The term assistance dog covers a much broader reality than that of a guide dog for the blind.

It refers to dogs trained to compensate for a motor or sensory disability, or sometimes certain medical or psychological conditions, with very different missions depending on the case.

Picking up an object, opening a door, signaling a doorbell, reacting during an epileptic seizure, detecting hypoglycemia, preventing elopement, or interrupting dangerous behavior: we are not talking about a single canine profession, but a set of distinct functions, with training frameworks and levels of scientific evidence that vary greatly.

The French legal framework

In France, the right of access for guide dogs and assistance dogs to public places and transport is not based on a single text.

It is primarily based on Article 88 of Law no. 87-588 of July 30, 1987, and then on subsequent texts, notably Law no. 2005-102 of February 11, 2005, which consolidates this right within the broader framework of equal rights and opportunities.

Official documents stipulate that these dogs may access public places, transport, and places allowing professional, educational, or training activities, without additional cost and without the obligation of a muzzle.

Prohibiting access to a guide dog or an assistance dog is punishable by a fine of up to 450 euros for an individual and 2,250 euros for a legal entity.

To fully benefit from this framework, the dog must be trained by a certified center and provide proof of this training through the certificate provided by the texts.

The PCH (Compensation Benefit for Disability) can also include animal assistance for a guide dog or an assistance dog.

The administrative sources in force indicate a ceiling of up to 6,000 euros over a period of 10 years, provided that the animal has been trained in a recognized structure within this framework.

Handi’chiens: a major player in France

In France, HANDI’CHIENS, created in 1989 and recognized as a public utility in 2012, is one of the major players in the sector.

The association states that it has four certified training centers and trains assistance dogs that are provided free of charge to children, adolescents, or adults with disabilities or vulnerabilities.

The process described by the association lasts approximately two years, with practically 16 months in a host family and then 6 months in a training center before the dog is handed over.

This model perfectly illustrates the "French-style" logic of assistance dogs: a long socialization in a real environment, followed by a more technical specialization phase in a center.

Main categories of assistance dogs

1. Mobility assistance dogs

These are the most clearly regulated assistance dogs in France.

They are trained to compensate for a motor disability by performing concrete tasks:

• picking up objects,
• opening or closing a door,
• operating a switch,
• helping with undressing,
• bringing a phone,
• sometimes triggering an alarm or assisting during a fall.

These are also the dogs for which the certification system for centers and trainers is most clearly structured in the French landscape.

2. Dogs for deaf or hard of hearing people

These dogs are trained to

•  identify precise everyday sound signals such as doorbells, alarm clocks, fire alarms, a child's cries, telephones, etc.
• alert their owner with a coded behavior, often physical contact and guiding them towards the source.

Their role is that of a sensory mediator between the person and the domestic or social environment.

3. Seizure response dogs and alert dogs

Two often confused realities must be distinguished.

Seizure response dogs are trained to act during or after a seizure: staying with the person, fetching help, triggering an alarm, limiting immediate danger, or positioning themselves to secure the environment.

Alert dogs, on the other hand, are supposed to warn before the onset of a seizure. On this point, the literature remains cautious. Catala and Grandgeorge's review published in 2018 explains that reported observations are real, but robust evidence remains insufficient to conclude a reliable and generalizable predictive ability.

In other words, there are reported cases of dogs seemingly anticipating certain seizures, sometimes spontaneously, but the difference between spontaneous behavior, secondary learning, and true pre-ictal detection remains a subject of research.

4. Hypoglycemia detection dogs

"Diabetes alert dogs" are trained to detect, through olfaction, metabolic changes associated with hypoglycemia.

In the laboratory, some results are encouraging: the study by Hardin et al. showed that dogs could be conditioned to recognize sweat samples related to hypoglycemia using smell alone.

However, real-world results are much more nuanced.

The study by Gonder-Frederick et al. showed an average sensitivity of 35.9% for hypoglycemic episodes during waking hours, with great variability from one dog to another.

This does not mean that these dogs never help; it means that they cannot be considered a reliable substitute for medical glycemic monitoring.

5. Assistance dogs for autism

Assistance dogs for individuals with autism spectrum disorders are trained for several functions:

• reducing the risk of elopement,
• improving transitions,
• supporting emotional regulation,
• calming,
• facilitating social interactions,
• sometimes interrupting certain repetitive or self-injurious behaviors.

The systematic review by Lindsay et al., covering 29 studies and 1,121 children or young people in 6 countries, concludes that a majority of studies report positive effects in at least one physical, psychological, or social domain, even if the overall level of evidence remains moderate.

6. Psychiatric assistance dogs

There are also dogs trained to accompany individuals with PTSD, severe anxiety disorders, or other psychiatric conditions.

Tasks may include:

•  interrupting a dissociative episode,
• creating a safe space in public places,
• reminding them to take medication,
• soliciting a relative or third party in case of crisis.

Recent literature, particularly among veterans, suggests a real benefit as a complement to usual care, but this field is still developing, with persistent debates on training standards, indications, and the welfare of the dog itself.

What these dogs have in common: health, career, and retirement

Even if their missions differ, assistance dogs share several common realities.

First, their service life is often long, around 8 to 10 years in large structured programs.

Secondly, they most often work in domestic or urban environments, on hard surfaces, with daily repetitions.

This activity profile logically leads to locomotor wear comparable to that already well-documented in guide dogs.

This explains the importance of orthopedic selection in serious programs, particularly in the most commonly used breeds such as Labradors, Golden Retrievers, or German Shepherds.

Serious breeding and training structures integrate this dimension from initial selection, precisely because a dog that has to work for a long time cannot be built on a fragile joint base.

Finally, these dogs share a very strong relational dependence on the human-dog team.

Poorly prepared retirement, a sudden withdrawal from activity, or a sudden decrease in stimulation can be difficult to cope with.

For the veterinarian, this means that they are not just monitoring a "pet" dog, but a former working animal with a particular locomotor, behavioral, and relational profile.

This observation is even more true when the dog has spent years in a mission involving permanent human contact.

In conclusion

Assistance dogs do not form a homogeneous block.

Between mobility assistance dogs, dogs for the hard of hearing, seizure response dogs, diabetic alert dogs, autism assistance dogs, or psychiatric assistance dogs, the missions, scientific evidence, and expectations are not at all the same. The French framework is relatively solid for guide dogs and assistance dogs from certified centers, but many medical or psychiatric specialties remain more heterogeneous in their practices and their level of scientific validation.

The common thread, however, remains clear: these are working dogs, and they must be monitored as such, with real attention paid to selection, joint health, mental load, and retirement.

Sources:

1. Law no. 87-588 of July 30, 1987, Article 88, on the access of guide dogs and assistance dogs
2. Law no. 2005-102 of February 11, 2005 for equal rights and opportunities
3. Code of Social Action and Families, provisions relating to the education certificate from a certified center
4. Service-Public.fr, Disability Compensation Benefit (PCH)
5. HANDI’CHIENS, institutional website
6. HANDI’CHIENS, presentation of the association and the training process
7. Catala A. and Grandgeorge M., Dog alerting and/or responding to epileptic seizures: A scoping review, 2018
8. Dalziel D.J. et al., Seizure-alert dogs: A review and preliminary study, 2003
9. Hardin D.S. et al., Dogs Can Be Successfully Trained to Alert to Hypoglycemia Samples from Patients with Type 1 Diabetes, Diabetes Therapy, 2015
10. Gonder-Frederick L.A. et al., Diabetes Alert Dogs (DADs): An assessment of accuracy and implications, 2017
11. Lindsay S. et al., The impact of service dogs on children, youth and their families: A systematic review, 2021
12. Leighton S.C. et al., clinical trial on psychiatric assistance dogs for PTSD, 2024
13. Kiiroja L. et al., review on PTSD assistance dogs and animal welfare, 2025