Biobehavioral Effects of Therapy Dog Visitation in Elderly Intensive Care Unit Patients: A Pilot and Feasibility Study | HABRI

Biobehavioral Effects of Therapy Dog Visitation in Elderly Intensive Care Unit Patients: A Pilot and Feasibility Study

Principal Investigator

Sandy Branson, PhD, MSN, RN (Cizik School of Nursing, The University of Texas Health Science Center)

Full Text


Elderly patients who are admitted to the intensive care unit (ICU) are at risk for psychosocial stress that negatively affects mental and physical health. Managing psychosocial stress in the ICU requires an innovative approach with low-risk and cost-effective interventions. Limited empirical evidence suggests the potential efficacy of a brief therapy dog visit (TDV) in improving psychosocial stress and biological responses to psychosocial stress in ICU patients. Due to the novelty of testing TDV in an ICU setting, a pilot and feasibility study is needed to inform, justify and improve rigor of a future randomized control trial (RCT).


To assess the preliminary efficacy of a 10-minute TDV in reducing biobehavioral stress responses including anxiety, salivary cortisol, interleukin-1ß, and C-reactive protein in elderly ICU patients, obtain initial estimates in outcomes for sample size calculations for a larger trial, and assess the rates of consent, completion of psychosocial data and saliva collection, and the best way to facilitate the TDV.


ICU patients ≥60 years old were randomly assigned to a 10-min AAA intervention or control/usual ICU care. Attitudes toward pets were assessed before the intervention. Self-reported stress and anxiety and salivary stress biomarkers were collected before and after the intervention and the usual care condition.


The majority of patients were ineligible due to lack of decisional capacity, younger age, inability to provide saliva specimens, or critical illness. Though 15 participants were randomly allocated (Animal Assisted Activities (AAA) = 9; control = 6), only 10 completed the study. All participants completed the questionnaires; however, saliva specimens were significantly limited by volume.

AAA was associated with decreases in stress and anxiety. Biomarker results were variable and revealed no specific trends associated with stress responses. Barriers to recruitment included an insufficient number of patients eligible for AAA based on hospital policy, difficulty finding patients who met study eligibility criteria, and illness-related factors.

Recommendations for future studies include larger samples, a stronger control intervention such as a visitor without a dog, greater control over the AAA intervention, and use of blood from indwelling catheters for biomarkers.