Autism-specific therapy and support

What “autism-specific” means here

The guideline organised its recommendations by therapy goals. One of them is improving social interaction and communication. ^{[1, Ch. C.4]} This article describes only this goal. Other goals, such as language, daily-living skills or the treatment of anxiety and depression, had chapters of their own (see The S3 guidelines on autism). According to the guideline, any therapy should be planned with clearly defined, concrete goals and for a limited period. ^{[1, Rec. 1]} More on planning is in the article What support helps?.

Small children (toddler and preschool age)

For toddlers and preschool children, regardless of developmental age and cognitive skills, the guideline recommended:

• Early start. Autism-specific therapy should be carried out as soon as autism is suspected; after 12 months at the latest, the diagnosis should be checked again before therapy is continued (consensus-based, strong consensus). ^{[1, Rec. 9]} Therapy should start early, at the latest from the age of 2 years or directly after the diagnosis (evidence level 2, grade of recommendation A). ^{[1, Rec. 10]}
• Type of therapy. Individualised, development-oriented and behavioural-therapy-based, following scientifically tested therapy manuals, including guidance for parents or primary caregivers on supportive parent–child interaction. ^{[1, Rec. 10]}
• Duration and scope. At least one year and at least two hours per week; in addition, parents or caregivers should interact and practise with the child in everyday life. ^{[1, Rec. 10]}
• Parent guidance. According to the guideline, parent training alone without the child is not enough. There should be regular feedback on implementation, for example by video, and parental strain should be taken into account in planning. ^{[1, Rec. 10]}
• Nursery. Educators and integration assistants should additionally be guided in supportive interaction with the child. ^{[1, Rec. 10]}
• Contents. Among other things, a standardised-as-possible assessment of the child’s skills at the start, attention to motivation, interests and needs (avoiding over- and under-demand), the family’s needs, training in joint attention, imitation and play, and promoting self-initiated communication. ^{[1, Rec. 10]}
• Methods. Exercises in situations that are as natural and close to everyday life as possible so that what is learned carries over into daily life, and behavioural-therapy techniques such as reinforcement with natural, preferably social reinforcers. ^{[1, Rec. 10]}

According to the guideline, therapy can be continued until the child is integrated into the (inclusive) nursery or the (special) school (consensus-based). ^{[1, Rec. 11]} The recommendation on contents and methods had evidence level 2 to 3 and grade of recommendation A to B. ^{[1, Rec. 10]}

School children and adolescents

Without intellectual disability. School children and adolescents should be offered a manualised, scientifically tested group therapy, limited to 3 to 6 months, to promote social interaction with peers (evidence level 1 to 2, grade of recommendation A, strong consensus). The format described was a closed group of about 4 to 5 children or adolescents and two therapists, 12 to 18 double sessions of 90 minutes each, weekly, and 2 to 3 parent information evenings. Parents should support the homework in everyday life. ^{[1, Rec. 14]} The contents included, for example, clear group rules, education about autism, training in greetings and conversations, recognising one’s own and others’ feelings, building friendships and emotion regulation. ^{[1, Rec. 14]} If a group is not possible, the contents can also be delivered in time-limited individual therapy, with many exercises and homework for everyday life (consensus-based). ^{[1, Rec. 15]}

With intellectual disability. Here the guideline recommended (consensus-based) a clear structuring of everyday life and demands that match development and competence. For older children and adolescents, promoting daily-living skills moves more into the foreground; a time-limited group therapy over 3 to 6 months can be offered, involving parents and school. Individual therapy alone to promote social interaction with peers should not be carried out. ^{[1, Rec. 13]} Several associations (including Autismus Deutschland e. V. and behaviour-therapy associations) submitted dissenting votes on this: support should, where possible and sensible, also take place in group format alongside individual therapy and involve parents and school. ^{[1, Rec. 13]}

Adults

Without intellectual disability. Adults should be offered a manualised group therapy limited to 3 to 6 months to promote social interaction (evidence level 2, grade of recommendation A). The format provided is a closed group of about 6 to 8 adults and two therapists, 12 to 18 double sessions of 90 minutes each. The contents included, among other things, education about autism including aspects of identity, conversation techniques, dealing with wishes for relationships and sexuality, dealing with stress and loneliness, and emotion regulation. ^{[1, Rec. 17]} Individual therapy is an alternative when a group cannot be implemented (consensus-based). ^{[1, Rec. 19]} In addition, imitation- and synchronisation-based dance or movement therapy in a group can be offered (grade of recommendation 0, that is, an open recommendation because of low evidence). ^{[1, Rec. 18]}

With intellectual disability. The focus is on daily-living skills and adaptive behaviour in structured activity services; a time-limited group therapy can be offered in addition (consensus-based). ^{[1, Rec. 16]}

What the guideline did not recommend for this goal

For toddlers, preschool and primary school children, the guideline named methods that, in randomised controlled trials of overall good to adequate quality, were not sufficiently effective in promoting social interaction (and other aspects). They should not be used for this goal, since more effective therapies are available: ^{[1, Rec. 12]}

• Hanen “More Than Words”,
• pure parent training without involving the child,
• pure structuring measures without special training of interactive, communicative and daily-living skills,
• improvisation-based music therapy (also with involvement of the family),
• isolated cognitive, computer- or iPad-based training,
• auditory integration therapy,
• sensory integration therapy according to Ayres,
• horse-riding therapy and other animal-assisted therapies,
• facilitated communication,
• eclectic approaches.

This recommendation had evidence level 2 to 3 and grade of recommendation A, with a consensus (not a strong consensus). The Deutsche Musiktherapeutische Gesellschaft submitted a dissenting vote and demanded that the point on music therapy be deleted. ^{[1, Rec. 12]} This explicitly refers only to the goal of social interaction in children up to primary school age; a method that was not recommended here may be described for a different goal. ^{[1, Foreword]}

A separate chapter of the guideline also named methods that should never be used in autism because they are harmful to health or ethically questionable. More on this in the article Recognising controversial and dubious offers.

Side effects and limits

According to the guideline, no undesirable effects of psychosocial interventions have so far been known when carried out correctly; but they have also rarely been studied systematically (evidence level 1 to 3, strong consensus). ^{[1, Statement 2]} The associations that submitted a dissenting vote on the group therapies pointed out that randomised controlled trials mostly examined only very time-limited group formats. These had shown a certain added benefit and should, where possible, be part of treatment alongside individual therapy and parent or institutional counselling. ^{[1, Rec. 14, 17]} For school children without intellectual disability, a consensus-based recommendation (majority agreement, more than 50 to 75 per cent) also stated: in therapies it should be taken into account that many people with autism first need a longer period to build a therapeutic relationship and that transferring what has been learned into everyday life is often made harder by reduced ability to generalise and by reduced flexibility. ^{[1, Rec. 47]}

How you can assess a service

The recommendations are not instructions to buy or reject a particular service. Helpful questions are: What concrete goal does the therapy pursue? How long is it planned for and when will it be reviewed? Does the service work with parents or caregivers and does it avoid disregarding strengths and interests? Do group size and scope suit the age? The planning principles (clear goals, limited period, review, involvement of those affected) are described in the article What support helps?. Who bears the costs is explained in Financing autism therapy.

Sources and literature

1. DGKJP, DGPPN et al. Autismus-Spektrum-Störungen im Kindes-, Jugend- und Erwachsenenalter. Teil 2: Therapie. Interdisziplinäre S3-Leitlinie, AWMF-Reg.-Nr. 028-047, Langversion, Stand Text 02.05.2021 (Vorwort; Empf. 1; Kap. C.4 mit Empf. 9–19 und Aussage 2). – Guideline (in German); validity expired on 23 March 2026 according to the DGKJP overview ^[2]
2. DGKJP. Aktuelle Leitlinien (Eintrag 028-047: „Gültig bis 23.03.2026“). https://www.dgkjp.de/wissen/aktuelle-leitlinien/ – Professional society (website in German)