Yes and no. Anxiety can come back, in the sense that periods of higher anxiety often return across the course of a life, particularly during times of stress, change, or vulnerability. What can also happen, and tends to happen more often for people who have done meaningful therapy, is that the anxiety returns into a different relationship with you. The thing you've been calling better is rarely the absence of anxiety. It tends to be a changed way of responding when the anxiety shows up. That changed response is what holds, even when the anxiety itself fluctuates.
What the research describes
This is one of the questions where the research is reasonably clear, and somewhat reassuring in its honesty. Let me give you the numbers in a way that's useful, then I'll say what I think they mean.
Evidence-based therapy for anxiety produces meaningful improvement in the majority of people who complete a course of treatment, and the gains tend to be durable over time. A recent meta-analysis covering 441 trials and over 33,000 patients found that response rates across the anxiety disorders sit between 32 and 38 per cent at end of treatment, with significant additional groups showing partial response (Cuijpers et al., 2024). For those who respond, follow-up studies indicate that the gains are broadly maintained over time, with modest declines over longer periods (Papola et al., 2024).
A long-term Australian study followed older adults with anxiety and depression for ten years after a course of CBT. Around 70 per cent remained in remission of their primary diagnosis at ten years; relapse occurred for roughly one third (Johnco et al., 2024). The relapse number is real, and worth knowing. So is the remission number.
What the research does not tell you is which of those numbers you will be in. The predictors of maintenance, including depth of post-treatment improvement, continued use of skills, lower comorbidity, are statistical and partial. They influence the odds. They don't decide your outcome.
What "coming back" actually tends to mean
There is a question worth slowing down on here. What is actually meant by anxiety coming back?
In one version, the question imagines anxiety as a thing that was there, was removed by treatment, and could return again as the same thing. In this version, recovery is binary; the anxiety is either gone or back; and the goal of therapy is the complete absence of anxiety.
This version doesn't match how anxiety actually moves through a life. Most people have some anxiety. Some have a lot. The kind of work people do in therapy doesn't usually move them from a lot to none. It moves them to a place where the anxiety, when it shows up, no longer runs things in the way it used to.
The more accurate question, then, is not will it come back, but when it does show up again, will I respond to it the way the old me did? The research and what clinicians observe in practice suggest the answer to the second question is usually no, even when the anxiety itself returns. The reflexive avoidance, the catastrophising, the inability to be with the feeling, these are the patterns that tend to change. The change holds, even when the feeling rises again.
This is the part that doesn't show up in the response-rate numbers, because the response-rate numbers measure symptom reduction. They don't measure the part where the same symptom now produces a different response in the person carrying it.
What helps it hold
A few patterns are worth naming.
What tends to help recovery hold over time, across the research and across what clinicians observe in practice, is continued use of the things therapy taught you. The skills. The noticing. The willingness to approach what was being avoided. Recovery is less a state you arrive at than a way of operating that gets practised, becomes familiar, and stays available even when life gets harder.
What also helps is treating a return of anxiety not as proof that the work failed, but as a signal worth noticing. Anxiety often returns at points of meaningful change or stress, sometimes for reasons that are clear, sometimes for reasons that take a while to see. Treating those moments as information, rather than as a verdict on whether you're really better, is part of what keeps the relationship to the anxiety changed.
A return of anxiety doesn't undo the recovery you did. The recovery lives in the way you respond, and that part doesn't disappear just because the feeling has come back.
What I would say in closing is that will my anxiety come back if I get better? is one of the more honest questions a person can ask before starting therapy. The most honest answer I can give you is that some version of it probably will, at some point. What I would want you to know is that the work people do in therapy doesn't depend on the anxiety being gone forever for it to have been real. The work changes how you meet the anxiety the next time. That is worth doing, and it lasts. Even when the anxiety doesn't.
For how we work with anxiety, what sessions cost, and how to begin, see Working with anxiety in adults.
