An OCD treatment plan sets exposure and response prevention as the primary intervention, then breaks it into objectives a client can actually complete between sessions. Each objective names the trigger, the ritual being prevented, the frequency, and the measure used to track it. Goals describe the life the client wants back; objectives describe the exposures that get them there.
What makes an OCD plan different from a general anxiety plan?
Most anxiety plans are built around reducing distress. An OCD plan is built around tolerating it. The therapeutic work is not calming the client down when an intrusive thought arrives, it is helping them not neutralise it: no checking, no washing, no mental reviewing, no reassurance seeking from the clinician or the family.
That changes how objectives are worded. An objective like "client will report reduced anxiety about contamination" rewards avoidance, because the fastest route to low anxiety is more washing. An objective like "client will touch the bathroom door handle and delay handwashing for 30 minutes, five days in the coming week" is behavioural, countable, and pointed at the mechanism that actually maintains the disorder.
OCD is also more common than many clinicians assume. The Australian Bureau of Statistics National Study of Mental Health and Wellbeing found that 3.6% of Australians aged 16 to 85 had a 12-month obsessive-compulsive disorder in 2020 to 2022, putting it in the same band as generalised anxiety disorder at 3.8% and panic disorder at 3.7%.
How many sessions will an Australian OCD plan actually get?
This is the constraint that shapes every other decision in the plan, and most treatment plan templates ignore it.
The UK NICE guideline on OCD defines low intensity psychological treatment as up to 10 therapist hours per patient, and reserves it for clients whose functional impairment is mild. For adults with moderate functional impairment it recommends the choice of an SSRI or "more intensive CBT (including ERP; more than 10 therapist hours per patient)". For severe functional impairment it recommends combined SSRI and CBT including ERP.
Under Better Access, eligible clients can claim a Medicare benefit for up to 10 individual mental health treatment services per calendar year. In other words, a full year of subsidised individual therapy in Australia lands at the ceiling of what that guideline calls the low intensity tier, for a condition where moderate presentations are told to expect more.
Three practical consequences for the written plan:
- State the session budget in the plan itself. Sequencing 10 sessions is a different clinical task from sequencing 20, and a plan that pretends otherwise will quietly run out of room somewhere around the hierarchy's middle rungs.
- Make between-session ERP the engine, not homework. If the client completes four exposures a week, the year holds roughly 200 exposure practices rather than 10.
- Write the session 10 decision into the plan at the start: refer on, move to group or self-directed maintenance, continue privately, or return to the GP for a medication review. Deciding this in advance is far easier than improvising it when the sessions have run out.
Which outcome measures belong in an OCD plan?
Three sit at different levels, and a good plan is explicit about which one answers which question.
The Yale-Brown Obsessive Compulsive Scale (Y-BOCS, Goodman and colleagues, 1989) is the severity benchmark. It is clinician-administered, 10 items scored 0 to 4 for a total range of 0 to 40, with separate subtotals for obsessions and compulsions, and it was deliberately designed so the score does not depend on which obsessions or compulsions the client has.
The Obsessive-Compulsive Inventory-Revised (OCI-R, Foa and colleagues, 2002) is the self-report workhorse: 18 items across six subscales, with empirically derived cut scores that distinguished people with and without OCD across 215 patients with OCD, 243 with other anxiety disorders and 677 non-anxious participants.
The Obsessive Beliefs Questionnaire-44 (OBQ-44) measures the beliefs underneath the behaviour. The Obsessive Compulsive Cognitions Working Group's 2005 factor analysis produced three factors: responsibility and threat estimation, perfectionism and intolerance for uncertainty, and importance and control of thoughts. Those three factors are an unusually good source of cognitive objectives, because each one names a testable belief rather than a mood.
One Australian wrinkle worth planning around. The MBS descriptor for item 2715 requires a GP preparing a Mental Health Treatment Plan to administer an outcome measurement tool, and names only the K10 and DASS 21 as examples. At review, it requires "re-administration of the outcome measurement tool used in the assessment stage". So an OCD-specific measure sits alongside the referring GP's tool rather than replacing it. Track the Y-BOCS or OCI-R for the clinical work, and keep the GP's measure in the file so the plan review can close its own loop. Our guides to DASS-21 scoring and mental health treatment plan referrals cover that side of the paperwork.
What do OCD goals and objectives look like in practice?
Goals are the client's language. Objectives are yours. Each objective should name a trigger, the response being prevented, a frequency, and how it will be measured.
Goal: Get the morning back so I can leave for work by 8am.
- Build a graded exposure hierarchy of 12 to 15 morning triggers with subjective units of distress ratings, by session three.
- Complete exposure to the two lowest rungs (touching the bin lid, leaving without a final stove check) with full response prevention, four mornings per week.
- Reduce the morning routine from 95 minutes to under 45 minutes, self-recorded daily.
- Eliminate the final door-lock check by week eight, tracked on the daily log.
Goal: Stop needing my partner to tell me it's fine.
- Map all current reassurance-seeking behaviours, including mental reviewing and internet checking, by session two.
- Partner attends one session to agree a scripted response to reassurance requests.
- Reduce reassurance requests from an average of 12 per day to under three per day by week ten, tracked on a tally card.
- Practise responding to an unanswered "what if" with an agreed acceptance statement, daily.
Goal: Have intrusive thoughts without believing I'm dangerous.
- Complete a thought-action fusion experiment in session, then repeat it at home twice weekly.
- Reduce the OBQ-44 importance and control of thoughts subscale, re-administered at session six and at the final session.
- Write and listen to an imaginal exposure script three times per week without neutralising.
- Explain in the client's own words why suppressing a thought increases it, by session five.
Goal: Keep the gains.
- Draft a written relapse prevention plan naming three early warning signs and the exposures that address them, by the final session.
- Schedule two self-directed exposure practices per week for the three months after discharge.
- Agree the trigger for re-contacting the practice, and give the client the GP's review date in writing.
For the structure around these, see our guide to writing a therapy treatment plan and the sibling anxiety treatment plan goals and objectives. PractaLuma is AI-native practice management software for Australian mental-health practices, so care plans, standardised assessments and progress notes share one client record and a goal written at intake can be scored against at review without re-keying it. You can see the full feature set on our features page and current plans on pricing.
What should the plan say about medication?
Psychologists do not prescribe, but the plan should record the medication position and who owns it. The RACGP's OCD guidance for GPs (Johnson and Blair-West, Australian Family Physician, 2013) notes that ERP and SSRIs have the strongest evidence, that the SSRI doses required in OCD are usually higher than those used for depression, that the number needed to treat for SSRIs in OCD has been calculated at six to 12, and that not all SSRIs are subsidised on the Pharmaceutical Benefits Scheme for OCD. It also flags that ERP dropout rates of 25% have been reported, which is a good reason to write engagement itself into the early objectives.
If you are writing back to the referrer about any of this, our template for a psychologist's letter to a referring GP covers the format.
How do you know whether the plan is working?
An international Delphi consensus published in World Psychiatry (Mataix-Cols and colleagues, 2016) settled on definitions that are useful reference points: treatment response as a 35% or greater reduction in Y-BOCS score plus a Clinical Global Impression-Improvement rating of much or very much improved, partial response as a 25% to 35% reduction, and remission as no longer meeting diagnostic criteria or, where a structured interview is not feasible, a Y-BOCS score of 12 or below with a CGI-Severity rating of one or two.
Read those as trial definitions, because that is what the authors recommend them for, not as a clinical pass mark. They are still more defensible than "client reports improvement", and a plan that states its own change threshold up front is easier to review honestly at session six than one that does not.
Watch for the OCD-specific trap as well: a falling symptom score can reflect successful avoidance rather than recovery. A client who has quietly stopped cooking will report fewer stove checks. Always read the score against the exposure log and the client's functioning, not on its own. The same principle applies at intake, which our guide to writing a psychology intake assessment covers in more detail.
Frequently asked questions
How many goals should an OCD treatment plan have? Two to four, each with three or four objectives. With 10 subsidised sessions, a plan with eight goals is a plan that will not finish any of them.
Can I use the Y-BOCS if I am not a psychiatrist? The Y-BOCS is clinician-administered rather than restricted to one profession, and the scale was validated with multiple raters. Training in its administration matters more than discipline, and the OCI-R is a reasonable self-report alternative where a structured interview is not practical.
Do I have to use the K10 if the GP did? For the GP's own plan review, the MBS descriptor for item 2715 ties re-administration to the tool used at assessment. Your clinical measure can be different. Keeping both in the record is the simplest way to satisfy the review and still measure what you are treating.
What if the client refuses exposure work? Make engagement the first objective rather than skipping ERP. Psychoeducation, a values conversation and a single, very low rung exposure will often do more than substituting a different therapy.
Should the family be named in the plan? Where family accommodation maintains the rituals, yes. Write the accommodation behaviours and the agreed replacement responses into the objectives, with the family member's role stated explicitly.
This article is general information for clinicians, not clinical or legal advice. Treatment decisions should follow your professional judgement, your supervisor's guidance, and the current guidance of AHPRA and your professional body.