A few standardised tools researchers actually use
- PIADS, the Psychosocial Impact of Assistive Devices Scale. A 26-item questionnaire that measures the perceived impact of an assistive device, including an assistance dog, on functional independence, wellbeing and quality of life. It is useful here because it was designed to work across many different kinds of assistive technology, not just dogs.
- Quality of life measures. Broader scales that track day-to-day wellbeing, participation and daily functioning, sometimes measured before a dog is placed and then again at several points afterward.
- Self esteem and psychological wellbeing scales. Standardised, validated instruments such as the Rosenberg Self Esteem Scale, used to assess whether living with an assistance dog is associated with measurable differences in self-esteem.
- Satisfaction measures. Tools such as the Quebec User Evaluation of Satisfaction with Assistive Technology, which assess satisfaction with an assistive technology and the service around it.
One well known UK example is a longitudinal study run by the charity Dogs for Good, which followed 307 people who had applied for an assistance dog, surveying them before placement and then again at six, twelve, twenty four and thirty six months afterward. Tracking the same people over multiple years like this allows a study to examine how outcomes change over time, rather than relying on a single snapshot, though a service evaluation of this kind cannot on its own prove that the dog caused any change observed.
PIADS Resource Centre · Dogs for Good Longitudinal Service Evaluation
The honest picture is more mixed than most people expect
A 2021 systematic review pulled together 27 separate studies on the psychosocial effects of guide, hearing, medical and mobility assistance dogs, covering 147 statistical comparisons in total. The result is worth stating plainly: 68 percent of those comparisons found no statistically significant difference, 30 percent found a positive effect in the expected direction, and 2 percent found a negative one. Where positive effects did show up, they tended to involve psychological wellbeing, emotional functioning, self esteem and vitality. The review did not include psychiatric service dogs or emotional support dogs, focusing instead on guide, hearing, medical and mobility assistance dogs for people with physical disabilities, so its findings should not be assumed to apply equally to every assistance dog category.
This is not a discouraging finding so much as an honest one, and it says as much about the state of the research as it does about assistance dogs themselves. The review's authors pointed to real methodological weaknesses across many of the studies, including small sample sizes, inconsistent reporting, and a general failure to account for other factors that could explain the differences researchers were seeing. A statistically non-significant result in a small, imperfectly designed study does not prove a dog made no difference. It means the study did not find sufficiently strong statistical evidence of a difference, which is not the same as proving there is no difference, and the evidence available today is not yet strong or consistent enough to make confident, sweeping claims in either direction.
Systematic Review, Psychosocial Effects of Assistance Dogs, 2021
Training success and life impact are not the same measurement
It is worth being clear about a distinction that often gets blurred. Whether a dog successfully completes assistance dog training, the kind of metric we discussed in an earlier FootNotes piece on suitability, is a different question from whether that dog, once placed, actually improves a person's life once it gets there. An organisation can report a training success rate without ever properly measuring life impact, and the reverse is also true. Both matter, but they answer different questions, and a responsible programme needs to be honest about which one it is actually reporting when it talks about "results."
Why we are not claiming outcomes we have not measured yet
Cohort 1 has not yet reached placement, so we do not have impact results to report, and we are not going to imply otherwise with a well chosen photo or story. What we can say honestly is how we intend to measure this properly once placements begin: using established, validated measures like the ones described above, where they are appropriate to the outcomes we want to assess, rather than inventing our own metrics, tracking people over time rather than at a single point, and being willing to report a null or mixed result if that is what the data actually shows.
We think this matters because the field's own evidence, including the 68 percent of statistical comparisons in one systematic review that found no statistically significant difference, is a reason for more rigorous measurement, not less. A story about one person is real and worth telling, but it is not a substitute for asking, carefully and honestly, whether our dogs are actually making the difference we hope they will.
