Published by Unseen Progress, an independent publisher of caregiver research. Last reviewed 2026-05-10. Part of the child anxiety research overview.
Short answer. Most SSRI side effects parents notice in the first one to three weeks — mild nausea, restlessness, sleep changes, an increase in jitteriness — are transient and resolve as the dose stabilises (Walkup et al., 2008; March et al., 2007). The research-backed answer is rarely "stop the medication"; it is "wait through the activation window, watch the specific red-flag list with the prescriber, and avoid making the medication decision from inside week one."
The Child/Adolescent Anxiety Multimodal Study (CAMS) — the largest randomised trial of SSRIs for childhood anxiety — found that sertraline plus CBT produced the highest response rates (about 80%), with sertraline-alone at roughly 55% and placebo at 24% (Walkup et al., 2008). The Pediatric OCD Treatment Study (POTS Team, 2004) found a parallel result for paediatric OCD: sertraline plus CBT was superior to either alone, and both active arms substantially exceeded placebo. The Treatment for Adolescents with Depression Study (March et al., 2007) extended the picture to comorbid depression. Across these trials, side-effect profiles were generally manageable, with most adverse events being mild, transient, and concentrated in the first two to four weeks.
The FDA black-box warning about suicidal ideation in children and adolescents starting SSRIs is real and should not be dismissed — but its underlying data show a small absolute increase in thoughts, not completions, and the warning has not changed the consensus that SSRIs are first-line pharmacological treatment for moderate-to-severe paediatric anxiety and OCD when used with appropriate monitoring (AACAP Practice Parameter, 2007/2020).
When parents say "the SSRI is making things worse," they almost always mean one of three things:
1. An activation effect in the first 1–3 weeks — the child seems more wired, restless, irritable, or sleepless. This is the most common first-week pattern and usually settles. 2. A GI side effect — nausea, loose stools, appetite change. Typically resolves inside two weeks; taking the dose with food often helps. 3. A felt sense that "this isn't my child" — emotional blunting, less spontaneous play, a flatness the parent has not seen before. This one warrants a clinical conversation, not a unilateral stop.
Each of these has a different research-backed response, and none of them can be evaluated from a single hard day.
SSRIs reach steady-state plasma levels at around four to six weeks; therapeutic effect on anxiety often appears between weeks four and eight (Walkup et al., 2008). Inside the first three weeks, the body is adjusting to the molecule and side effects cluster heavily. "He was more anxious the first ten days than he was before we started — we almost stopped" is one of the most common parent reports in the CAMS open-label literature. Most of these families who held the dose saw the activation settle.
The list that should trigger an urgent call to the prescriber — not a wait-and-see — is short and specific: new or increasing suicidal statements, agitation that escalates rather than settles, mania-like presentations (racing speech, no need for sleep, grandiosity), rash or allergic features, or severe persistent GI symptoms. The black-box-warning literature (Hammad et al., 2006) is clear that the risk window is the first month and the response is monitoring, not avoidance of the medication entirely.
Some children on SSRIs report a flattening of affect at higher doses (Read et al., 2014, in adult samples; smaller paediatric literature). This is not the activation effect and does warrant a conversation about dose. It is also distinguishable from the activation pattern by timing — blunting tends to emerge after the dose has been stable for weeks, not in the first ten days.
The honest comparison is not "child on SSRI vs. ideal child" — it is "child on SSRI vs. child with untreated moderate-to-severe anxiety." Untreated paediatric anxiety carries its own substantial functional cost: school avoidance, social withdrawal, comorbid depression risk, and the long-term consolidation of avoidance behaviour. The CAMS team and POTS team both frame the side-effect conversation against this baseline, not against zero.
From recent parent threads:
The third quote captures the central measurement problem: at therapeutic dose, the child is genuinely calmer, which can be misread as "not herself." The first quote shows the activation window. The second is the cost-benefit math parents do at week one, before the trial has had time to produce data — and the research-backed answer is that week one is not enough trial.
1. Agree with the prescriber, before starting, what the activation window looks like and what specific red flags would trigger an early call. 2. Log side effects daily for the first four weeks with timing — first-week activation looks different from week-four blunting, and unaided memory blurs the two. 3. Re-evaluate at the four-to-six-week mark, not earlier — that is when therapeutic effect on anxiety is expected to be visible (Walkup et al., 2008). 4. If considering discontinuation, taper with the prescriber. SSRI discontinuation syndrome in children is real (Foa et al., 2002, paediatric SSRI safety review) and abrupt stops can rebound anxiety symptoms.
---
Unseen Progress publishes long-form caregiver research and builds research-backed daily trackers for the families covered. See the full child anxiety research overview for the complete framework.