OCD and Comorbid Conditions: Schemas, Rigidity, and Risk

This article is for informational purposes only and does not constitute clinical advice or a substitute for assessment by a qualified professional.

Quick summary

• OCD almost never arrives on its own. Around a quarter of people with OCD also meet criteria for obsessive-compulsive personality disorder (Pozza et al., 2021), and roughly one in five to two in five have co-occurring PTSD (Fontenelle et al., 2012).
• The instinct to “treat the depression first, then do ERP” has weaker evidence behind it than most people assume. Pre-treatment depression severity did not significantly predict the effectiveness of CBT in a meta-analysis of 16 randomised trials (Olatunji et al., 2013).
• Only one thing reliably justifies delaying ERP: risk that cannot be managed while you do it. Everything else calls for adapting ERP, not postponing it.
• Perfectionism — not personality disorder as a whole — was the single trait that predicted poorer ERP outcomes in one trial (Pinto et al., 2011), while a cognitively focused CBT produced better outcomes in the same comorbidity (Gordon et al., 2016).
• NICE is explicit that intrusive sexual, aggressive, or death-related thoughts in OCD are common and are frequently misread as indicating risk (National Institute for Health and Care Excellence, 2005).

What you won’t find elsewhere

Most articles on this topic give you a list of conditions that co-occur with OCD and a vague instruction to “seek integrated treatment”. This one gives you the decision I actually make in the room: a three-question test — unsafe, unlearnable, unfinishable — that separates the one situation where ERP genuinely should wait from the far more common situations where ERP should simply be built differently. It also corrects a statistic that circulates widely in clinical training on this topic and instead reports what the primary sources actually say.

OCD and Comorbid Conditions: The 3 Questions I Ask Before Starting ERP

I have spent the past decade working almost exclusively with OCD, and in that time, I can count on one hand the number of clients who walked in with OCD and nothing else. OCD and comorbid conditions are the norm in my practice, not the complication — and the question I am asked most often by clients and supervisees is which one to treat first.

The honest answer is that “first” is usually the wrong frame. What determines whether exposure and response prevention (ERP) works is rarely the treatment order. It is whether the other condition makes ERP unsafe, unlearnable, or unfinishable — and only the first of those three is a reason to wait.

How often does OCD come with something else?

Co-occurring conditions are the rule in OCD, not the exception. Obsessive-compulsive personality disorder (OCPD) — a pervasive pattern of perfectionism, rigidity, and preoccupation with control and order — is present in around 25% of people with OCD, according to the first systematic review and meta-analysis of its prevalence in this group (Pozza et al., 2021). Post-traumatic stress disorder co-occurs in roughly 19% to 41% of cases, and people carrying both tend to present with a more severe overall clinical picture (Fontenelle et al., 2012). Depression is so common alongside OCD that NICE treats it as a routine part of the assessment picture rather than an unusual finding (National Institute for Health and Care Excellence, 2005).

None of this is news to anyone who works in this field. What is more interesting is how badly the field handles it.

How do OCD and comorbid conditions change ERP?

Less than you would expect, and different from what you would expect. The received wisdom is that comorbidity dilutes ERP, so you should stabilise everything else and then run the protocol on a clean sample. The evidence does not support that story well.

In their meta-analysis of sixteen randomised controlled trials, Olatunji and colleagues found that CBT produced a large advantage over control conditions at post-treatment (Hedges’s g = 1.39), with a medium effect maintained at follow-up (g = 0.43). They then tested whether that effect shrank in people who started treatment more depressed. It did not — pre-treatment depression severity was not significantly associated with a reduced effect size (Olatunji et al., 2013). Baseline OCD severity did not predict a smaller effect either.

Read that carefully, because it cuts against something most of us were taught. It does not mean depression is irrelevant, and it does not mean a severely depressed person will find ERP easy. It means the assumption that depression must be cleared out of the way before ERP can work is doing more clinical work than the data can carry.

NICE makes a related point without fanfare. When a young person’s OCD is not responding, the guideline advises considering additional or alternative interventions for the comorbid conditions and psychosocial factors involved — and then adds that the person will still require evidence-based treatment for their OCD (National Institute for Health and Care Excellence, 2005). Comorbidity changes what you add. It does not cancel what they are owed.

This is one of the most misunderstood aspects of complex OCD, and I have watched experienced clinicians get it wrong — myself included, earlier in my career. I have delayed ERP for months on the grounds of “stabilising” someone, only to realise I had spent that time treating my own discomfort with their presentation rather than their OCD. Meanwhile, the OCD had a free run.

The three questions: unsafe, unlearnable, and unfinishable

Before I start ERP with anyone carrying more than one diagnosis, I ask three questions in order. They sort the situation into “wait” or “adapt” — and the sorting matters more than the diagnosis does.

Is it unsafe?

Would doing exposure work now create a level of risk I cannot manage inside this treatment? This is the only question that can produce a genuine “wait”. Active suicidal intent, current self-injury, I have no plan around, or an eating disorder in a physically unsafe range all mean the same thing: safety first, ERP after. NICE is clear that the risk of self-harm and suicide should be assessed in everyone diagnosed with OCD, with particular attention where depression is also present, and that the impact of the compulsions themselves forms part of that assessment (National Institute for Health and Care Excellence, 2005).

Is it unlearnable?

ERP works by giving your brain new information that contradicts what OCD predicts. If nothing can be encoded, nothing is learned. Heavy dissociation during exposures, intoxication, or a level of anhedonia so flat that no disconfirmation registers — these are capacity problems, and they call for adaptation. Shorter exposures. Grounding is built into the session rather than bolted on. Sometimes a period of behavioural activation runs alongside ERP rather than in front of it. What they do not call for is a six-month queue.

Is it unfinishable?

Can this person actually do the between-session work? Chronic illness, caring responsibilities, a job with no slack in it, or a fluctuating condition all constrain what is realistic. Rigidity constrains it in a subtler way, which I come to below. The answer here is almost never to delay — it is to build a smaller hierarchy that gets finished rather than an elegant one that gets abandoned.

Only “unsafe” stops the clock. The other two change the shape of the work.

Why rigidity is the obstacle people miss

Rigidity interferes with ERP not by making it frightening but by making it feel wrong. This is a genuinely different problem, and it needs a different response.

The evidence here is more interesting than the textbooks admit. In a trial of manualised exposure and ritual prevention with medicated OCD outpatients, comorbid OCPD predicted poorer outcomes — but when the individual OCPD criteria were tested separately, only perfectionism predicted worse outcomes once other factors were accounted for (Pinto et al., 2011). It was not the personality disorder. It was one trait inside it.

And then the picture turns. In a sample of 92 people receiving a more cognitively focused variant of CBT without medication, comorbid OCPD predicted better outcomes (Gordon et al., 2016). Two studies, two directions. The reasonable conclusion is not that one is wrong — it is that whether rigidity interferes appears to depend on what the therapy asks the person to do with it.

That fits what I see. A client whose rules are treated as symptoms to be overridden will dig in, because to them the rules are not symptoms — they are standards, and standards are the things they have been praised for their whole life. A client whose rules are treated as testable predictions will often engage because testing is something perfectionists are good at. The difference is not the client’s rigidity. It is whether I picked a fight with it or recruited it.

Perfectionism also has the property of hiding inside the ERP itself. “Am I doing the exposure correctly?” is a reasonable question the first time and a compulsion by the fifth. So is redoing an exposure that felt insufficiently distressing.

Where schemas fit

Schemas explain why some people experience ERP as an attack on who they are. An early maladaptive schema is a broad, self-referential pattern — built from memories, emotions, thoughts, and bodily sensations, formed early and elaborated across a lifetime — that shapes how a person reads themselves and their relationships (Young et al., 2003).

The clinical use is narrow but real. When someone with a defectiveness schema resists an exposure, the resistance is often not about the feared outcome at all. It is that complying would confirm something they already believe about themselves — that they are the kind of person who gets things wrong. You can run a beautifully designed hierarchy straight past that and wonder why nothing shifts.

I do not treat schemas instead of OCD. I use them to work out why an exposure that should be tolerable is not, and then I design around it.

What the numbers on risk actually say

Suicide risk in OCD is real, frequently underestimated, and frequently mislocated. In a population-based study of 36,788 Swedish patients, people with OCD had substantially raised odds of dying by suicide (OR = 9.83) and of attempting suicide (OR = 5.45) compared with matched controls in unadjusted models (Fernández de la Cruz et al., 2017).

Here is the part that should change practice. Among those in the OCD cohort who died by suicide, 43.5% had no other recorded psychiatric diagnosis — and in the subgroup with no recorded comorbidity at all, the odds of death by suicide were, if anything, higher than in the full cohort (Fernández de la Cruz et al., 2017). Risk in OCD is not simply depression wearing an OCD coat. Screening for it only when depression is on the chart will miss people.

The mirror-image error is just as costly. NICE states plainly that intrusive sexual, aggressive, or death-related thoughts are common in OCD at any age and are often misinterpreted as indicating risk, and advises clinicians who are uncertain to consult someone with specific OCD expertise (National Institute for Health and Care Excellence, 2005). A person terrified by a harmful thought is describing OCD. A person planning harm is describing something else. Confusing the two has consequences in both directions — unnecessary safeguarding referrals for the first, missed risks for the second.

If you are not sure which you are looking at, the distinction is not one to be settled from an article. That is what a structured OCD assessment is for.

One statistic worth retiring

You may encounter the claim that people with OCD are “15 times more likely” to be suicidal. I have not been able to locate a primary source reporting that figure. The meta-analytic estimate is a pooled effect size of Hedges’s g = 0.66 for suicidality in OCD relative to controls (Angelakis et al., 2015), and the population-level odds ratios are the ones above. The real numbers are alarming enough. Inflated ones make the case easier to dismiss.

Key takeaways

• Comorbidity is the norm in OCD. It changes how ERP is delivered far more often than whether it is delivered.
• Only unmanageable risk justifies delaying ERP. Capacity problems and bandwidth problems call for adaptation.
• Perfectionism, specifically, is the trait most implicated in poorer ERP outcomes — and it responds better to being recruited than to being overridden.
• Suicide risk in OCD is not a proxy for depression. Screen for it independently.

FAQ

Should my depression be treated before I start ERP for OCD? Not automatically. In a meta-analysis of sixteen randomised trials, how depressed people were before starting made no significant difference to how much CBT helped their OCD (Olatunji et al., 2013). Severe depression may mean ERP needs pacing differently or running alongside behavioural activation, but “get better first, then we’ll start” is a higher bar than the evidence sets. If your low mood is largely driven by what OCD has taken from your life, waiting can make things worse.

Can I have ERP if I have PTSD as well? Yes, though it needs care. PTSD co-occurs with OCD in roughly 19–41% of cases, and people with both tend to present more severely (Fontenelle et al., 2012). The practical difficulty is that trauma-related safety behaviours and OCD compulsions can look identical from the outside while doing entirely different jobs.

Does having OCPD mean ERP won’t work for me? No. The evidence is genuinely mixed: one trial found comorbid OCPD predicted poorer ERP outcomes, with perfectionism the only individual trait that held up as a predictor (Pinto et al., 2011), while another found comorbid OCPD predicted better outcomes in a more cognitively focused CBT (Gordon et al., 2016). That suggests the approach matters more than the diagnosis. Expect the work to take longer and to spend real time on the rules themselves.

How do I know if my violent or sexual intrusive thoughts mean I’m dangerous? The distress is the clue. In OCD, these thoughts are unwanted, repugnant, and experienced as alien — which is exactly why they generate so much fear. NICE notes that such thoughts are common in OCD and are often misread as indicating risk (National Institute for Health and Care Excellence, 2005). Understanding why they stick is the whole subject of the intrusive thoughts in the OCD explained cluster.

I keep asking my therapist whether I’m doing the exposure right. Is that a problem? Possibly. Checking that you have understood an instruction is reasonable once. Returning to it repeatedly is usually the compulsion that leads to treatment. The way out is the same as with any reassurance compulsion, which is covered properly in how to break the reassurance-seeking cycle.

When to seek professional help

Speak to someone if OCD is taking up more than an hour a day, if it is shaping decisions you would otherwise make differently, or if you have been treated for OCD before and it stalled without anyone explaining why. That last one matters — a stalled ERP course is information, not a verdict.

Start with your GP, who can refer you to NHS talking therapies. To find an accredited therapist directly, use the BABCP register at babcp.com, which lists practitioners accredited to deliver CBT and ERP. OCD Action provides advocacy, information, and support groups.

If you are having thoughts of ending your life, please contact your GP, NHS 111, or the Samaritans on 116 123 (free, 24 hours). If you are in immediate danger, call 999.

Author bio

Federico Ferrarese is a BABCP-accredited Cognitive Behavioural Psychotherapist and a BPS Chartered Psychologist (registration no. 00001005090), with an MSc in Applied Neuroscience. He specialises in OCD and Exposure and Response Prevention, working online with clients across the UK and internationally in English and Italian. Read more about Federico

References:
Angelakis, I., Gooding, P., Tarrier, N., & Panagioti, M. (2015). Suicidality in obsessive compulsive disorder (OCD): A systematic review and meta-analysis. Clinical Psychology Review, 39, 1–15. https://doi.org/10.1016/j.cpr.2015.03.002
Fernández de la Cruz, L., Rydell, M., Runeson, B., D’Onofrio, B. M., Brander, G., Rück, C., Lichtenstein, P., Larsson, H., & Mataix-Cols, D. (2017). Suicide in obsessive–compulsive disorder: A population-based study of 36 788 Swedish patients. Molecular Psychiatry, 22(11), 1626–1632. https://doi.org/10.1038/mp.2016.115
Fontenelle, L. F., Cocchi, L., Harrison, B. J., Shavitt, R. G., do Rosário, M. C., Ferrão, Y. A., de Mathis, M. A., Cordioli, A. V., Yücel, M., Pantelis, C., Mari, J. J., Miguel, E. C., & Torres, A. R. (2012). Towards a post-traumatic subtype of obsessive–compulsive disorder. Journal of Anxiety Disorders, 26(2), 377–383. https://doi.org/10.1016/j.janxdis.2011.12.001
Gordon, O. M., Salkovskis, P. M., & Bream, V. (2016). The impact of obsessive compulsive personality disorder on cognitive behaviour therapy for obsessive compulsive disorder. Behavioural and Cognitive Psychotherapy, 44(4), 444–459. https://doi.org/10.1017/S1352465815000582
National Institute for Health and Care Excellence. (2005). Obsessive-compulsive disorder and body dysmorphic disorder: Treatment (NICE guideline CG31). https://www.nice.org.uk/guidance/cg31
Olatunji, B. O., Davis, M. L., Powers, M. B., & Smits, J. A. J. (2013). Cognitive-behavioral therapy for obsessive-compulsive disorder: A meta-analysis of treatment outcome and moderators. Journal of Psychiatric Research, 47(1), 33–41. https://doi.org/10.1016/j.jpsychires.2012.08.020
Pinto, A., Liebowitz, M. R., Foa, E. B., & Simpson, H. B. (2011). Obsessive compulsive personality disorder as a predictor of exposure and ritual prevention outcome for obsessive compulsive disorder. Behaviour Research and Therapy, 49(8), 453–458. https://doi.org/10.1016/j.brat.2011.04.004
Pozza, A., Starcevic, V., Ferretti, F., Pedani, C., Crispino, R., Governi, G., Luchi, S., Gallorini, C., Lochner, C., & Coluccia, A. (2021). Obsessive-compulsive personality disorder co-occurring in individuals with obsessive-compulsive disorder: A systematic review and meta-analysis. Harvard Review of Psychiatry, 29(2), 95–107. https://doi.org/10.1097/HRP.0000000000000287
Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003). Schema therapy: A practitioner’s guide. Guilford Press.