Anthrozos

A multidisciplinary journal of the interactions of people and animals

ISSN: 0892-7936 (Print) 1753-0377 (Online) Journal homepage: http://www.tandfonline.com/loi/rfan20

The Effect of Animal-Assisted Therapy on Stress

Responses in Hospitalized Children

Chia-Chun Tsai, Erika Friedmann & Sue A. Thomas

To cite this article: Chia-Chun Tsai, Erika Friedmann & Sue A. Thomas (2010) The Effect of
Animal-Assisted Therapy on Stress Responses in Hospitalized Children, Anthrozos, 23:3, 245-258

To link to this article: http://dx.doi.org/10.2752/175303710X12750451258977

Published online: 28 Apr 2015.

Submit your article to this journal

Article views: 522

View related articles

Citing articles: 8 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=rfan20

Download by: [UNO Criss Library] Date: 10 February 2017, At: 09:35
 ANTHROZOS VOLUME 23, ISSUE 3 REPRINTS AVAILABLE PHOTOCOPYING ISAZ 2010
PP. 245258 DIRECTLY FROM PERMITTED PRINTED IN THE UK
THE PUBLISHERS BY LICENSE ONLY

The Effect of Animal-Assisted

Therapy on Stress Responses
in Hospitalized Children
Chia-Chun Tsai*, Erika Friedmann and
Sue A. Thomas
* Yuanpei University Department of Nursing, Taiwan, R. O. C.
University of Maryland School of Nursing, Baltimore, USA

Address for correspondence:
Chia-Chun Tsai,
6F No.89 Wen Ren Street,
Chupei, Hsinchu 302,
Taiwan, R.O.C.
E-mail: cctsai002@gmail.com
ABSTRACT Hospitalization is a major, stressful experience for children. The
stress associated with childrens hospitalization may lead to negative physio-
logical and psychological sequelae. Pediatric healthcare professionals can de-
velop interventions to decrease childrens stress during hospitalization.
Although Animal-Assisted Therapy (AAT) frequently is used to alleviate the
stress of hospitalization, little scientific evidence exists on its efficacy in that sit-
uation. This study examined the effects of AAT on cardiovascular responses,
state anxiety, and medical fear in hospitalized children. A quasi-experimental,
repeated measures design was used. Children (8 girls, 7 boys, aged 7 to 17
years) participated in AAT and comparison visits on two consecutive days;
they were assigned to AAT (n = 9) or comparison (n = 6) visits first. Childrens

Anthrozos DOI: 10.2752/175303710X12750451258977

systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate
(HR) were measured 18 times: 3 pre-, during, and post-visit measurements
each. State anxiety and medical fear were measured after each visit. Repeated
measures ANOVAs were used for data analysis. SBP decreased from before
to during to after AAT, while SBP decreased from before to during the com-
parison visit and increased from during to after the comparison intervention.
The decreases in SBP after AAT continued after the intervention was over.
The findings suggest that the cardiovascular effect of AAT may continue for at
least a few minutes after the AAT ends. Childrens anxiety and medical fear did
not differ after the AAT visit compared with the comparison intervention. This
exploratory study indicates that AAT can decrease physiological arousal in
hospitalized children and therefore may be useful in helping them cope bet-
ter in a hospital setting.

Keywords: animal-assisted therapy, blood pressure, children, stress

In 2004, more than 2.57 million children in the US under the age
of 15 were hospitalized, with an average length of stay of 4.5
days (DeFrances and Podgornik 2006). Hospitalization is a
245

major, stressful event for children, leading to a high level of anxiety and fear
 The Effect of Animal-Assisted Therapy on Stress Responses

(Scavnicky-Mylant 1987; Tiedeman and Clatworthy 1990; Bossert 1994; Hart and Bossert
1994; Mahat and Scoloveno 2003). Stress has negative health consequences for children. As
many as 35% of all children in America experience stress-related health problems (Kaplan
2000). The long-term physiological effects of stress are likely to cause nonpsychiatric gas-
trointestinal, cardiovascular, or other sequelae later in life (Nutter, Larsen and Sylvester 2006).
Anxiety also affects childrens social and academic functioning (Nutter, Larsen and Sylvester
2006). For several decades, hospitalization has been known to produce undesirable emo-
tional responses in pediatric populations. Children experience both fear (Broome et al. 1990;
Hart and Bossert 1994) and anxiety (Astin 1977; Scavnicky-Mylant 1987; Tiedeman and
Clatworthy 1990) when they are hospitalized.
One goal of pediatric health care professionals is to develop interventions to decrease chil-
drens stress during hospitalization. Animal-assisted therapy (AAT) involves the introduction of
a companion animal to an individual who is not that animals owner, with the expectation of the
individual benefiting while the animal is present (Friedmann and Tsai 2006). Animal-assisted
therapy promotes emotional comfort and decreases loneliness, anxiety, and physical stress re-
sponses (Calvert 1989; Barker and Dawson 1998; Churchill et al. 1999; Friedmann, Thomas
and Eddy 2000; Friedmann and Son 2009). It has been used to alleviate the stress of hospi-
talization and health care experiences in a wide variety of health care settings and has the po-
tential to be effective at reducing psychological and physiological indicators of hospitalization
stress in children (Kaminski, Pellino and Wish 2002; Wu et al. 2002).
Most studies of the effects of AAT on stress involve adult patients and/or elderly individu-
als. For example, AAT has been shown to be effective in reducing chronic psychological and
physiological stress and acute stress responses to stressful tasks in adults and adult patients
(Muschel 1984; Barba 1995; Brodie and Biley 1999; Cole 1999; Cole and Gawlinski 2000;
Connor and Miller 2000; Miller and Ingram 2000; Jorgenson 2002; Stanley-Hermanns and
Miller 2002; Barker, Pandurangi and Best 2003; Friedmann and Tsai 2006; Friedmann et al.
2007; Friedmann and Son 2009). Only a few studies have evaluated the physiological and
psychological effects of AAT on children (Nagengast et al. 1997; Hansen et al. 1999; Havener
et al. 2001; Kaminski, Pellino and Wish 2002; Wu et al. 2002; Sobo, Eng and Kassity-Krich
2006). Two of the previous studies that examined the physiological effects of AAT on hospi-
talized children used dogs in the interventions. In one study, pediatric patients stress levels,
as indicated by HR and respiratory rates, decreased during 10 to 20 minutes of interaction with
dogs (Wu et al. 2002). In the other study, the effects of a dog were compared with the effects
of a child-life program play therapy intervention designed to minimize the stress and anxiety
that often accompany a hospital stay and medical procedures. Blood pressure (BP) responses
to the play therapy and AAT were reported for children randomly assigned to one of the inter-
ventions. Heart rate (HR) was higher in the AAT group than in the play therapy intervention
group (Kaminski, Pellino and Wish 2002).
The presence of a dog can influence childrens physiological stress responses during out-
patient health procedures. The presence of a dog during a dental procedure moderated the
Anthrozos

stress of children who reported distress before the procedure (Havener et al. 2001). Two similar
studies demonstrate the importance of self-comparisons in evaluating the stress-ameliorating
effects of AAT. In the first study young childrens BPs and HRs decreased more during a mock
medical examination with a friendly dog present than one without the dog present (Nagengast
et al. 1997). A similar study, without a crossover design, found no effect of the presence of a
246

friendly dog on physiological responses of young children who were assigned to either a dog
 Tsai et al.

or no-dog condition during a physical examination, but the children showed less behavioral dis-
tress when the dog was present (Hansen et al. 1999). The presence of an unfamiliar but friendly
dog also reduced physiological stress responses to a moderately stressful task in healthy chil-
dren, in a study that used a crossover design (Friedmann et al. 1983).
AAT has been shown to have positive effects on psychological indicators of stress of hos-
pitalized children (Kaminski, Pellino and Wish 2002; Wu et al. 2002; Bouchard et al. 2004;
Sobo, Eng and Kassity-Krich 2006). In studies without a comparison group, AAT reduced per-
ceptions of pain after surgery (Sobo, Eng and Kassity-Krich 2006) and generated positive feel-
ings on the part of children and their parents (Wu et al. 2002; Bouchard et al. 2004). Caregivers
of children who received either play therapy or AAT reported that both interventions improved
the moods of hospitalized children: parents thought their childrens moods were better after
AAT (Kaminski, Pellino and Wish 2002). Adolescents hospitalized in a psychiatric setting with
a resident dog responded positively to the presence of the dog. They viewed the dog as a
component of the milieu, as a friend, a therapist, a comforter, and a distractor (Bardill and
Hutchinson 1997).
Since stress and its effects on an individuals health are a major concern for nurses and
health care professionals, multiple approaches to managing stress have been advocated (Baun,
Oetting and Bergstrom 1991). Little is known about the impact of AAT on reducing physiolog-
ical and psychological stress responses in hospitalized children. Professionals who serve hos-
pitalized children can benefit from a better understanding of the effects of AAT on stress
reduction during hospitalization. This study focused on evaluating the effectiveness of AAT for
reducing physiological (SBP, DBP, and HR) and psychological (fear and anxiety) indicators of
stress in hospitalized children, compared with an alternative intervention. It extends previous
studies by using a crossover design to evaluate the effect of AAT on physiological and psy-
chological aspects of stress in hospitalized children. It was hypothesized that BP and HR would
decrease more in response to AAT than to a comparison intervention (completion of a puzzle)
and that anxiety and fear would be lower after AAT than after the comparison intervention.

Methods
Human subject approval was obtained from the three facilities where the study was conducted
and from the University of Maryland. Informed consent by parents and assent by participants
were obtained prior to participation in the study.
Participants
Convenience sampling was used to recruit a sample of 15 children aged 7 to 17 years with
acute or chronic conditions who were admitted to the pediatric units of participating hospitals
during the research period. The recruitment criteria were that children had to be 7 to 17 years
of age, without significant cognitive impairment, expected to have a minimum of a one-night
stay (two days of hospitalization), able to speak and read sufficient English to participate in the
study, and had adequate physical abilities to participate in testing (e.g., be able to interact with
Anthrozos

a dog or assemble a puzzle). To control for factors that influence BP and HR reactions, chil-
dren who were experiencing severe pain, fever, dehydration, or had NPO (nothing by mouth)
orders were excluded. For security, and to minimize risks from exposure to the dog, only chil-
dren who were not on strict contact or respiratory isolation restrictions and not allergic to or
afraid of dogs were eligible for inclusion. Only children who, along with their parents/guardians,
247

were willing to have a dog visit were recruited for participation in the study.
 The Effect of Animal-Assisted Therapy on Stress Responses

Staff at the three participating institutions screened all children admitted to each facility to
identify those who were potentially eligible for inclusion. They then asked parents if they were
willing to have their child participate in the study. If an affirmative response was received, the
study team was notified and they spoke with both the parent and the child to obtain
consent/assent. Potential participants were told that the study would include the child
participating in a dog visit and a visit to complete a puzzle.
Instruments
A demographic data sheet that included age, gender, grade level, race/ethnicity, previous hos-
pitalization and AAT experiences, and pet ownership was completed by the parent/guardian.
Childrens physiological responses, BP and HR, were measured using a noninvasive BP mon-
itor the Dinamap Pro 400 oscillometric BP monitor (GE HealthCare, FL) with an appropriate
size external cuff. The BP monitor automatically inflates at intervals in 1-minute increments as
set by the investigator. This monitor prints the systolic BP (SBP), diastolic BP (DBP), and HR
and time after each measurement. The Dinamap oscillometric BP monitor is the most com-
monly used noninvasive apparatus to measure BP and HR. The validity and reliability of it are
well established (Ramesy 1979; Yelderman and Ream 1979; Park and Menard 1987). This in-
strument has been used in many studies examining BP responses to stressors (Lynch et al.
1980; Lynch et al. 1981; Friedmann et al. 1982; Lynch et al. 1982; Friedmann et al. 1983) and
in studies of children (Murphy and McGarvey 1994; Boyce et al. 1995). The BP accuracy
meets or exceeds ANSI/AAMI standard SP-10.
In this study, the Child Medical Fear Scale (CMFS) was used to measure medical fear.
It is designed to identify medical fear of children 6 years of age or older. The revised CMFS
consists of 17 items of medical fear for health-care experiences and uses a 3-point, forced-
choice format (1 = not at all, 2 = a little, and 3 = a lot) (Broome et al. 1992). Scores range
from 17 to 51; low scores reflect children who are less fearful of medical experiences, while
high scores indicate greater fear. The reliability coefficients of internal consistency, using
Cronbachs alpha, are 0.93 and 0.87 for the total test (Broome et al. 1988; Broome et al.
1990). Test-retest reliability is 0.84, over a two-week period, indicating a high degree of
stability. Content validity has also been established: there was 78% agreement amongst
three experts on the validity of all items. The CMFS demonstrates concurrent validity, with
a moderately positive relationship with the Fear Survey Schedule (r = 0.71, p < 0.05)
(Broome and Hellier 1987) and the Baker-Wong Faces Scale (r = 0.54, p < 0.01) (Broome
et al. 1990).
Level of state anxiety was measured by the State-Anxiety scale of the State-Trait Anxiety
Inventory for Children (STAI-C) (Spielberger 1970). The STAI-C consists of two 20-item
scales: a State-Anxiety scale that measures transitory anxiety reactions to particular situa-
tions, and a Trait-Anxiety scale that measures a stable, relatively constant aspect of anxiety.
Each item is scored from 1 (very or not) to 3 (very or not). The total possible scores range
from 20 to 60, with higher scores indicating greater anxiety. The STAI-C has excellent psy-
Anthrozos

chometric properties, with high reliability and validity (Spielberger et al. 1973). The test-
retest reliability coefficients range from 0.65 to 0.72 for the State-Anxiety scale, and from
0.44 to 0.94 for the Trait-Anxiety scale (Kendall and Ronan 1990). In an early study by
Spielberger et al. (1973), internal consistency estimates ranged from 0.78 to 0.87. The
State-Anxiety scale was used in this study to quantify childrens anxiety due to particular
248

situations during hospitalization.

 Tsai et al.

Interventions
Two interventions were included in the study: AAT and a comparison intervention (person with
a puzzle to complete). Each child completed both interventions. The order of the interventions
was varied based on the days AAT teams visited each facility and the day of patient admission.
A dog, along with a volunteer (its handler), who regularly visited each facility, served as the
intervention for the AAT visit. The volunteers and their dogs had been recruited, screened,
trained, and certified by pet therapy programs, for example the Delta Society Pet Partners Pro-
gram. They visit the hospitals on a regular basis through their child-life or recreational therapy
programs. Before the dogs are allowed into the hospital, they must be evaluated by a veteri-
narian, to make sure they are healthy and up-to-date on vaccinations and have a good tem-
perament. A dog was only allowed inside a childs room if the parent and child approved. During
the AAT visit, the dog was positioned on the bed next to the child or stayed by the bedside or
chair (if the child was sitting in it). The child was allowed to pet, touch, and brush the dog.
A research assistant who assembled a puzzle with the child served as a comparison in-
tervention. Puzzle completion was chosen as an age-appropriate activity that involved quiet
interaction, could be accomplished while in the hospital bed or chair, and required some move-
ment and attention by the child. Two differently sized puzzles were used depending on the
childs age. A frame tray puzzle with 40 pieces (9.5 inches 10 inches) was used for children
aged 7 to 11 years and a frame tray puzzle with 60 pieces (14.5 inches 10 inches) was used
with those aged 12 to 17 years.
Procedure
Each child participated in two interventions, administered at the same time of day on two con-
secutive days. One session was a 6- to 10-minute AAT session and the other session was a
6- to 10-minute puzzle session. Both sessions took place in the patients private or semi-
private room with limited interruptions, or in the playroom. Before the study, each child was
assigned to either the AAT-visit-first group or the comparison (puzzle)-visit-first group. The
order of the dogs presence was determined by the days animals visited each facility, the day
of the week of the childs admission to the hospital, and the childs availability based on
procedures and medical condition.
Before the first session, parents/guardians completed a demographic data sheet. The
childs arm was measured to determine the appropriate BP cuff size [child/small adult (1826
cm) or adult (2333 cm)]. The Dinamap Pro 400 oscillometric BP monitor, with an appropri-
ately sized external cuff, was placed on the childs left arm to measure his/her BP and HR.
Blood pressure and HR were recorded every 2 minutes for each participant for 6 minutes
before the intervention (baseline), during the 6- to 10-minute intervention visit, and for six
minutes post-intervention. During the AAT session, the dog was brought to the child and was
positioned on the childs bed next to the child for 6 to 10 minutes. During the comparison visit,
the research assistant assembled a puzzle with the child in the childs room for 6 to 10 minutes.
The questionnaires to assess medical fear and anxiety were completed after the intervention
session. The next day, data collection was repeated with the second type of intervention (AAT
Anthrozos

or comparison).
Data Analysis
Descriptive statistics, multiple regression analyses, t-tests, and analyses of variance with re-
peated measures (RMANOVAs) were used to test the hypotheses in this study. Data were
249

screened for missing data, outliers, and non-normal distributions prior to analysis.
 The Effect of Animal-Assisted Therapy on Stress Responses

Demographic data were analyzed using descriptive statistics. To examine the baseline differ-
ences between children assigned to AAT-first and AAT-second (comparison-first) groups,
t-tests were used.
Separate, four way RMANOVAs were used to compare mean SBP, DBP, and HR pre, dur-
ing, and post the AAT and comparison interventions. The between subjects or grouping fac-
tor was which intervention occurred first [FIRST (2 levels): AAT visit on the first day, AAT visit
on the second day]. The three within subjects factors were intervention [VISITTYPE (2 levels):
AAT and comparison], period [ACTIVITY (3 levels): pre-visit, during visit, and post-visit] and
measurement [MINUTE (3 levels): measure 1average of first 2 minutes, measure 2aver-
age of middle 2 to 4 minutes, measure 3average of last two minutes within each period]. The
main hypotheses were tested by the interaction of VISITTYPE and ACTIVITY. The results of
two-tailed tests are presented, and because of the directional hypothesis, p values less than
0.1 were considered significant.
The relationships between the childrens medical fear and state anxiety and their personal
characteristics (e.g., age, gender, and previous hospitalization experience) were examined with
t-tests, Pearsons correlations, and multiple regression analyses. Separate two way
RMANOVAs were used to compare the childrens medical fears and state anxiety measured
after the AAT and comparison visits. The between subjects factor was group [FIRST (2 levels):
AAT visit on the first day, AAT visit on the second day] and the within subjects factor was in-
tervention [VISITTYPE (2 levels): AAT and comparison].

Results
Fifteen hospitalized children (8 girls, 7 boys), aged from 7 to 17 years (M = 10.97, SD = 3.01),
participated in the study. Baseline demographic data for these children are presented in Table
1. Seven children were Black or African-American (46.7%) and six children were White (40.0%).
Eight of the children (53.3%) had previous hospitalization experiences. Only 4 (26.7%) children
indicated that they had been visited previously in the hospital by pets. Eight (53.3%) children
had pets at home (3 dogs, 3 cats, 1 fish), and one had both dogs and cats.
There were no significant differences in gender (2 =1.61, p = 0.21), race (2 = 2.30,
p = 0.512), or age (t(1, 13) = 0.651, p = 0.527) between the AAT-visit-first and the comparison-visit-
first groups. Childrens baseline SBP (t(13) = 0.23, p = 0.824), DBP (t(13) = 1.86, p = 0.086), and
HR (t(13) = 1.23, p = 0.242) did not differ significantly between the two groups (see Table 2).
Physiological Outcomes
For the physiological outcomes, the main test of the hypotheses is the interaction between
VISITTYPE and ACTIVITY. In the RMANOVA for SBP, there was a significant two-way interac-
tion between VISITTYPE and ACTIVITY (F(2,26) = 7.88, p = 0.002; see Figure 1). Systolic BP de-
creased from pre- to during to post-AAT visits. In contrast, SBP decreased from pre- to during
the comparison visits, but then increased back to pre-comparison visit levels within a few min-
utes after the intervention ended. Systolic BP was significantly lower after AAT than before
Anthrozos

AAT (t(14) = 3.09, p = 0.008). There was no significant difference in SBP pre- and post-com-
parison visit (t(14) = 0.73, p = 0.477). For SBP, there also was a significant three-way interac-
tion between FIRST, VISITTYPE, and ACTIVITY (F(2, 26) = 3.56, p = 0.043), indicating that the
order of the visits made a difference in childrens SBP responses. Graphs of the three-way in-
teractions are presented in Figures 2 and 3. In both the children in the AAT-first (Figure 2) and
250

the comparison-visit first (Figure 3) groups, SBP decreased in a linear manner from pre- to
 Tsai et al.

Table 1. Demographic characteristics of the children (n = 15) in the study.

Variable n (%)
Gender
Boy 7 (46.7)
Girl 8 (53.3)
Race
White 6 (40.0)
Black (African American) 7 (46.7)
Asian 1 (6.7)
Ethnicity
Hispanic or Latino 1 (6.7)
Hospitalization History
Previously Hospitalized 8 (53.3)
Previous Overnight Hospitalization 6 (40.0)
Surgery 2 (13.3)
Previous Hospital Pet Visit 4 (26.7)
Pets at Home 8 (53.3)
Dog 3 (20.0)
Cat 3 (20.0)
Both 1 (6.7)
Fish 1 (6.7)

Table 2. Means and standard deviations (SD) for systolic blood pressure
(SBP), diastolic blood pressure (DBP), and heart rate (HR) at baseline for
children in the AAT-visit-first and comparison-visit-first groups.
AAT Visit First (n = 9) Comparison Visit First (n = 6)
Mean SD Mean SD
SBP (mm Hg) 105.78 5.88 106.94 11.37
DBP (mm Hg) 63.59 5.97 58.56 3.42
HR (bpm) 92.96 17.28 103.22 13.39

during to post-AAT visit. Simple effects tests revealed that the decrease in SBP from pre- to
post-AAT in the AAT-visit-first group was not significant. In the comparison-visit-first group,
SBP decreased significantly from pre- to post- (t(5) = 3.87, p = 0.012)) and from during to after
the AAT visit (t(5) = 3.72, p = 0.014). For both the AAT-visit-first and comparison-visit-first
groups, SBP decreased from pre- to during the comparison visit and then increased from dur-
ing to after the comparison visit. The decreases in SBP from pre- to during the comparison visit
(t(8) = 4.22, p = 0.003) and the increase from during to post the comparison visit (t(8) = 2.21,
Anthrozos

p = 0.058) were significant for the group that had AAT first, but not for the group with the com-
parison visit first (pre- to during comparison visit: t(5) = 1.92, p = 0.113; during to post-com-
parison visit: t(5) = 1.95, p = 0.109).
For DBP, there was a significant interaction between VISITTYPE and ACTIVITY (F(2, 26) =
2.54, p = 0.098). Diastolic BP in the combined AAT-first and comparison-visit-first groups in-
251

creased from pre- to during both AAT(t(14) = 2.01, p = 0.064) and comparison interventions
 The Effect of Animal-Assisted Therapy on Stress Responses

118
Intervention
116
 AAT

Systolic Blood Pressure (mmHg)

114  Puzzle

112

110

108

106

104 

102

100 

98
Pre-Visit Visit Post-Visit
Time of Assessment

Figure 1. Mean (+/-SEM) systolic blood pressure pre-, during, and post-
AAT and comparison visits for all children.

118
Intervention
116
 AAT
Systolic Blood Pressure (mmHg)

114  Puzzle

112

110 

108
 
106

104

102 

100

98

96
Pre-Visit Visit Post-Visit
Time of Assessment

Figure 2. Mean (+/-SEM) systolic blood pressure pre-, during, and post-
AAT and comparison visits for children with the AAT visit first.
Anthrozos

(t(14) = 2.04, p = 0.060; see Figure 4). Diastolic BP decreased significantly after AAT visits
(t(14) = 2.94, p = 0.001), but remained elevated and did not decrease after comparison inter-
vention visits. The order of intervention type did not have an effect on DBP responses to the
interventions (F(2,26) = 0. 961, p = 0.41). There was a significant main effect of ACITIVITY
(F(2,26) = 4.54, p = 0.02). Diastolic BP was significantly higher during the combined visits
252

(M = 65.15, SEM = 0.884 mmHg) than pre- (M = 62.48, SEM = 1.02 mmHg; t(14) = 3.16,
 Tsai et al.

118
Intervention
116
 AAT

Systolic Blood Pressure (mmHg)

114  Puzzle

112

110 

108
 
106

104

102 

100

98

96
Pre-Visit Visit Post-Visit
Time of Assessment

Figure 3. Mean (+/-SEM) systolic blood pressure pre-, during, and post-
AAT and comparison visits for children with the comparison visit first.

72
Intervention
70  AAT
Diastolic Blood Pressure (mmHg)

 Puzzle
68


66


64

 
62

60


58

56
Pre-Visit Visit Post-Visit
Time of Assessment

Figure 4. Mean (+/-SEM) diastolic blood pressure pre-, during, and post-
AAT and comparison visits for all children.
Anthrozos

p = 0.007) the combined visits and tended to be higher than post-intervention visits (M =
63.78, SEM = 1.09 mmHg; t(14) = 2.13, p = 0.052).
For HR, there was no significant interaction between ACTIVITY and VISITTYPE (F(2,26) =
4.80, p = 0.82). The main effect of ACTIVITY was significant (F(2, 26) = 3.44, p = 0.047). HR in-
creased significantly (t(14) = 2.72, p = 0.017) from pre- (M = 98.77, SEM = 3.9 bpm) to during
253

(M = 101.14, SEM = 3.9 bpm) the combined interventions and did not change significantly
 The Effect of Animal-Assisted Therapy on Stress Responses

(t(14) = 1.48, p = 1.61) from during to post (M =101.18, SEM = 3.92 bpm) visits. There were no
differences according to VISITTYPE or FIRST and there were no significant interactions; the
changes in HR did not depend on the type of visit or the order of the visits. There was no
evidence that the two types of visits differentially affected HR.
Psychological Outcomes
On the Child Medical Fear Scale, a score of 34 indicates a medium level of medical fear, and
scores above 34 indicate greater fear. Children reported a moderate level of medical fear when
values after both visits were combined (M = 31.03, SD = 1.10). After the AAT visits, five chil-
dren (33%) reported scores higher than 34, while four children reported scores higher than 34
after the comparison visit. Scores on the State-Anxiety scale can range from 20 to 60. A
moderate level of anxiety is indicated by a score of 40; higher scores indicate greater anxiety.
Children had low state anxiety (M = 27.93, SD = 0.49), with a range of 24 to 31.5. No child
reported scores greater than 40 for state anxiety after either the AAT or comparison visits.
Hospitalized girls (M = 27.75, SD = 4.62) tended to have lower medical fear scores after
the comparison visit than hospitalized boys (M = 33.43, SD = 6.02, t(13) = 2.06, p = 0.06).
There were no differences according to gender in medical fear scores after the AAT visit or in
anxiety after either the AAT or the comparison visits (t(13) = 1.64, p = 0.124). Childrens previ-
ous hospitalization experience was not related to medical fear or state anxiety while they were
hospitalized after either type of visit. Older hospitalized children had a lower level of medical fear
after both the AAT (r = 0.62, p < 0.05) and the comparison visits (r = 0.60, p < 0.05), and
they tended to have lower state anxiety post the AAT visit (r = 0.44, p < 0.10). In multiple re-
gression analysis, only age had a relationship with childrens medical fear and state anxiety.
Older children tended to have lower medical fear scores after both the AAT (t(3, 11) = 2.189,
p = 0.051) and comparison visits (t(3, 11) = 1.87, p = 0.088) (see Table 3).

Table 3. Means (M) and standard deviations (SD) for the medical fear and state anxi-
ety scores according to gender and previous hospitalization experience, and correla-
tions (r) for the relationship of age to medical fear and state anxiety.
Medical Fear State Anxiety
Variable Post Post Post Post
AAT Visit Comparison Visit AAT Visit Comparison Visit
Gender
Boy [M (SD)] 34.43 (6.90) 33.43 (6.02) 28.43 (1.51) 27.71 (1.98)
Girl [M (SD)] 29.38 (5.01) 27.75 (4.62) 26.25 (3.20) 29.38 (2.97)
Previous
Hospitalization
No [M (SD)] 31.14 (5.81) 29.57 (6.35) 27.86 (2.79) 28.14 (3.58)
Yes [M (SD)] 32.25 (7.07) 31.13 (5.82) 26.75 (2.71) 29.00 (1.51)
Age (r) 0.62** 0.60** 0.44* 0.20
Anthrozos

n = 15; *p < 0.10; **p < 0.05

Examination of medical fears with ANOVAs revealed that VISITTYPE (AAT or comparison)
did not have a significant effect on the CMFS scores (F(1, 13) = 3.79, p = 0.073). The results in-
dicated that childrens level of medical fear after the AAT visit (M = 31.73, SEM = 1.63) were
254

similar to those after the comparison visit (M = 30.40, SEM = 1.52). There were no interaction
 Tsai et al.

effects for FIRST and VISITTYPE on medical fear (F(1, 13) = 1.28, p = 0.28). There was no
evidence that the AAT visit reduced medical fear.
There was no significant interaction between FIRST and VISITTYPE for state anxiety
(F(1, 13) = 0.192, p = 0.67). State anxiety did not differ after the AAT and the comparison visits
according to which visit occurred first. There also was no significant effect of VISITTYPE.
(F(1, 13) = 1.732, p = 0.21). There was no evidence that the AAT visit affected childrens anxiety.

Discussion
The findings from this exploratory study suggest that AAT may lower the SBP of hospitalized
children. Systolic BP decreased from before to during to after AAT; SBP decreased from be-
fore to during and increased from during to after the comparison intervention. The decreases
in SBP after AAT continued after the intervention was over. These findings suggest that the
effect of the AAT visit lasted beyond the time of the intervention itself.
Diastolic BP and HR increased from before to during both interventions. This is likely due
to the physical activity and communication involved in interacting with the dog and the
researcher. Previous studies document that when people talk, BP and HR increase from 10
to 50% (Thomas and Friedmann 1994). It is only natural for people to talk to dogs while they
are interacting with them (Friedmann 1995; Friedmann, Thomas and Eddy 2000). While the
comparison intervention was designed to require movements similar to those that occur dur-
ing AAT, informal observations suggest that the amount of physical exertion during interaction
with the dog was greater than the physical exertion during the comparison visit. Further, the
participants may have talked more during AAT than during comparison visits. The increases
in cardiovascular measures may be due to exertion and or speaking and could mask or even
overwhelm the anti-arousal effects of the AAT visit. Differences in speaking in the two inter-
ventions could also interfere with assessment of their effectiveness at reducing stress.
In the current study, the children were resting in their rooms when the dog visited; they
were not undergoing a stressful event. In previous within-subject design studies that showed
cardiovascular benefits from the presence of a dog, childrens cardiovascular stress indicators
were lower while undergoing stressful events, such as physical examination (Nagengast et al.
1997) or reading aloud (Friedmann et al. 1983), when a dog was present than when it was not
present. If the current study was held in a more stressful situation, the effects of AAT on
physiological and psychological outcomes might have been greater.
There was no evidence that the children who participated in the study were experiencing
psychological stress. Their anxiety levels were low. Thus, no intervention could be expected
to reduce anxiety. This finding is consistent with those that demonstrate that having a dog
present does not reduce physiological stress indicators during dental procedures, in the gen-
eral population. Havener et al (2001) found that a dogs presence was effective at reducing
these indicators for children who were distressed prior to their dental procedure.
The results of this study indicated that hospitalized girls tended to have lower medical
fear than hospitalized boys, after the comparison visit. Older children had lower medical fears
Anthrozos

after both visits, and tended to have lower state anxiety after the AAT visit. The results of the
current study are difficult to compare with previous studies, since medical fear and anxiety
were assessed after AAT and the comparison visits. Other studies did not include evaluation
after interventions. It is important to evaluate whether these scales are sufficiently discrimi-
natory, such that scores would change with moderate relaxation or distraction over a short
255

period of time.
 The Effect of Animal-Assisted Therapy on Stress Responses

The results of the psychological stress responses indicated that the childrens level of med-
ical fear after the AAT visit tended to be higher than after the comparison visit. There was no
significant difference in state anxiety after AAT and comparison visits. However, since fear and
anxiety were measured only after the visits, it is not clear whether these effects were due to
the visits or to the situations and procedures that were conducted during the hospitalization.
There is no literature on the effect of AAT on stress/medical fear or anxiety in children.
As an exploratory piece of research, this study has a number of limitations. The conven-
ience sampling, the small sample size, and the inability to randomize assignment limits gen-
eralizeability to all hospitalized children. A power analysis indicates that a sample of 40 children
was required. Difficulties in recruiting participants due to the few children hospitalized for two
days without the exclusion criteria and without competing care demands and schedules of the
AAT volunteers, made it impossible to obtain a larger sample. In one three-month period, no
eligible children were admitted to any of the three participating hospitals. Post-hoc power
analysis reveals a power less than 0.30. Thus, the lack of significant findings on several meas-
ures may be due to type II error and does not indicate that the presence of the dog had no ef-
fect. The post-test-only design for the medical fears and anxiety scales was particularly
problematic, in light of the inability to randomize children to the different orders. By continuing
research, we can evaluate the use of AAT in this area as well. Replication of this exploratory
study with a larger sample size, with a randomized design, and across various different age
groups of children is needed.
This study has demonstrated that AAT is more effective than a visit by a person at
alleviating some signs of physiological stress in inpatient pediatric settings. This occurred
even though children in this study had very low hospitalization stress and anxiety. Greater ef-
fects are expected in more stressful situations. Distraction and stress relief are the primary
benefits of AAT intervention. Thus, the use of AAT to alleviate stress during hospitalization is
an acceptable intervention that nurses or other health professionals can provide to meet the
physiological and psychological needs of pediatric patients. Animal-assisted therapy also
may be of use in outpatient and rehabilitation settings. In addition to its effect in pediatric
patients, AAT may provide physiological and psychological benefits for the patients families
and pediatric nurses.
The good news is that hospitalization was not as stressful for the children who participated
in this study as might be expected based on previous reports. Future research focused on pe-
diatric patients who are experiencing distress during hospitalization are likely to demonstrate
the effectiveness of AAT more definitively.
In general, it may be accepted that hospitalized children who interact with a dog may
benefit from improved physical and psychological health experiences. Therefore, this study
lends some support to the therapeutic benefit of AAT. Through practice and such research, pe-
diatric nurses and other health professionals are aware that AAT may be a useful intervention
which can play a part in decreasing the stress in hospitalized children. Pediatric nurses and
other health professionals can begin to analyze the exact relationship between AAT and
Anthrozos

hospitalized childrens stress status.

Acknowledgements
Thanks to Shady Grove Adventist Hospital, Mt. Washington Pediatric Hospital, University of
Maryland Medical Center, the participating hospitals, and to the child-life and dog visitation
256

teams for their assistance with data collection.

 Tsai et al.

References
Astin, E. W. 1977. Self-reported fears of hospitalized and nonhospitalized children aged ten to twelve. Maternal
Child Nursing Journal 6: 1724.
Barba, B. E. 1995. The positive influence of animals: Animal-assisted therapy in acute care. Clinical Nurse
Specialist 9: 199202.
Bardill, N. and Hutchinson, S. 1997. Animal-assisted therapy with hospitalized adolescents. Journal of Child and
Adolescent Psychiatric Nursing 10: 1724.
Barker, S. B. and Dawson, K. S. 1998. The effects of animal-assisted therapy on anxiety ratings of hospitalized
psychiatric patients. Psychiatric Services 49: 797801.
Barker, S. B., Pandurangi, A. K. and Best, A. M. 2003. Effects of animal-assisted therapy on patients anxiety,
fear, and depression before ECT. Journal of ECT 19: 3844.
Baun, M. M., Oetting, K. and Bergstrom, N. 1991. Health benefits of companion animals in relation to the
physiologic indices of relaxation. Holistic Nursing Practice 5: 1623.
Bossert, E. 1994. Stress appraisals of hospitalized school-age children. Child Health Care 23: 3349.
Bouchard, F., Landry, M., Belles-Isles, M. and Gagnon. J. 2004. A magical dream: A pilot project in animal-
assisted therapy in pediatric oncology. Canadian Oncology Nursing Journal 14: 1417.
Boyce, W. T., Chesney, M., Alkon, A., Tschann, J. M., Adams, S., Chesterman, B., Cohen, F., Kaiser, P., Folkman,
S. and Wara, D. 1995. Psychobiologic reactivity to stress and childhood respiratory illnesses: Results of two
prospective studies. Psychosomatic Medicine 57: 411422.
Brodie, S. J. and Biley, F. C. 1999. An exploration of the potential benefits of pet-facilitated therapy. Journal of
Clinical Nursing 8: 329337.
Broome, M. E., Bates, T. A., Lillis, P. P. and McGahee, T. W. 1990. Childrens medical fears, coping behaviors,
and pain perceptions during a lumbar puncture. Oncology Nursing Forum 17: 361367.
Broome, M. E. and Hellier, A. P. 1987. School-age childrens fears of medical experiences. Issues in
Comprehensive Pediatric Nursing 10: 7786.
Broome, M. E., Hellier, A., Wilson, T., Dale, S. and Glanville, C. 1988. Measuring childrens fears of medical
experiences. In Measurement of Nursing Outcome: Measuring Client Outcomes, 201214, ed. O. L. Strick-
land. New York: Springer.
Broome, M. E., Lillis, P. P., McGahee, T. W. and Bates, T. A. 1992. The use of distraction and imagery with
children during painful procedures. Oncology Nursing Forum 19: 499502.
Calvert, M. M. 1989. Humanpet interaction and loneliness: A test of concepts from Roys adaptation model.
Nursing Science Quarterly 2: 194202.
Churchill, M., Safaoui, J., McCabe, B. W. and Baun, M. M. 1999. Using a therapy dog to alleviate the agitation
and desocialization of people with Alzheimers disease. Journal of Psychosocial Nursing and Mental Health
Services 37: 1622.
Cole, K. M. 1999. Animal-assisted therapy. In Nursing Interventions: Effective Nursing Treatments, 508519, ed.
G. M. Bulechek and J. C. McCloskey. Philadelphia, PA: W. B. Saunders Company.
Cole, K. M. and Gawlinsk, A. 2000. Animal-assisted therapy: The humananimal bond. AACN Clinical Issues
11: 139149.
Connor, K. and Miller, J. 2000. Help from our animal friends. Nursing Management 31: 4246.
DeFrances, C. J. and Podgornik, M. N. 2006. 2004 National hospital discharge survey.Vital and Health Statistics
371: 120.
Erickson, F. 1972. Stress in the pediatric ward. Maternal Child Nursing Journal 1: 113116.
Friedmann, E. 1995. The role of pets in enhancing human wellbeing: Physiological effects. In Waltham Book of
HumanCompanion Animal Interactions, 3354, ed. I. Robbins. Oxford, UK: Pergamon.
Friedmann, E., Katcher, A. H., Thomas, S. A., Lynch, J. J. and Messent, P. R. 1983. Social interaction and
blood pressure: Influence of animal companions. Journal of Nervous and Mental Disease 171: 461465.
Anthrozos

Friedmann, E. and Son, H. 2009. The humancompanion animal bond: How humans benefit. Veterinary Clinics
of North America: Small Animal Practice 39: 293326.
Friedmann, E., Thomas, S. A., Cook, L. K., Tsai, C. C. and Picot, S. J. 2007. A friendly dog as potential
moderator of cardiovascular response to speech in older hypertensives. Anthrozos 20: 5163.
Friedmann, E., Thomas, S. A. and Eddy, T. J. 2000. Companion animals and human health: Physical and
cardiovascular influences. In Companion Animals & Us, 125142, ed. A. Podberscek, E. S. Paul and J. A.
257

Serpell. Cambridge: Cambridge University Press.

 The Effect of Animal-Assisted Therapy on Stress Responses

Friedmann, E., Thomas, S. A., Kulick-Ciuffo, D., Lynch, J. J. and Suginohara, M. 1982. The effects of normal
and rapid speech on blood pressure. Psychosomatic Medicine 44: 54553.
Friedmann, E. and Tsai, C. C. 2006. The animalhuman bond: Health and wellness. In Handbook on Animal-
Assisted Therapy, 95117, ed. A. H. Fine. San Diego, CA: Elsevier.
Hansen, K. M., Messinger, C. J., Baun, M. M. and Megel, M. 1999. Companion animals alleviating distress in
children. Anthrozos 12: 142148.
Hart, D. and Bossert, E. 1994. Self-reported fears of hospitalized school-age children. Journal of Pediatric
Nursing 9: 8390.
Havener, L., Gentes, L., Thaler, B., Megel, M. E., Baun, M. M., Driscoll, F. A., Beiraghi, S. and Agrawal, S. 2001.
The effects of a companion animal on distress in children undergoing dental procedures. Issues in
Comprehensive Pediatric Nursing 24: 137152.
Jorgenson, J. 2002. Animal-assisted therapy. In Complementary Alternative Therapies in Nursing,152162, ed.
M. Snyder and R. Lindquist. New York: Springer Publishing Company.
Kaminski, M., Pellino, T. and Wish, J. 2002. Play and pets: The physical and emotional impact of child-life and
pet therapy on hospitalized children. Childrens Health Care 31: 321335.
Kaplan, P. S. 2000. A Childs Odyssey. Belmont, CA: Wadsworth.
Kendall, P. C. and Ronan, K. R. 1990. Assessment of childrens anxieties, fears, and phobias: Cognitive
behavioral models and methods. In Handbook of Psychological and Educational Assessment of Children,
223244, ed. R. W. Kamphaus. New York: Guilford Press.
Lynch, J. J., Long, J. M., Thomas, S. A., Malinow, K. L. and Katcher, A. H. 1981. The effects of talking on the
blood pressure of hypertensive and normotensive individuals. Psychosomatic Medicine 43: 2533.
Lynch, J. J., Thomas, S. A., Long, J. M., Malinow, K. L., Chickadonz, G. and Katcher, A. H. 1980. Human
speech and blood pressure. Journal of Nervous and Mental Disease 168: 526534.
Lynch, J. J., Thomas, S. A., Long, J. M., Malinow, K. L., Friedmann, E. and Katcher, A. H. 1982. Blood pressure
changes while talking. Israel Journal of Medical Sciences 18: 575579.
Mahat, G. and Scoloveno, M. 2003. Comparison of fears and coping strategies reported by Nepalese school-
age children and their parents. Journal of Pediatric Nursing 18: 305313.
Miller, J. and Ingram, L. 2000. Perioperative nursing and animal-assisted therapy. AORN Journal 72: 477483.
Murphy, J. K. and McGarvey, S. T. 1994. Modernization in the Samoas and childrens reactivity: A pilot study.
Psychosomatic Medicine 56: 395400.
Muschel, I. J. 1984. Pet therapy with terminal cancer patients. Journal of Social Casework 65: 451458.
Nagengast, S. L., Baun, M. M., Megel, M. and Leibowitz, J. M. 1997. The effects of the presence of a companion
animal on physiological arousal and behavioral distress in children during a physical examination. Journal of
Pediatric Nursing 12: 323330.
Nutter, D. A., Larsen, L. H. and Sylvester, C. 2006. Anxiety Disorder: Generalized Anxiety, eMedicine.
<http://emedicine.medscape.com/article/916933- overview> Accessed on September 18, 2006.
Park, M. K. and Menard, S. M. 1987. Accuracy of blood pressure measurement by the Dinamap monitor in
infants and children. Pediatrics 79: 907914.
Ramesy, M. 1979. Noninvasive automatice determination of mean arterial pressure. Medical Biological
Engineering Computer 17: 1118.
Scavnicky-Mylant, M. 1987. The hospitalized school-age childs capacity for an appraisal of threat. Western
Journal of Nursing Research 9: 503526.
Sobo, E. J., Eng, B. and Kassity-Krich, N. 2006. Canine visitation (pet) therapy: Pilot data on decreases in child
pain perception. Journal of Holistic Nursing 24: 5157.
Spielberger, C. D. 1970. State-Trait Anxiety for Children. Palo Alto, CA: Consulting Psychologists Press.
Spielberger, C. D., Edwards, C. D., Lushene, R., Monturoi, J. and Platzek, D. 1973. State-Trait Anxiety Inventory
for Children: Sampler Set: Manual, Test Boolet, Scoring Key. Palo Alto, CA: Mind Garden.
Stanley-Hermanns, M. and Miller, J. 2002. Animal-assisted therapy. American Journal of Nursing 102: 6976.
Anthrozos

Thomas, S. A. and Friedmann, E. 1994. The cardiovascular effects of rate of verbal communication. Journal of
Cardiovascular Nursing 9: 1626.
Tiedeman, M. E. and Clatworthy, S. 1990. Anxiety responses of 5- to 11-year-old children during and after
hospitalization. Journal of Pediatric Nursing 5: 334343.
Wu, A. S., Niedra, R., Pendergast, L. and McCrindle, B. W. 2002. Acceptability and impact of pet visitation on
a pediatric cardiology inpatient unit. Journal of Pediatric Nursing 17: 354362.
258

Yelderman, M. and Ream, A. K. 1979. Indirect measurement of mean blood pressure in the anesthetized patient.
Anesthesiology 50: 253256.