Anthrozoös

A multidisciplinary journal of the interactions of people and animals

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

Animal-Assisted Therapy in the Treatment of

Substance Dependence

Martin C. Wesley, Neresa B. Minatrea & Joshua C. Watson

To cite this article: Martin C. Wesley, Neresa B. Minatrea & Joshua C. Watson (2009) Animal-
Assisted Therapy in the Treatment of Substance Dependence, Anthrozoös, 22:2, 137-148, DOI:
10.2752/175303709X434167

To link to this article: https://doi.org/10.2752/175303709X434167

Published online: 28 Apr 2015.

Submit your article to this journal

Article views: 1546

View related articles

Citing articles: 10 View citing articles

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

https://www.tandfonline.com/action/journalInformation?journalCode=rfan20
AZ VOL. 22(2).qxp:Layout 1
ANTHROZOÖS
Address for correspondence:
Dr. Martin Wesley,
School of Professional
Counseling,
Lindsey Wilson College,
210 Lindsey Wilson Street,
Columbia, KY 42728, USA.
E-mail: wesleym@lindsey.edu
3/31/09 10:43 PM Page 137
VOLUME 22, ISSUE 2 REPRINTS AVAILABLE PHOTOCOPYING © ISAZ 2009
PP. 137–148 DIRECTLY FROM PERMITTED PRINTED IN THE UK
THE PUBLISHERS BY LICENSE ONLY
Animal-Assisted Therapy
in the Treatment
of Substance Dependence
Martin C. Wesley,* Neresa B. Minatrea† and
Joshua C. Watson‡
*School of Professional Counseling, Lindsey Wilson College, Kentucky,
USA
† Department of Counseling and Student Affairs, Western Kentucky
University, USA
‡ Department of Counselor Education, Mississippi State University, USA
ABSTRACT The purpose of this study was to evaluate the effect of animal-
assisted therapy (AAT) on the therapeutic alliance with an adult, residential, sub-
stance abuse population in group therapy. We used randomized samples and
controlled conditions to establish the effects of chosen variables that influence
outcome. A total of 231 people took part in the study (control group [without
therapy dog]: n = 96; experimental group [with therapy dog]: n = 135) in which
there were 26 group sessions. The results of the study indicate that, overall, the
therapeutic alliance is enhanced with the addition of a therapy dog: the AAT
group had a more positive opinion of the therapeutic alliance, as measured using
the Helping Alliance Questionnaire (HAQ-II), than the control group (ANOVA:
F(1,229) = 25.44, p < 0.001). Clients seeking treatment for a dual diagnosis, clients
Anthrozoös DOI: 10.2752/175303709X434167
with state social service involvement, and clients seeking treatment for alcohol ad-
diction had similar opinions of the therapeutic alliance, whether in the experi-
mental group or the control group. Males, females, pet owners, court ordered
clients, and clients seeking treatment for polysubstance, cannabis, and metham-
phetamine dependence all were more positive about the therapeutic alliance if
they were in the experimental group than if they were in the control. This study
demonstrates that addiction professionals could increase treatment success by
adding this complementary, evidence-based practice.
Keywords: animal-assisted therapy, drug abuse, group counseling, sub-
stance abuse treatment, therapeutic alliance
❖
The purpose of this study was to evaluate the effect of animal-
assisted therapy (AAT) on the therapeutic alliance with an adult,
residential, substance abuse population in group therapy. Sub-
137
stance abuse treatment is in need of complementary evidence-based
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 138

Animal-Assited Therapy in the Treatment of Substance Dependence

treatment techniques and options that enhance the therapeutic alliance and reduce stress in
therapy. In 2003, there were 1.84 million individual substance abuse treatment admissions
(SAMHSA 2006a). In the same year, 28% of clients in treatment left against professional ad-
vice and the facility terminated an additional 9%. Only 44% successfully completed treatment
(SAMHSA 2006b). As these figures indicate that treatment for alcohol and drug dependence
can be less than effective (Cuttler and Fishbain 2005), new treatment strategies are needed
to help with client retention in treatment and increase treatment efficacy.
The therapeutic alliance is a complex phenomenon of the collaborative relationship or bond
that is formed between the client and the therapist. It is a dynamic process that is influenced
by the developmental histories of both the client and the therapist. Factors such as gender,
race, age, and drug use attitudes may impinge on the therapeutic alliance (Meissner 1996). A
positive therapeutic alliance determines treatment success (Meier et al. 2006). Studies have
consistently found that the quality of the therapeutic alliance relates to treatment outcome and,
in fact, can be the most important variable contributing to therapeutic success or failure
(Orlinsky, Ronnestad and Wilutzki 2003).
Sobriety is the desired therapeutic outcome for a substance-abusing client seeking re-
covery. The problem in all addiction treatment and research has been how to get clients to
attach to, and stay in, treatment long enough for a successful outcome. Studies confirm that
enhancing the therapeutic alliance can increase sobriety rates for clients seeking treatment
for their substance abuse problem (Miller, Taylor and West 1980; Connors et al. 1997).
Over the past 40 years, AAT has moved from being an unassuming, obscure therapy to
becoming a successful, research-based therapy (Hines 2003). The professional community
overlooked AAT as a viable therapy for years (Arkow 1993); however, it is now a healthy and
positive experience for most populations. AAT is not a specific technique; rather, the therapy
dog or other animal is an adjunct to the treatment process, helping to engage the client in the
therapeutic program (Fine 2006).
AAT is an established strategy for enhancing the therapeutic alliance with specific popula-
tions (Lefkowitz 2005) and reducing stress (Allen et al. 2001). However, the literature reveals
very little about how AAT can contribute to the treatment of individuals suffering from chem-
ical dependency. Since most animal–human interaction research is based on anecdotal
success or limited or poor data (Barba 1995; Garrity and Stallones 1998; Wilson and Barker
2003), good, empirical studies are needed to confirm the results. Fine and Mio (2006) recently
posed the question of whether a trained therapy animal can enhance the therapeutic alliance.
In their call for additional research, they suggested that “the gold standard for research of this
type is to have a random assignment to an AAT format and a traditional therapy format and
then to determine which format led to a better therapeutic outcome” (p. 517). The present
quantitative study is a direct result of these recommendations.
AAT can strengthen the therapeutic alliance and reduce stress in several ways. One way
to use the therapy animal is as an interest topic to engage the client in verbal interactions.
Moreover, the animal can reduce the client’s anxiety about being in the therapist’s office or in
Anthrozoös

a group setting, making the client less resistant to the therapeutic intervention. The client seems
to trust the therapy animal well before they trust the human therapist (Chandler 2005).
Additionally, if the client bonds with the animal as an adjunct to the therapeutic team, the client
may be more committed to completing his or her therapeutic goals (Cieslak 2001).
Another benefit from incorporating AAT is that it provides the client a surrogate for
138

therapeutic touch (Chandler 2005). Often it is difficult for a therapist to know when touching is
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 139

Wesley et al.

appropriate and so they usually avoid this interaction. Touching may be misconstrued as com-
forting behavior or may accidentally sexually provoke the client, leading to possible incompe-
tency charges (Koocher and Keith-Spiegel 1998). Porous boundaries found with many
addiction therapists, who may be in recovery themselves, intensify this concern (Bissell and
Royce 1987). Therapy dogs are safe for therapist and client and can introduce physical touch
without breaking ethical boundaries. They also introduce therapeutic touch to residential
patients, who may not have social support from family or friends.
A further benefit from integrating animals into counseling involves physiological changes.
Physiological indicators of stress and anxiety such as heart rate and blood pressure have been
shown to decrease when a therapy animal is present (Baun et al. 1984; Nagengast et al. 1997).
According to one study by Odendaal (2000), six neurochemicals associated with a decrease in
blood pressure were positively associated with animal interaction. Psychological stress can con-
tribute to drug and alcohol relapse (Brady and Sonne 1999). Haleem (1996) states that stress-
ful experiences can reliably induce a relapse to drugs or alcohol. Traditional confrontational group
therapy in drug and alcohol treatment centers has the potential for increasing stress levels and
intensifying the chance of relapse, thus resulting in the therapeutic relationship being damaged
(Miller, Rollnick and Conforti 2002). Clients in group therapy are free to express themselves ver-
bally when psychological stress is reduced (Yalom 2005). In a therapeutic milieu, a therapy dog
has the capacity to enhance the therapeutic environment, calm the client and foster healthy at-
tachments. The therapy animal seems to lower stress and anxiety often associated with being
in group therapy and residential treatment, including those who are court ordered or came to the
residential treatment setting from jail or prison (Bardill and Hutchinson 1997; Campbell-Berg
2000). A person coping with the stress of compulsory, residential or inpatient treatment may feel
like his or her world is tumbling down around them, yet the therapy animal remains constant.
Social support theory (Hupcey 1998) and attachment theory (Ainsworth 1989; Flores 2004)
formed the basis for the theoretical perspective of the present project. These theories stress
the positive health effect of human social companionship and were easily adaptable to AAT be-
cause animals are also a source of social support and attachment for many individuals. Many
people consider their pet as a “member of the family” or a private confidant. Companion ani-
mals can also increase the frequency of human social contact (Serpell 2002).
Reviewing and developing mechanisms to achieve healthy attachments, improve thera-
peutic alliance, while decreasing stress, consequently would improve the therapeutic experi-
ence for those going through treatment for chemical dependency. The integration of AAT in a
counseling setting targeting individuals abusing or addicted to chemicals merits further inves-
tigation. The literature review aforementioned indicates that a therapy animal appears to
enhance the therapeutic climate (Cieslak 2001; Chandler 2005), facilitate appropriate touch
(Chandler 2005), and reduce stress (Baun et al. 1984; Nagengast et al. 1997); however,
empirical evidence does not exist to support its use with a substance abuse population.
Therefore, the purpose of this study was to test the effectiveness of AAT with an adult, resi-
dential, substance abuse population in group therapy.
Anthrozoös

Research Questions and Hypothesis

This study sought to answer how AAT affects the therapeutic alliance, as measured using the
Helping Alliance Questionnaire (HAQ-II). We hypothesized that clients in an AAT group session
would show higher ratings of the therapeutic alliance, compared with clients in a group therapy
139

session which had no therapy dog.

AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 140

Animal-Assited Therapy in the Treatment of Substance Dependence

Methods
Therapy Dog and Setting
The certified therapy animal for this study was a dog, a Beagle-cross named Mitzi Ann. Her
handler and owner, a doctoral level therapist, conducted the group sessions for this study. At
the time of this study, the dog weighed 27 pounds, stood 17 inches high at the shoulder and
was about seven years of age (she came from an animal shelter, so her exact age was not
known). Mitzi Ann quickly demonstrated her skills by providing comfort, service, and nurture
to her owner/handler during her own struggles with ovarian cancer. To enhance this training,
Mitzi Ann completed three ten-week sessions of advanced obedience training, followed by
three eight-week sessions of agility training. Mitzi Ann first earned her international registration
as a Delta Pet Partner in July 2004. Delta Society provides Pet Partner registration to animals
and their handlers for AAT. Delta’s test examiners qualified her to work with any mental or med-
ical condition; all age groups; and within any environmental settings (e.g., intensive care units,
hospitals, prisons, schools, physical rehabilitation units, nursing homes, treatment centers,
group homes). At the time of this study, Mitzi Ann and her handler had provided approximately
157 hours of service work integrating counseling with AAT. Furthermore, Mitzi Ann’s multiple
trainings furnished a firm foundation for utilizing her in this project.
Participants
The research facility was located within the city limits of Bowling Green, Kentucky, USA, which
has a population of about 50,000. The facility was a division of the larger community mental
health agency entrusted to provide services to the indigent population in a ten-county area of
South Central Kentucky. Participants in this research study (a) were enrolled in residential treat-
ment at the designated facility during the research phase of this research project, (b) had a
diagnosis of substance dependence, (c) signed a written consent form, (d) were able to under-
stand and communicate in English, and (e) had been able to attend and participate in group
therapy for approximately one hour. The participants eligible for this study appeared both phys-
ically and emotionally stable, and presented no risk of harm to themselves, others, or the dog.
There were 231 participants and 26 group sessions. Fourteen groups were conducted for
both the control and experimental groups. Two sessions from the control group were removed
from the analysis due to contaminating factors: one group session was interrupted by the fa-
cility fire alarm sounding; the second group was removed because the therapy dog entered
the group setting by accident. This resulted in uneven group and individual numbers, but this
did not influence the results. There were 135 individuals in the experimental group (with the
therapy dog) and 96 individuals in the control group (without the therapy dog). Although some
intervening variables may have influenced the results of this study, they were routine for the
population and setting and were not disproportionate to the control or experimental groups.
These potential variables included facility emergencies, family problems, and weather changes.
Males represented 49.4% of the participants, while females represented 50.6%, making
for a balanced sample. The sample was 90.5% Caucasian, with African Americans represent-
Anthrozoös

ing the remaining percentage. Over 50% of the participants were under the age of 25 years and
none of the participants were over the age of 55. The two study groups were selected through
drawing numbers (which had been randomly assigned to individual participants) from a hat.
Table 1 gives full background details of the participants in each group.
The participants in this study were all seeking treatment for substance dependency in a
140

residential treatment facility. All of the study participants were diagnosed with substance
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 141

Wesley et al.

Table 1. Background details of the participants.

Background Variable Experimental Group Control Group
(n = 135) (n = 96)
Age (yrs)
18–25 62 (45.9%) 54 (56.3%)
26–35 14 (10.4%) 41 (42.7%)
36–45 47 (34.8%) 1 (1.0%)
46–55 12 (8.9%) 0 (0%)
Gender
Male 70 (51.9%) 44 (45.8%)
Female 65 (48.1%) 52 (54.2%)
Race/Ethnicity
White/Caucasian 124 (91.9%) 85 (88.5%)
Black/African American 11 (8.1%) 11 (11.5%)
Marital Status
Single 45 (33.3%) 70 (72.9%)
Married 21 (15.6%) 0 (0%)
Living Together (unmarried) 14 (14.6%) 3 (3.1%)
Separated (not divorced) 8 (5.9%) 0 (0%)
Divorced 47 (34.8%) 23 (24.0%)
Sexual Orientation
Heterosexual 132 (97.8%) 78 (81.3%)
Homosexual or Bisexual 3 (2.2%) 18 (18.7%)
Have Children 88 (65.2%) 72 (75.0%)
Highest Level of Education
Grade School 3 (2.2%) 13 (13.5%)
High School 100 (74.1%) 83 (86.5%)
Trade School 32 (23.7%) 0 (0%)
Family Annual Income
Under $10,000 64 (47.4%) 72 (75.0%)
$10,000–$20,000 21 (16.6%) 3 (3.2%)
$20,001–$30,000 41 (34.4%) 3 (3.2%)
$30,001–$40,000 3 (2.2%) 18 (18.7%)
$40,001–$50,000 6 (4.4%) 0 (0%)
Court Ordered for Treatment 99 (73.0%) 46 (48.0%)
Drug Court Involvement 2 (1.5%) 20 (20.9%)
Probation or Parole 38 (28.1%) 69 (71.9%)
Social Service Involvement 55 (40.7%) 23 (24.0%)
Anthrozoös

dependency to at least one substance; about 69% were seeking treatment for more than
one substance. Over 50% of the participants were seeking treatment for alcohol (55%)
and/or methamphetamines (57.1%) dependency. Table 2 gives a full account of the
141

substance abuse characteristics of the participants. The sample also represented clients
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 142

Animal-Assited Therapy in the Treatment of Substance Dependence

wanting treatment for mental health problems; more than 59% of the clients were seeking
treatment for a dual diagnosis. Table 3 gives a full account of the mental health character-
istics of the participants.
Finally, the participants provided varying characteristics and attitudes related to animals
and pets. Almost 70% of the them either currently owned or had owned a pet within the last
year. All of the clients indicated that they did not hate animals, but all viewed pet ownership
differently. Just over 85% of the participants believed that the primary purpose of pet
ownership was for companionship, while others viewed the primary purpose as recreation
or security and safety. Table 4 gives an account of the pet history and attitude characteristics
of the participants.

Table 2. Substance abuse characteristics of the participants.

Type of Substance Abuse Experimental Group Control Group
(n = 135) (n = 96)
Alcohol 85 (63.0%) 42 (43.8%)
Marijuana 70 (51.9%) 44 (45.9%)
Methamphetamine 85 (63.0%) 47 (49.0%)
Heroin 9 (6.7%) 11 (11.5%)
Cocaine 29 (21.5%) 41 (42.7%)
Rx Opiates 45 (33.3%) 29 (30.2%)
Benzodiazepines 9 (6.7%) 11 (11.5%)
Hallucinogens 1 (0.08%) 11 (11.5%)
More than One Substance 98 (72.6%) 63 (65.6%)

Table 3. Mental health characteristics of the participants.

Mental Health Issues Experimental Group Control Group
(n = 135) (n = 96)
Dual Diagnosis 71 (52.6%) 65 (67.7%)
Mood Disorders 43 (31.9%) 53 (55.2%)
Anxiety Disorders 48 (35.6%) 54 (56.6%)
Abuse History 9 (6.7%) 11 (11.5%)
Social/Relational Problems 23 (17.0%) 46 (47.9%)
Anger Issues 14 (10.4%) 34 (35.4%)

Table 4. Pet history and attitudes of the participants.

Pet History and Attitudes Experimental Group Control Group
(n = 135) (n = 96)
Anthrozoös

Current Pet Owner or Owner in Past Year 114 (84.4%) 45 (46.9%)

Lost a Pet in the Last Year 36 (22.7%) 3 (3.1%)
Primary Purpose of Pet Ownership
Companionship 121 (89.6%) 76 (79.1%)
Safety and Security 1 (1.0%) 9 (9.4%)
Recreation 13 (9.4%) 11 (11.5%)
142
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 143

Wesley et al.

Study Procedures
Participants were oriented to the study prior to participation. Orientation included answering
questions and reviewing forms and questionnaires, including limits of confidentiality. After this
step, participant randomization occurred through the drawing of names from a hat. We at-
tempted to design, conduct, and report the findings in a manner consistent with the Ameri-
can Counseling Association’s (ACA) ethical standards for research and publication. Observed
standards included prior Institutional Review Board (IRB) approval for the research study,
which included instrumentation, methodology, and data analysis. Other observed standards
included a comprehensive informed consent for the clients, random assignment of partici-
pants, and confidentiality of information, complying with all HIPPA regulations, and federal
and state laws.
The experimental and control group sessions were conducted in the mornings, consistent
with the agency’s policy and schedule. After each group session, participants from both groups
completed the Helping Alliance Questionnaire (HAQ-II). Each group session lasted approxi-
mately one hour, with a 15-minute break for the therapist between each group. In total, there
were 26 group sessions and these were conducted over a period of three weeks.
Choice theory (Glasser 1998) provided the therapeutic philosophy of the cognitive-
behavioral therapy used by the therapist for the experimental and control groups. The group
therapy sessions had specific treatment goals to change the client’s movement along the
stages of change (Prochaska, DiClemente and Norcross 1993). The goals may have varied
depending on the specific group and individual needs. The therapist used her skills as an ex-
perienced therapist to address many of these specific goals in the process group by using
specific techniques based on her chosen theory and the needs of the client and group at the
time of treatment.
The specific AAT techniques used in this study were based on group and individual treat-
ment goals, as outlined by Gammonley et al. (1997). They outline numerous AAT techniques
in meeting treatment goals and these were utilized in this study. The therapy dog used in this
study participated in training involving techniques that were conducive to specific group or in-
dividual treatment goals. For example, the therapy dog removed tissues from the tissue box
and presented them to any participant who cried or sneezed. The therapy dog had a reper-
toire of about 30 “tricks,” used as psycho-educational tools for the therapist within the group.
The dog also naturally moved from one group participant to another, making herself available
for physical touch. In addition, all her “tricks” could be completed as interactive activities with
the participants. The dog was available when called, was nonjudgmental, and was predictable
in her responses.
Instrumentation
The first author of this paper developed a questionnaire to obtain background information
about the participants, such as gender, race, and drug of choice. This form also consisted of
fill-in-the-blank, multiple choice and short-answer questions. In addition, the Pet Attitude Scale
Anthrozoös

(PAS) (developed by Templer et al. 1981), which is an 18-item Likert style questionnaire used
to measure favorableness toward pets, was included. Statistical measures include a Cron-
bach’s alpha of 0.91 and test-retest reliability of 0.92. Construct validity was established by
principal-component factor analysis and correlated with four different personality instruments
in the construction and validation study. The PAS consists of three primary factors labeled (1)
143

love and interaction, (2) pets in home, and (3) joy of pet ownership. An individual can score
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 144

Animal-Assited Therapy in the Treatment of Substance Dependence

between 38 and 266, with higher scores indicating a more favorable attitude toward pets. The
responses from this questionnaire were used to determine participant appropriateness for this
study. Clients overly fearful of animals or who abused animals in the past were excluded from
the study.
The Helping Alliance Questionnaire Revised (HAQ-II) developed by Luborsky et al. (1996)
was used to evaluate the therapeutic alliance. This 19-item questionnaire measures the strength
of the client–therapist alliance. Each item is rated on a 6-point Likert scale (1 = I strongly disagree
and 6 = I strongly agree); the instrument’s authors reversed the negatively worded items. This in-
strument has a high internal consistency (above 0.90 for both patient and therapist versions at
sessions 2, 5 and 24), strong test-retest reliability (0.79) on the patient version, and moderate to
high convergent validity with the California Psychotherapy Alliance Scale (CALPAS). According
to Duncan, et al. (2004), the HAQ-II is the gold standard of alliance self-report scales.
Data Analysis
We made comparisons of the mean differences between groups with and without the therapy
dog, using a one-way analysis of variance (ANOVA). A significance level of p < 0.05 was adopted.

Results
After each group session (with or without the therapy dog present), participants rated their in-
terpretation of the therapeutic alliance by completing the HAQ-II questionnaire. A significant dif-
ference was found between mean HAQ-II ratings (F(1,229) = 25.44, p < 0.001, 2 = 0.100), with
those in the AAT groups reporting a more positive opinion of the therapeutic alliance. A review
of the effect sizes calculated for the variable suggests that the addition of AAT contributed most
to the differences observed for mean HAQ-II ratings (see Figure 1).

116

114

112
Mean HAQ-II Score

110

108

106

104

102
Anthrozoös

0
100
Experimental Control
Group Group
Figure 1. Mean HAQ-II total scores for the AAT experimental group (n =
135) and the control group (n = 96). Error bars represent 95% confidence
144

intervals.
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 145

Wesley et al.

Demographic subgroups, for which there was enough statistical power, were identified for
further examination of the therapeutic alliance. Analysis was conducted to determine if AAT af-
fected the therapeutic alliance, as measured on the Helping Alliance Questionnaire (HAQ-II),
in these subgroups. The demographic population subgroups further tested were: a) gender,
b) court ordered clients, c) clients with social service (Child Protective Service) involvement, d)
pet owners, e) dual diagnosis clients, and f) drug dependence: polysubstance, alcohol,
marijuana, and methamphetamines.
The results of this inquiry were mixed. All population subgroups indicated significant differ-
ences between the AAT and control groups, except for dual diagnosis clients (F(1,134) = 0.130,
p > 0.719); clients with social service (child protective service) involvement (F(1,76) = 0.062,
p > 0.804); and clients seeking treatment for alcohol (F(1,125) = 2.91, p > 0.091). For the remain-
ing population subgroups tested, it was found that clients in the AAT groups reported a more
positive opinion of the therapeutic alliance than did individuals in the groups without the therapy
dog: a significant difference was found between mean HAQ-II ratings for males (F(1,112) = 34.62,
p < 0.000); females (F(1,115) = 4.75, p < 0.031); court ordered clients (F(1,143) = 37.16, p < 0.000);
pet owners (F(1,157) = 40.59, p < 0.000); polysubstance dependence clients (F(1,159) = 27.85,
p < 0.000); cannabis dependence clients (F(1,112) = 24.96, p < 0.000); and clients seeking
treatment for methamphetamine dependence (F(1,130) = 29.57, p < 0.000).

Discussion
This study revealed several significant findings. There was confirmation of the hypothesis
that the experimental group (with the therapy dog) would be significantly more positive
about the therapeutic alliance than the control group (without the therapy dog). Therefore
the therapeutic alliance is enhanced with the addition of a therapy dog within a group
setting with adult clients in a residential, drug abuse treatment setting. These positive re-
sults were also seen with the subgroup samples of males, females, court-ordered clients,
pet owners, and clients seeking treatment for polysubstance, cannabis, and metham-
phetamine dependence.
Increasing the therapeutic alliance can significantly improve the probability of recovery suc-
cess for clients seeking treatment for substance dependence. Our findings support those of
previous researchers (Duncan and Miller 2000; Orlinsky, Ronnestad and Wilutzki 2003) who
have advocated that the quality of the alliance and not the technique or theory is most pre-
dictive of recovery outcome success. Therefore a client receiving AAT is more likely to benefit
from the treatment process, have higher retention rates, and less frequent episodes of drug
and alcohol relapse (Meier et al. 2006).
Results regarding the sample subgroups proved varied. The study indicated that AAT is a
viable complementary therapy for both male and female clients; for clients seeking treatment
for polysubstance dependence, cannabis dependence, and methamphetamine dependence;
for clients who are court ordered into treatment; and for clients who own or have recently
owned pets. These clients are most likely to come to treatment under the worst of circum-
Anthrozoös

stances. Many have been living in shelters, on the streets, or have come directly from jail or
prison. Nearly all have felony convictions with charges pending. These clients have not had the
comforts of home for some time. They could perceive the therapy animal as a novelty to the
therapeutic process. The dog might remind them of better days and give them a sense of
145

home and normalcy.

AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 146

Animal-Assited Therapy in the Treatment of Substance Dependence

AAT was not found to be effective for clients seeking treatment for a dual diagnosis, clients
who were being investigated by social services (child protective services), and clients seeking
treatment for alcohol dependence. Clients seeking treatment for alcohol dependence reported
that AAT did improve the therapeutic alliance; however, the result was not significant. Interest-
ingly, AAT was a viable option for clients seeking treatment for other forms of dependence:
methamphetamine, cannabis, and polysubstance. Since drug-dependent clients and alcohol-
dependent clients often represent different age groups, this may account for the difference.
In contrast, the results for the dual-diagnosis clients and clients involved with social services
subgroups did not appear to benefit at all from the session of AAT in a group setting. The
reason for this is unknown. It is possible that clients with a dual diagnosis and clients with
social service or child protective service involvement require multiple AAT sessions before the
therapeutic alliance is improved.
Limitations
This study was limited to lower income, residential, treatment clients from South Central Ken-
tucky. While many factors may be generalized to other populations, the group participants
represented a homogeneous group. This included common demographic characteristics
specific to this area of South Central Kentucky.
Another limitation of this study concerns the knowledge by both groups of the therapist’s
love for animals. One reason for the increase in the therapeutic alliance could be the percep-
tion from the clients that the therapist was more caring and could be trusted more than some-
one who did not own or love animals. However, in this study the control group clients were
aware of the therapist’s love of animals and could have seen the therapy dog enter or exit the
facility with the therapist. This suggests that the control group may have had good opinions of
the therapeutic alliance and reduced stress levels simply because of this knowledge. Therefore,
the differences between groups could have potentially been greater if the therapist had con-
ducted the control group and experimental group sessions at different facilities. Unfortunately,
it was not possible to do this in the current study, and the change would have added new
variables which we would have had to have taken into account when analyzing the data.
This was an exploratory study investigating the effect of AAT for a substance abuse popu-
lation. The authors recommend continued research to validate the use of this therapy with sub-
stance abuse samples. Studies should further explore the efficacy of AAT with clients with a dual
diagnosis and clients with social service involvement. Researchers may want to explore why
AAT was effective for clients seeking treatment for illicit drugs but not for individuals seeking treat-
ment for alcoholism.
Recently, Fine and Mio (2006) called for additional empirical research evaluating the
effectiveness of AAT for various populations and settings. Our quantitative study followed these
recommendations by using randomized populations and controlled conditions to establish the
effects of chosen variables that influence outcome. This was an empirically sound foundational
study for the efficacy of AAT with a substance abuse population and for the enhancement of
Anthrozoös

the therapeutic alliance in a group treatment environment. The results of the study could have
significant implications for social change for the substance abuse treatment industry.

Conclusion
This study shows that AAT can significantly improve the quality of the therapeutic relationship,
146

which may result in higher treatment outcomes and reduced physiological indicators of
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 147

Wesley et al.

psychological stress. Residential treatment centers, and potentially outpatient treatment pro-
grams, can improve treatment outcome by an improved therapeutic alliance with a trained ther-
apy dog present in the group treatment. AAT can reduce stress and may improve retention and
motivation for participation in therapy, resulting in an increase in recovery success. AAT is a
viable option for addiction programs desiring positive change.

References
Ainsworth, M. D. S. 1989. Attachments beyond infancy. American Psychologist 44: 709–716.
Allen, K. M., Blascovich, J., Tomaka, K. and Kelsey, R. M. 2001. Presence of human friends and pet dogs as mod-
erators of autonomic responses to stress in women. Journal of Personality and Social Psychology 61: 582–589.
Arkow, P. 1993. Pet-Therapy: A Study and Resource Guide the Use of Companion Animals in Selected
Therapies. Colorado Springs, CO: The Humane Society of the Pikes Peak Region.
Barba, B. E. 1995. A critical review of the research on the human/companion animal relationship: 1988–1993.
Anthrozoös 8(1): 9–14.
Bardill, N. and Hutchinson, S. 1997. Animal-assisted therapy with hospitalized adolescents. Journal of Child and
Adolescent Psychiatric Nursing 10(1): 17–24.
Baun, M., Bergastrom, N., Langston, N. and Thoma, L. 1984. Physiological effects of human/companion animal
bonding. Nursing Research 15: 126–129.
Bissell, L. and Royce, J. E. 1987. Ethics for Addiction Professionals. 2nd edn. Center City, MN: Hazelden.
Brady, K. T. and Sonne, S. C. 1999. The role of stress in alcohol use, alcoholism treatment, and relapse. Alcohol
Research and Health 23(4): 263–271.
Campbell-Berg, T. 2000. A case study using animal-assisted therapy to promote abstinence in a group of indi-
viduals who are recovering from chemical addictions. Journal of Addictions Nursing 12(1): 31–35.
Chandler, C. K. 2005. Animal-Assisted Therapy in Counseling. New York: Routledge.
Cieslak, E. J. 2001. Animal-assisted therapy and the development of an early working alliance: The use of dogs
in therapy with young adults. Unpublished doctoral dissertation, The University of Wisconsin–Madison, USA.
Connors, G. J., DiClemente, C. C., Carroll, K. M., Longabaugh, R. and Donoban, D. M. 1997. The therapeutic
alliance and its relationship to alcoholism treatment participation and outcome. Journal of Consulting and
Clinical Psychology 65(4): 588–598.
Cuttler, R. B. and Fishbain, D. A. 2005. Are alcoholism treatments effective? The Project MATCH data. BMC
Public Health 5(75): 1–11.
Duncan, B. L. and Miller, S. D. 2000. The Heroic Client: Doing Client-Directed Outcome-Informed Therapy. San
Francisco: Jossey-Bass.
Duncan, B. L., Miller, S. D., Reynolds, L. R., Sparks, J. A., Claud, D. A., Brown, J., et al. 2004. The session rating
scale. <www.talkingcure.com/documents/TheSRSArticle.doc> Accessed January 10, 2007.
Fine, A. H. 2006. Handbook on Animal-Assisted Therapy: Theoretical Foundations and Guidelines for Practice.
2nd edn. San Diego, CA: Academic Press.
Fine, A. H. and Mio, J. S. 2006. The future of research, education, and clinical practice in the animal–human
bond and animal-assisted therapy: Part C: the role of animal-assisted therapy in clinical practice: the
importance of demonstrating empirically oriented psychotherapies. In Handbook on Animal-Assisted
Therapy: Theoretical Foundations and Guidelines for Practice. 2nd edn, 513–523, ed. A. Fine. San Diego,
CA: Academic Press.
Flores, P. J. 2004. Addiction as an Attachment Disorder. New York: Jason Aronson.
Gammonley, J., Howie, A. R., Kirwin, S., Zaph, S. A., Frye, J., Freeman, G., et al. 1997. Animal-Assisted Ther-
apy: Therapeutic Interventions. Renton, WA: Delta Society.
Garrity, T. F. and Stallones, L. 1998. Effects of pet contact. In Companion Animals in Human Health, 3–22, ed.
Anthrozoös

C. C. Wilson and D. C. Turner. Thousand Oaks, CA: Sage.

Glasser, W. 1998. Choice Theory: A New Psychology of Personal Freedom. New York: Harper Perennial.
Haleem, D. J. 1996. Adaptation to repeated restraint stress in rats: Failure of ethanol-treated rats to adapt in the
stress schedule. Alcohol and Alcoholism 31: 471–477.
Hines, L. M. 2003. Historical perspectives on the human–animal bond. American Behavioral Scientist 47(1): 7–15.
Hupcey, J. E. 1998. Clarifying the social support theory-research linkage. Journal of Advanced Nursing 27:
147

1231–1241.
AZ VOL. 22(2).qxp:Layout 1 3/31/09 10:43 PM Page 148

Animal-Assited Therapy in the Treatment of Substance Dependence

Koocher, G. P. and Keith-Spiegel, P. 1998. Ethics in Psychology: Professional Standards and Cases. 2nd edn.
New York: Oxford University Press.
Lefkowitz, C. M. 2005. Animal-assisted prolonged exposure: A treatment for survivors of sexual assault suffer-
ing with posttraumatic stress disorder (Doctoral dissertation, Widener University, Institute for Graduate Clinical
Psychology, USA). Dissertation Abstracts International B 66/06, 3415.
Luborsky, L., Barber, J., Siqueland, L., Johnson, S., Najavits, L., Frank, A. et al. 1996. The Revised Helping
Alliance Questionnaire (HAQ-II): Psychometric properties. The Journal of Psychotherapy Practice and
Research 5: 260–271.
Meier, P. S., Barrowclough, C. and Donmall, M. C. 2005. The role of the therapeutic alliance in the treatment of
substance misuse: a critical review of the literature. Addiction 100(3): 267–268.
Meier, P. S., Donmall, M. C., McElduff, P., Barrowclough, C. and Heller, R. F. 2006. The role of the early thera-
peutic alliance in predicting drug treatment dropout. Drug and Alcohol Dependence 83(1): 57–64.
Meissner, W. W. 1996. The Therapeutic Alliance. New Haven, CT: Yale University Press.
Miller, W. R., Rollnick, S. and Conforti, K. 2002. Motivational Interviewing: Preparing People for Change. 2nd edn.
New York: The Guilford Press.
Miller, W. R., Taylor, C. A. and West, J. C. 1980. Focused versus broad-spectrum behavior therapy for problem
drinkers. Journal of Consulting and Clinical Psychology 48: 590–601.
Nagengast, S. L., Baun, M. M., Megel, M. and Leibowitz, J. M. 1997. The effects of the presence of a com-
panion animal on physiological arousal and behavioral distress in children during a physical examination.
Journal of Pediatric Nursing 12: 323–330.
Odendaal, J. S. 2000. Animal-assisted therapy—Magic or medicine? Journal of Psychosomatic Research 49(4):
275–280.
Orlinsky, D. E., Ronnestad, M. H. and Wilutzki, U. 2003. Fifty years of process-outcome research: Continuity and
change. In Bergin and Garfield’s Handbook of Psychotherapy and Behavioral Change. 5th edn, 307–390, ed.
M. J. Lambert. New York: Wiley.
Prochaska, J. O., DiClemente, C. C. and Norcross, J. C. 1993. In search of how people change: Applications
to addictive behaviors. American Psychologist 47: 1102–1114.
SAMHSA. 2006a. Discharges Who Left Against Professional Advice: 2003 (Dasis Report Issue 28). Washing-
ton DC: Office of Applied Studies, Substance Abuse and Mental Health Services Administration.
SAMHSA. 2006b. Trends in Substance Abuse Treatment Admissions: 1993 and 2003 (The Dasis Report Issue
12, 2006). Washington DC: Office of Applied Studies, Substance Abuse and Mental Health Services
Administration.
Serpell, J. A. 2002. Anthropomorphism and anthropomorphic selection—Beyond the “Cute Response.” Society
& Animals 10(4): 437–454.
Templer, D. I., Salter, C. A., Baldwin, R. and Veleber, D. M. 1981. The construction of a pet attitude scale. The
Psychological Record 31: 343–348.
Wilson, C. C. and Barker, S. B. 2003. Challenges in designing human–animal interaction research. American
Behavioral Scientist 47(1): 16–28.
Yalom, I. D. 2005. The Theory and Practice of Group Psychotherapy. 5th edn. New York: Basic Books.
Anthrozoös
148