OCD vs Generalised Anxiety Disorder

Both conditions are made largely of repetitive worrying thoughts, both keep people awake, and both respond to therapy. They are confused often enough that people spend years in treatment aimed at the wrong one.

Generalised anxiety disorder, GAD, is excessive worry about real-life concerns, on more days than not, that is difficult to control and comes with physical symptoms such as restlessness, tension and poor sleep. Obsessive-compulsive disorder, OCD, is a loop: an unwanted thought arrives, causes intense distress, and something is done to neutralise it.

Four differences hold up reliably.

1. The compulsion

This is the one that decides it.

GAD worries. OCD worries and then does something about it, and the something works for a few minutes and has to be repeated.

The trap is that most people picture only the visible kind of compulsion. In practice the something is often entirely mental: replaying a memory, arguing with the thought, silently listing evidence about your character, scanning your body for a reaction, repeating a phrase to cancel an image, or trying to figure it out one more time.

So the question is not "do I do rituals". It is: after the thought, do I perform anything whose purpose is to make the feeling stop, and does it need repeating?

2. What the worry is about

That last row has a clinical name: ego-dystonic, meaning at odds with your sense of yourself. GAD worry is usually the opposite. It feels like an accurate, if exhausting, reflection of a genuinely uncertain world.

3. What an answer does

Give a person with GAD a genuine piece of reassuring information and it often helps for a while. The mortgage is affordable, the scan was clear, the meeting went fine. The worry may return, but the information landed.

Give a person in an obsessive loop the same reassurance and something different happens. It works for perhaps ten minutes, and then a new "but what if" arrives from a slightly different angle, and the reassurance has to be sought again, at a slightly higher dose.

That difference is diagnostic, and it is also why reassurance is treated as a compulsion in OCD rather than as support.

4. The physical picture

GAD tends to come with a persistent, low-grade physical state: muscle tension, restlessness, fatigue, difficulty concentrating, disturbed sleep. It is a body braced more or less constantly.

OCD tends to spike. Intense distress at the moment the thought arrives, sharp relief after the ritual, then a return. People with OCD often describe their day as a series of peaks rather than a plateau.

Neither pattern is universal, but the shape over a day is often more informative than any single moment.

The one test worth running today

If you want something practical rather than a taxonomy, use this. It works in the moment and it does not require a diagnosis.

The distinction that test draws is between productive worry, which identifies an action, takes it and closes, and unproductive worry, which rehearses a catastrophe to stay ready when no action exists or the actions are exhausted. The second one obeys every rule on this site: fed by answers, strengthened by rituals of checking, and quieted by the same skills.

Why the treatments differ, and it matters

It would be easy to conclude the labels do not matter. Underneath both sits one shared process that researchers increasingly call repetitive negative thinking, and a 2025 review made the case for treating it as a process that runs across diagnoses rather than inside any one of them. The core skills genuinely overlap.

But there is one practical place where getting it wrong is expensive.

General talking therapy can make OCD worse. A kind therapist who answers the question "am I sure" for the fortieth time is providing reassurance, and reassurance is the compulsion. This is the single most common reason people describe years of supportive therapy that changed nothing.

The treatment with the strongest evidence for OCD is exposure and response prevention, usually shortened to ERP, in which you gradually approach what you have been avoiding while practising not doing the ritual. Modern research, particularly the inhibitory learning work of Michelle Craske and colleagues, suggests it works by teaching your brain through direct experience that the feared outcome does not arrive, rather than by argument.

So: if the loop pattern above fits, ask for ERP by name, and describe the mental rituals as well as any visible ones.

And if it is both

Having more than one condition at a time is the norm in this area rather than a complication. Around 2.3 percent of people meet criteria for OCD at some point, and a large share of them also meet criteria for an anxiety disorder or for depression.

The good news is that this changes the treatment plan far less than people fear, because the engine underneath is shared. Learning to leave a question unresolved, and to stop discharging discomfort with a ritual, reaches both.

For the wider comparison, including ADHD, depression, bipolar and the fear of psychosis, see is it something else. For the mechanism and the first move, how to stop intrusive thoughts. And the free two-minute self-check will name which pattern your experience fits.