Manas Dave examines topics covered in our sister journal Evidence-Based Dentistry.

figure 1

©Orawan Wongka/iStock/Getty Images Plus

Introduction

The article by Massouda et al. ‘Evaluating effects of animal-assisted therapy on paediatric dental care patients: A pilot clinical trial'1 was published in the Journal of the American Dental Association in June 2025 and a commentary ‘Are therapy animals the key to happier dental visits for children?' was published in Evidence-Based Dentistry in December 2025.2

Background

Dental anxiety is apprehension or a reaction to feeling threatened in a dental setting. This can manifest with several outcomes such as disruptive behaviour, increased pain response during dental procedures and avoidance of seeking dental help, increasing the risk of infection and emergency visits.3 Animal-assisted therapy (AAT) is a non-pharmacological intervention that uses a trained therapy animal to achieve treatment goals. For example, dogs can sense elevated stress hormones and respond based on their training to calm the patient. The aim of this prospective controlled clinical trial was to assess the effects of AAT on objective and subjective measures of dental anxiety.

Methods

The objective measures of anxiety included: salivary cortisol, alpha-amylase, video-coded behaviours and heart rate. The subjective measures included self-reported measures of anxiety, fear, pain and next visit expectations. Participants aged 7–14 years who were scheduled to receive invasive dental procedures were contacted for enrolment into the trial.

Unstimulated saliva samples were collected and processed using saliva collection swabs. Video data was collected using a wireless camera (GoPro) and were coded according to the paediatric dental pain, anxiety and fear coding approach. Participants were either selected to be in the control group (no AAT) or the intervention group (AAT) through consecutive enrolment and then allocated by the weekday of their appointment as the therapy dog and handler were only available on certain days.

Results

  • A total of 39 participants were recruited (AAT: n = 18; control: n = 21) into the trial. The participants were well-matched for age and sex. Those identifying as Latin American represented a majority of both groups

  • Restorations were performed in 61% of all participants (AAT: 67%, control 57%) and extractions performed in 36% of patients (AAT: 33%, control 38%). One control participant received local anaesthetic injection only (no procedure due to behavioural challenges)

  • There were no differences between groups for salivary cortisol or alpha-amylase at any time point

  • AAT participants had a longer mean duration of relaxed lower body posture during the procedure than control participants. Post procedure anxiety indices (CFSS-DD and MCDAS) showed a significant positive correlation with the number of verbal indicators during the procedure

  • Nitrous oxide use was higher in the AAT (89%) group than the control group (71%)

  • Post-procedure, the control group reported significantly higher pain on the Wong-Baker FACES scale (WBS) than the AAT group (p = 0.001). Pain creased post-procedure in 53% of controls versus 21% of AAT participants.

Conclusions

The authors stated:

‘There was less variation in HR range among AAT participants than the control group, with decreases in HR after key events, such as procedure start, which may suggest a decrease in DA [dental anxiety]. No differences were noted in fear and anxiety scales, and salivary cortisol and α-amylase levels. The AAT group had longer durations of relaxed lower body and reported significantly less pain at the end of the procedure than the control group…'

A key strength of this pilot trial is its use of multiple, complementary anxiety readouts, behavioural video coding and heart-rate dynamics alongside validated child-reported scales and salivary biomarkers.

Commentary

A key strength of this pilot trial is its use of multiple, complementary anxiety readouts, behavioural video coding and heart-rate dynamics alongside validated child-reported scales and salivary biomarkers. This provides more of a nuanced picture than self-reporting alone. The pragmatic clinic-based delivery also supports feasibility and acceptability of animal-assisted therapy in real paediatric dental workflows. However, allocation by appointment day (rather than randomisation), a modest sample size, unequal nitrous oxide use between groups and a heterogeneous control condition introduce risk of confounding and limit causal inference. The study is therefore better viewed as hypothesis-generating. Future research should prioritise adequately powered randomised designs (ideally with a single, standardised control), prespecify primary outcomes and key timepoints, and control for sedation/analgesia and procedure complexity, to clarify whether the signal for reduced pain and more relaxed behaviour reflects a true anxiolytic effect of AAT and to define which children and procedures benefit most.