Sertraline enters a blinded autism anxiety trial after years of uncertainty
Adults with an autism diagnosis frequently receive SSRIs for anxiety, yet the evidence base has remained thin where it matters most: in the autistic population itself. STRATA was designed to answer a direct clinical question with a pragmatic, multicentre randomized trial, comparing sertraline with placebo in adults whose anxiety is substantial enough to register on the Generalised Anxiety Disorder Assessment, or GAD-7. The study protocol reflects a familiar problem in everyday practice. Clinicians often reach for an SSRI because options are limited, but the balance between symptom relief and adverse effects has not been defined with confidence in autistic adults.
The trial uses a double-blind, placebo-controlled design across five recruiting centres in the UK and Australia, with allocation at the level of the individual. Adults with an autism diagnosis and a GAD-7 score of 10 or higher at screening are randomized 1:1 to either 25 mg sertraline or placebo. Dosing can then be titrated flexibly up to 200 mg, mirroring real-world prescribing rather than imposing a rigid research-only regimen. That design choice matters because tolerance can vary widely, especially when anxiety treatment is layered onto sensory sensitivity, communication differences, and other features that may complicate medication adherence and adverse-effect recognition.
A single clinical endpoint may decide whether this drug earns a routine place in care.
Trial design mirrors everyday prescribing while preserving internal rigor
STRATA is built as a pragmatic, two-parallel group trial, which means the study is meant to reflect the realities of service delivery rather than an artificially controlled laboratory setting. Participants are followed for 16 weeks for the primary endpoint, with additional assessments at 24 and 52 weeks. Brief safety checks occur at 1 to 2, 4, 8, 12, and 36 weeks, allowing investigators to track tolerability over time as the dose is adjusted. Data collection can be completed online or on paper, and visits may occur by video call, face-to-face, or by telephone according to participant preference. That flexibility is not cosmetic; it lowers barriers for autistic adults who may find in-person research logistics difficult or exhausting.
The main outcome is GAD-7 scores at 16 weeks post-randomisation. Secondary outcomes extend well beyond generalized anxiety symptoms. The protocol includes adverse effects, proportionate change in GAD-7 scores including a 50% reduction threshold, social anxiety, obsessive-compulsive symptoms, panic attacks, repetitive behaviours, meltdowns, depressive symptoms, a composite depression and anxiety measure, functioning and disability, and quality of life. Those domains are clinically relevant because anxiety in autism rarely presents in isolation. It often interacts with repetitive behaviors, shutdowns, sensory overload, and depressive symptoms, making it harder to capture benefit through a single symptom scale alone.
Why sertraline remains plausible, but not yet proven, in autistic adults
Sertraline is a selective serotonin reuptake inhibitor with a long track record in general adult psychiatry, but autism has repeatedly exposed gaps between pharmacologic plausibility and actual benefit. Serotonergic modulation can reduce worry, anticipatory anxiety, and panic symptoms, and it may improve functional capacity when hyperarousal is a major driver of distress. In autistic adults, though, the response may be less predictable. Medication sensitivity, activation, gastrointestinal effects, sleep disruption, and changes in emotional regulation can all influence whether a drug helps or simply adds burden. The trial therefore treats adverse effects as a central endpoint rather than a side issue.
The protocol also includes outcomes that may be especially important in autism clinics, such as repetitive behaviours and meltdowns. Those are not decorative additions; they reflect the way anxiety can manifest behaviorally when verbal description of internal state is limited or when distress escalates into observable dysregulation. Measuring functioning and disability alongside symptom scores should help determine whether any anxiety reduction translates into meaningful daily gains. A lower GAD-7 score would have limited value if it comes with worsened tolerability or no improvement in participation, independence, or quality of life.
Cost-effectiveness and recruitment strategy are embedded from the start
STRATA also incorporates an economic evaluation comparing the cost-effectiveness of sertraline and placebo. In publicly funded systems, that question is inseparable from efficacy. If the drug lowers anxiety but requires repeated monitoring, generates adverse-effect burden, or fails to improve functioning, the net value may be modest. By embedding health economic analysis, the investigators are positioning the study to inform both prescribing decisions and service planning.
A QuinteT Recruitment Intervention is also built into the project to optimize recruitment and informed consent. In trials that rely on nuanced self-report outcomes and sustained follow-up, recruitment quality can shape the validity of the entire dataset. The linked carer sub-study adds another practical layer by assessing carer burden, an outcome that can shift when anxiety symptoms change even if the participant’s own self-reported distress evolves more slowly. Qualitative interviews at different stages of the trial will probe experiences of participating and taking the study medication, offering context for adherence, acceptability, and the day-to-day realities of trial participation.
What this protocol can and cannot answer today
Because this is a study protocol, it does not yet provide efficacy results, safety rates, or comparative cost-effectiveness findings. It does, however, establish a rigorous framework for testing a question that clinicians encounter constantly: whether sertraline meaningfully reduces anxiety in autistic adults without creating a trade-off in tolerability or function. The multicentre structure strengthens external relevance, while the flexible dosing schedule and multiple follow-up time points should improve the chance that the trial captures both benefit and harm in a way that resembles practice.
Registered trial identifiers include ISRCTN15984604, EudraCT 2019-004312-66, and ANZCTR ACTRN12621000801819. The key unanswered issue now is not the rationale for the study, but the magnitude of benefit that will emerge when the blind is broken and the participant-level data are analyzed against placebo over 16 weeks and beyond.