Table 5. Outcomes and conclusions.

No.	1st Author	Assessment type (inc. measures)	Results summary	Feasibility, tolerability acceptability & dropout	Limitations & future directions	Author conclusions
1	Hamama, L.	Self-report questionnaires PCL-C, CESD; Likert scales (subjective wellbeing and coping with stressful life events)	• I pre to post = Rapid decline in PTSD symptoms, & risk for PTSD diagnosis. No sig improvement to depression or subjective wellbeing or coping. • No sig diffs between I and C by end of intervention on PTSD symptoms	No dropouts evident or reported. Potential subject pool 20 (I)	Small sample size, poss selection bias	• Group therapy contributes to trauma healing. • Sense of control, mastery aided by canines
2	Hanselman, J. L.	Self-report questionnaires STAS-TAS, CABS, BDI-II; subjective mood thermometers (tension, confusion, fatigue, depression); facilitator observations	• Sig reduction in emotional & behavioural anger; • Sig increase in animal bonding; • Sig increase in depression; • observed improvements in pos. behaviour when canines present	• No dropouts evident or reported. Potential subject pool unknown. • Increase in depression attributed to increased emotional awareness & reduced substance use.	More sessions (12) and longer duration (2h) required for behavioural change. Qualitative data should be increased. Parent feedback should be analysed & more data sought.	The results confirm previous studies that animals are beneficial in treatment.
3	Hartwig, E.	Self-report questionnaires BYI-II (5 scales)	• No sig diff between I and C on BYI-II. • Sig reduction in BAI, BDI, BDBI but not BANI, or BSCI in both I and C conditions	• No dropouts evident or reported. Potential subject pool unknown. • Requires extensive staff training & supervision, Difficulties incorporating canines into curriculum	• Canine activities not sufficiently experiential. Small sample size. Small counselling rooms. • Explore impact of canine temperament & energy.	HART curriculum produces sig decreases in anxiety, depression, anger and disruptive behaviour. CAT useful adjunct
4	Lange, A. M.	Structured interview and subsequent qualitative analysis	Participants report canine presence is beneficial for humour, calming, attendance, disclosure, self-soothing, feeling attached	• 2 of 5 participants were not available for the follow up interview. • Therapist training needed. • Canine assessment & liability issues must be addressed.	• Sample size too small for thematic analysis. • Need to explore efficacy with other presenting issues	Potential benefits include calming, humour relief, safety in disclosing, experiences of empathy, motivation to attend.
5	Lubbe, C.	Document analysis & semi-structured interview for thematic analysis	Five themes identified; facilitating relationship, communication, physical affection, socialisation, and self-esteem.	• Author’s observations and thematic analysis indicate that the client was well engaged, unlike previous counselling attempts. • Requires properly trained animal & handler, so possible limitations to transferability/scope.	Findings may not be attributable to canine presence	Canine presence promotes engagement and facilitates therapeutic process
6	Stefanini, 2015	Staff reported measures; C-GAS, format of hospital care, ordinary school attendance; and observational/ behavioural coding.	• I = Sig improvement in global functioning, format of care (inc. duration of hospital stay), school attendance over TAU (C). • I = Sig increases in observed in-session behavioural participation, animal interaction & affection, adult & peer socialisations, reduced withdrawal	I = Dropout was zero, attendance 100%. Potential subject pool unknown.	• Small sample size in single location. • Follow up at 6 and 12 months recommended. Analyse results by diagnosis. Explore mechanisms of interaction between patient, animal and operators.	• I = significant clinical and behavioural improvements over TAU. • Animals may act as a catalyst in the therapeutic process, esp. socialization but mechanisms of action require further study
7	Stefanini, 2016	Staff report C-GAS, and observational/ behavioural coding; YSR	• I = Sig decrease in internalizing probs, sig increase in total competence & global functioning over TAU (C). • I = Sig decrease in externalizing probs pre to post. Not sig compared to TAU (C) (rate of change). • I = Sig increase in observed in-session behavioural participation; interaction & affection with animal; socialization with adults and peers; sig. decrease in withdrawal.	I = Dropout was zero, attendance 100%. Potential subject pool unknown.	• Small sample size in single location. Findings should be replicated in various sites, ages and diagnoses. • Explore mechanisms of action in AAT	Hypothesis supported that AAT reduces emotional & behavioural symptoms and increases global competence and psychological functioning over TAU. AAT may be more effective for internalizing symptoms.

I = intervention group or experimental group, C = control or comparison group; TAU = Treatment as Usual; PTSD = Post Traumatic Stress Disorder

Assessments and Measures

BDI-II = Beck Depression Inventory (second edition)

BYI-II = Beck Youth Inventories (second edition) Including—Anxiety (BAI), Depression (BDI), Disruptive Behavior (BDBI), Anger (BANI), and Self Concept (BSCI).

C-GAS = Children’s Global Assessment Scale

CABS = Companion Animal Bonding Scale

CESD = The short Centre for Epidemiologic Studies Depression Scale

PCL-C = Post Traumatic Stress Disorder checklist–Civilian

STAS-TAS = State-Trait Anger Scale

YSR = Youth Self Report