Published by Unseen Progress, an independent publisher of caregiver research. Last reviewed 2026-05-10. Part of the selective mutism research overview.
Short answer. They are not treated in sequence — they are treated together, with selective mutism behaviours as the primary target and social anxiety as the underlying fuel. The research consistently finds that the majority of children with selective mutism also meet criteria for social anxiety disorder (Cunningham & McHolm, 2006; Bergman et al., 2013), and the two conditions share substantial neurobiological and behavioural overlap. The clinical literature does not support waiting until social anxiety is "treated" before working on speech — speech behaviour change is one of the most effective ways to reduce social anxiety in this population.
Cunningham, McHolm and Boyle's comparative study (Cunningham & McHolm, 2006) found that children with selective mutism scored as high as or higher than children with diagnosed social phobia on measures of social anxiety. The DSM-5 explicitly notes that selective mutism and social anxiety disorder are highly comorbid, with most clinical samples showing co-occurrence rates above 90%. The SMart Center's clinical materials and Aimee Kotrba's synthesis (Kotrba, 2015) describe selective mutism as, functionally, "social anxiety with a freeze response that collapses speech."
This matters for treatment ordering. If selective mutism were a separate condition that sat next to social anxiety, sequencing would be plausible — treat one, then the other. But because the selective mutism behaviour is one of the load-bearing expressions of the underlying social anxiety, removing it is itself anxiety treatment.
Caregivers and well-intentioned clinicians sometimes propose: "Let's reduce the anxiety first — through talk therapy, mindfulness, or relaxation — and then she'll naturally start speaking." The literature does not support this approach, for three reasons.
First, traditional anxiety-talk-therapy depends on the child being able to verbalise their experience to the therapist. A child with selective mutism, by definition, cannot speak to the therapist in the early sessions. The therapy modality collapses before the mechanism can do any work.
Second, the freeze response is reinforced every time the child enters a speaking-expected setting and does not speak. Each non-speaking exposure consolidates the avoidance pattern, regardless of how relaxed the child feels at home. Anxiety reduction in a low-demand setting does not generalise to a high-demand setting on its own (Bergman et al., 2013; Kurtz, 2020).
Third, the children who recover from selective mutism through structured behavioural treatment show concurrent reductions in social anxiety scores (Bergman et al., 2013). The two move together when speech behaviour changes — they do not need to be addressed in series.
The integrated behavioural model used by Bergman, Kurtz, McHolm, and the SMart Center treats selective mutism behaviours as the primary target while explicitly working on the underlying anxiety in parallel:
Before any new variable is added, confirm the child reliably speaks in the home setting with the parent. If not, work on this first — but as a behavioural target, not as standalone anxiety treatment.
Sliding-in (see the sliding-in question) addresses both conditions simultaneously. Each successful slide-in adds an exposure that lowers anxiety to the relevant setting and generalises the speaking behaviour. The two effects compound.
Once the child can speak to a clinician — even at low volume, even in short answers — CBT components can be layered in: identifying anxious thoughts, building a worry hierarchy, structured exposure planning. This sequencing matches the developmental capacity of the child rather than the order of the diagnoses.
For children whose social anxiety is severe enough to block participation in the behavioural protocol itself, an SSRI consultation is supported by the limited but consistent pediatric anxiety literature (Black & Uhde, 1994; Manassis et al., 2016). The medication is a tool to allow the behavioural protocol to work, not a replacement for it. See the SSRI question for detail.
Adolescents and teenagers benefit from explicit work on the "I am a person who doesn't talk" self-concept. This is identity-level CBT that becomes possible once the child can verbalise — and is one of the differences between treating a 5-year-old (almost no identity work needed) and a 14-year-old (identity work is central).
Not all selective mutism is reducible to social anxiety. A meaningful minority of children with selective mutism present without the full social anxiety profile — they are not avoiding social situations broadly, they are specifically not speaking in certain settings. These children typically respond fastest to behavioural treatment alone and rarely need CBT or medication.
Conversely, a meaningful minority of children with social anxiety speak in all settings (just anxiously). These children do not have selective mutism and should not be treated with sliding-in protocols designed for it.
The diagnostic question is therefore not "does my child have social anxiety?" — they probably do — but rather "is the dominant clinical picture the selective mutism behaviour, the broader social anxiety, or both equally?" The answer guides which components of the integrated protocol need most weight, not whether to do them in sequence.
1. Assume comorbidity. Walk into evaluation expecting both selective mutism and social anxiety to be on the table. The literature predicts it; treatment is designed for it. 2. Reject sequential proposals. If a clinician suggests "let's reduce the anxiety first, then work on speech," ask specifically how the anxiety reduction will generalise to the speaking-demand setting. The research does not support this sequencing. 3. Look for an integrated behavioural protocol. Bergman's IBT for SM, Kurtz's PCIT-SM, the SMart Center model, and the McHolm family treatment manual are all integrated approaches. The label varies; the structure converges. 4. Track both axes. Use a speech-tracking measure (frequency of speech in target settings) and an anxiety measure (e.g. a child-anxiety scale appropriate for the age). The two should move together. If anxiety is dropping but speech is not generalising, the protocol needs more behavioural exposure; if speech is generalising but anxiety remains high, the protocol may need more CBT. 5. For older children, plan for identity work in advance. Build in adolescent-led goal setting from session one; do not wait until speech behaviour has shifted to address how the adolescent sees themselves.
---
Unseen Progress publishes long-form caregiver research and builds research-backed daily trackers for the families covered. See the full selective mutism research overview for the complete framework.