A meta-analysis of 10 randomized controlled trials found that dog-assisted therapy significantly improved behavioral and psychological symptoms in dementia patients compared to standard care, particularly reducing agitation and depression. However, the evidence base remains weak due to small study samples and high risk of bias, limiting confidence in clinical applications.
Behavioral and psychological symptoms of dementia (BPSD) such as depression, agitation, and social withdrawal create substantial burdens for both patients and caregivers. These symptoms correlate with higher hospitalization rates, mortality, and caregiver stress. Current treatment often relies on pharmacological interventions, which carry risks of adverse effects in older populations. This systematic review and meta-analysis examined whether dog-assisted therapy (DAT) could provide a non-pharmacological alternative by analyzing 10 randomized controlled trials published through April 2025.
The pooled analysis demonstrated that dog-assisted therapy produced statistically significant improvements in overall behavioral and psychological symptoms compared to routine care (standardized mean difference = -0.46, p < 0.0001). When researchers examined specific symptom categories, DAT was particularly effective for agitation and depressive symptoms. Additionally, functional activity improved significantly with DAT versus control (standardized mean difference = 0.31, p = 0.03). However, cognitive function showed no statistically significant difference between the dog-assisted therapy group and control group, indicating that DAT does not appear to reverse cognitive decline itself.
The subgroup analyses provided more granular insight into where DAT's benefits concentrated. Agitation and depressive symptoms responded most robustly to the intervention, suggesting the mechanism may involve emotional regulation and behavioral engagement rather than cognitive restoration. This distinction matters clinically: DAT appears positioned as a behavioral management strategy rather than a disease-modifying therapy for dementia itself.
The authors emphasized a critical caveat: the overall evidence quality was rated as weak. The included studies relied predominantly on small sample sizes, and the meta-analysis found high overall risk of bias across the 10 trials. This means while the direction of effect appears consistent and the statistical significance meets traditional thresholds, the strength of inference remains limited. The authors recommend DAT as a non-pharmacological care option in long-term care facilities while calling for larger, higher-quality studies with extended follow-up periods to establish more robust clinical benefits.
If you or a family member is navigating dementia care in a long-term care setting, this research suggests dog-assisted therapy could be worth discussing with your care team as a complementary non-medication strategy, particularly if behavioral symptoms like agitation or depression are present. The intervention involves structured or unstructured interaction with a trained therapy dog during care.
The evidence does not support expecting cognitive improvement from DAT. Rather, benefits appear concentrated in symptom management and quality-of-life markers. This makes it reasonable to integrate alongside standard care, though it should not replace established behavioral management approaches.
The weak evidence tier means individual results will vary considerably. Small study populations mean the "typical" patient may not reflect your specific situation. Ask your facility whether they track outcomes from DAT interventions in their own population—real-world effectiveness often diverges from trial results.
| Detail | Information |
|---|---|
| Study Type | Systematic review and meta-analysis |
| Databases Searched | AgeLine, CINAHL, Cochrane Library, Embase, ProQuest, PubMed, Airiti Library |
| Study Design Included | Randomized controlled trials only |
| Number of Trials | 10 RCTs |
| Primary Outcome | Behavioral and psychological symptoms of dementia (BPSD) |
| Secondary Outcomes | Functional activity, cognitive function |
| Evidence Quality | Weak (small samples, high risk of bias across trials) |
| Journal | Hu li za zhi (The Journal of Nursing) |
| PubMed ID | 42151047 |
ProtocolEngine provides general health information based on published research. This is not medical advice. Consult a healthcare professional before starting any supplement or health protocol.