient ( g , p ti t
ith i p i g
t i g t b li
ft it’
seemed
do worse after
the Paro
Robot).
seemed to do worse after using the Paro
d tto d
ft using
i th
P
R b t) Further,
Robot). Further, more F th more
ll d research
h is needed
d d to ffurther
h d
l
h h patients the
h
controlled
controlled research is needed to further delineate in which patients the delineate
in which
Paro Robot is most effective and in what situations the Paro
Robot is most effective and in what situations the Paro Robot can Robot can
be expected to be maximally effective Also of interest would be to
be expected to be maximally effective. Also of interest would be to investigate the overall cost effectiveness of this intervention as
investigate the overall cost‐effectiveness of this intervention as compared
to standard pharmacological
and non pharmacological
compared to standard pharmacological and non‐pharmacological p
p
g
p
g
interventions. i t
ti