Exposure and response prevention

1. Introduction

Version followed: Foa, Yadin & Lichner (2012), Exposure and Response (Ritual) Prevention for OCD: Therapist Guide, 2nd ed. (Oxford, Treatments That Work). Its session structure is used as the spine. Component detail comes from its predecessor, Kozak & Foa (1997), Mastery of OCD: Therapist Guide, and from the Penn CTSA programme formats. Optional inhibitory-learning refinements follow Craske et al. (2014) and Jacoby & Abramowitz (2016), and are labelled "IL variant" wherever they depart from the habituation-based manual. [source: OUP and Psychology Tools listings of the 2012 guide; library TOC of Kozak & Foa 1997] [uncertain: the full text of either manual is not drawn on, only their tables of contents and secondary descriptions. Procedural detail beyond those headings is marked accordingly.]

Exposure and response (ritual) prevention (ERP, EX/RP) is the core behavioural treatment for obsessive-compulsive disorder (OCD). It combines two parts. First, the client deliberately confronts the situations, objects, thoughts and images that trigger obsessional distress (exposure: in vivo and imaginal). Second, they stop the compulsions, avoidance and neutralising that normally end that distress (response or ritual prevention). It began with Victor Meyer's 1966 inpatient work at the Middlesex Hospital, London. He exposed two severe cases to their most upsetting triggers while staff blocked their rituals, for example by turning off the water supply for a washer. It was then made suitable for outpatients by Rachman, Hodgson and Marks at the Maudsley, and was manualised by Foa and Kozak at Penn. [source: ocdhistory.net; Wikipedia ERP; Meyer 1966]

How change happens: in the traditional model, rituals and avoidance work as escape. They prevent the client from learning that feared outcomes do not happen and that distress subsides or can be borne without them. Prolonged, repeated exposure with rituals blocked activates the fear and supplies information that contradicts it. In the original habituation and emotional-processing account, this reduces distress within and across sessions. In the inhibitory-learning account, it builds new "cue → no harm / I can tolerate this" learning that competes with the old fear memory. Fear tolerance and violated expectancies are the targets, and fear reduction is a welcome by-product. Meyer framed it as "modification of expectations": the client learns that harm does not follow when the ritual is not done. [source: Craske et al. 2014; Jacoby & Abramowitz 2016; Meyer 1966 title; Foa & Kozak 1986 is not drawn on, so the emotional-processing summary is from the secondary literature [uncertain]]

2. Framework

[source: Psychology Tools TOC of Foa et al. 2012; Abramowitz, Foa & Franklin 2003; Penn CTSA intensive packet 2015; Penn CTSA OCD treatment page; Kozak & Foa 1997 TOC]

Formats documented by the developers' centre (Penn CTSA):

• Intensive: 15 sessions on every weekday for 3 weeks, about 2 hours each, following the Kozak & Foa (1997) manual (Abramowitz et al. 2003). The 2015 Penn intensive programme describes 17 × 90-minute sessions, with up to two therapist home visits and follow-up phone calls over several weeks afterwards. Out-of-town clients bring a friend or relative for at least the initial period.
• Twice-weekly: 15 sessions over 8 weeks, plus one 5–10 minute phone check between sessions to review homework (Abramowitz et al. 2003). Penn's standard outpatient option is 15–20 × 90-minute sessions, twice weekly for 8–10 weeks.
• Weekly: 15–20 weekly 90-minute sessions for relatively mild OCD (Penn CTSA page).
• Intensive treatment produced more improvement by week 3, but outcomes converged with twice-weekly treatment at 3-month follow-up. Choose intensive treatment for severe cases, poor insight, difficulty with response prevention, or heavy family accommodation (Jacoby & Abramowitz 2016).
• Other formats from the lineage: couple-based ERP, family-based ERP for children, group, telehealth. These are covered as adaptations in section 4.

Phases (2012 guide structure):

• Pace: Kozak & Foa's chapter on the middle phase ("From bête noire to paper tiger") centres on confronting the most difficult items within the main exposure phase rather than leaving them to the end. In the intensive version, the top items are typically reached around the middle of the programme. [uncertain: exact timing not verified]
• Typical exposure-session layout: review self-monitoring and homework → imaginal exposure → in vivo exposure → set homework for the next 24–48 h. Older Penn protocols split the time roughly half imaginal and half in vivo when both are used. [uncertain: from secondary recall]
• Criterion for advancing an item (manual version): distress during the exposure falls substantially, and across repetitions the client starts the item at a clearly lower level. IL variant: advance once the specific feared outcome has been tested and the client's rated expectancy of it has fallen. A fall in distress is not required (Craske et al. 2014).

3. Guiding principles

[source: Kozak & Foa 1997 TOC headings; Abramowitz et al. 2003; Craske et al. 2014; Jacoby & Abramowitz 2016; Abramowitz et al. 2018. One-line meanings are my practical glosses [synthesis].]

• Exposure must match the actual fear. Choose stimuli that evoke the client's own obsessional fear and feared consequence, not a generic version of it. A checker who fears causing a fire needs to leave appliances unchecked, not just to touch "dirty" things.
• Prolonged, repeated and without escape. Stay with each exposure long enough for the learning to happen. Repeat it daily, in session and as homework.
• Ritual prevention is total and continuous. The rules apply 24 hours a day, including covert neutralising and reassurance seeking, not only during sessions. Partial ritualising undermines the exposure.
• Full engagement, no distraction. The client attends to the trigger and to what actually happens. Distraction, rationalising and "safe" mental tricks are treated as avoidance.
• Understanding before persuading. Assess thoroughly and build a shared model before asking for any exposure. Treatment starts with "understanding and persuading" the client (Kozak & Foa 1997, ch. 7).
• Consistency. Follow the plan agreed in session 2. Ad-hoc changes, usually in the direction of avoidance, erode progress (Kozak & Foa 1997, ch. 13 "The importance of consistency").
• The client keeps control. Every exposure is chosen with consent. The therapist encourages but never forces, and offers intermediate steps rather than dropping an item (Kozak & Foa 1997, ch. 17 "Emphasizing the client's control").
• Graded but brisk. Begin with moderate items and move steadily upward. Don't stall on easy items for long.
• No reassurance from the therapist. Answering "Is it safe?" or "Did I lock it?" is a ritual by proxy. Return the uncertainty to the client.
• Address the feared consequence. Where a feared outcome cannot be reached in vivo (e.g., catastrophe, damnation, "becoming" someone), use imaginal exposure.
• Tolerance, not elimination, of anxiety (IL). Frame success as being able to feel anxiety, disgust or uncertainty and carry on, not as feeling calm. Don't describe anxiety as dangerous or as something treatment will remove (Jacoby & Abramowitz 2016).
• Violate expectancies (IL). Design each exposure so that the client's specific prediction is tested as strongly as possible. Hold off cognitive reassurance until after the exposure (Craske et al. 2014).
• Vary and generalise. Use multiple stimuli, settings and states, practise with and without the therapist, and end with "theme and variations" (Kozak & Foa 1997, ch. 15; Jacoby & Abramowitz 2016).
• Remove safety behaviours and safety signals. These include objects, people and checks that make the exposure feel "safe".
• Homework is the treatment. Daily self-exposure between sessions carries most of the dose.
• Recruit the social system. Reduce family accommodation and turn significant others into coaches rather than ritual partners (Kozak & Foa 1997, ch. 11; Abramowitz et al. 2018).
• Return to normal behaviour. End treatment by teaching non-OCD norms for washing, checking and tidying, not permanent avoidance of these behaviours (Kozak & Foa 1997, ch. 15).
• Measure. Track self-monitoring of rituals, in-session distress ratings (SUDS, 0–100) or expectancy ratings, and periodic clinician-rated OCD severity.
• Never real danger. Exposures create uncertainty, not actual risk to anyone. They must be legal, medically safe and ethical. [synthesis]

4. Intervention inventory

Components 4.1–4.16 make up the core manual package (Foa et al. 2012 / Kozak & Foa 1997). Components 4.17–4.25 are inhibitory-learning (IL) refinements, and 4.26–4.29 are format adaptations from the same research lineage. Steps are written in my own words. Example language is illustrative unless a short quote is cited. [source: chapter headings of both manuals; Abramowitz et al. 2003; Penn CTSA documents; Craske et al. 2014; Jacoby & Abramowitz 2016; Abramowitz et al. 2018]

4.1 Information gathering and functional assessment of OCD (Session 1)

• Purpose: map every obsession, trigger, avoided situation, overt and mental ritual, and feared consequence, so that exposures match the fear exactly. The 2012 guide supplies an information-gathering form (appendix); it is not reproduced here.
• Steps: (1) Ask for a typical day from waking, noting each point where obsessions arise and what the client does next. (2) For each obsession, specify the external cues (objects, places, people), internal cues (thoughts, images, impulses, bodily sensations) and the feared consequence ("If I don't wash, I'll get sick / make my child sick / feel this forever"). (3) List passive avoidance, including subtle forms (using elbows, wearing gloves, asking others to do tasks). (4) List rituals, including mental ones: praying, counting, reviewing, "good" thoughts, reassurance seeking, confessing, researching online. Record their duration and frequency. (5) Ask about the client's confidence that the feared consequence would really happen (insight). (6) Ask what family members do to accommodate. (7) Review history, previous treatments and medication.
• Example language (illustrative): "Walk me through yesterday from the moment you woke up. Where did the OCD first show up? What did you feel had to happen next?" / "If you didn't do that ritual, what's the worst thing you think would happen? And how sure are you, 0 to 100?"
• Watch for: shame-driven concealment, especially with taboo, sexual, violent or religious obsessions. Normalise these by naming common examples in a matter-of-fact way. Watch for rituals the client doesn't count as rituals, for overvalued ideas, and for comorbid depression or suicidality, which need assessment before exposure (see 5.9).

4.2 Psychoeducation and treatment rationale (Sessions 1–2)

• Purpose: give a shared model that makes exposure and ritual prevention feel logical, and secure informed consent. Kozak & Foa call this "persuading the client".
• Steps: (1) Describe OCD as obsessions (unwanted intrusions that cause distress) and compulsions (acts done to reduce distress or prevent harm). (2) Draw the client's own cycle: trigger → obsession → distress → ritual or avoidance → short-term relief → stronger obsession next time, because the client never learns the fear was unfounded or bearable. (3) Explain the two components: exposure confronts the fear, and ritual prevention stops the escape. Explain why both are needed. (4) Describe what treatment will feel like: distress will rise, the client will not be forced, and the work is gradual but steady. (5) Give realistic expectations, such as substantial improvement for most people who complete it and the importance of daily homework. (6) In the IL variant, frame the aim as learning to carry anxiety and uncertainty, not getting rid of them. (7) Invite questions and objections, and check understanding by asking the client to explain the rationale back.
• Example language (illustrative): "Every time you wash, the fear gets a little vote of confidence. ERP is about testing it instead: you'll touch what scares you and not wash, so that you, not the OCD, find out what really happens." / "I'm not going to promise this will feel calm. I am confident you can learn to handle the discomfort."
• Watch for: the client agreeing on the surface while privately planning to keep "just one" ritual. Watch for a fear that treatment will force them to do something harmful; clarify that exposures never involve real danger. Watch for families who need the same explanation (4.11).

4.3 Self-monitoring of rituals (from Session 1, throughout)

• Purpose: establish a baseline, sharpen awareness of triggers and rituals, and later reveal residual or substitute rituals.
• Steps: (1) Give a simple log with columns for time, situation or trigger, distress (0–100), urge, ritual done (what, how long), or ritual resisted. (2) Ask the client to record as close to real time as possible, not from memory at night. (3) Review the log at the start of every session and praise accurate recording, including recorded lapses. (4) Once ritual prevention begins, use the log to record urges and any slips.
• Example language (illustrative): "Write it down as soon as you can after it happens. Rough notes are fine. An honest log of a bad day is worth more than a perfect-looking one."
• Watch for: logging that turns into a checking or reviewing ritual (keep entries brief and time-limited), and blank logs that signal avoidance or shame.

4.4 Building the exposure hierarchy (Session 2)

• Purpose: produce a graded list of in vivo situations and imaginal scenarios that covers every theme and feared consequence.
• Steps: (1) Explain the 0–100 distress scale (SUDS) and anchor it with the client's own examples for 0, 50 and 100. (2) Brainstorm situations for each obsessional theme, drawing on the items gathered in 4.1. (3) Rate each one. (4) Organise them into one or more lists, with roughly 10–20 items per theme, spanning moderate (around 40–50) to maximum. (5) Plan the imaginal scenarios for feared consequences that can't be produced in vivo (4.7). (6) Mark which items will be done in session and which as homework. (7) In the IL variant, also record the specific feared outcome and its expected likelihood (0–100) for each item. Order items flexibly, by importance to the client's life and values or even randomly, rather than strictly by SUDS (Jacoby & Abramowitz 2016).
• Example language (illustrative): "Let's find things around a 40 or 50, hard but doable, and also be honest about the 90s and 100s. We'll get there, and I don't want us to leave them off the list."
• Watch for: lists that stop short of the real top items, items that are distressing but not relevant to the fear, and hidden conditions ("I could touch it if I could wash afterwards"). Kozak & Foa's sample plans group items by theme. [uncertain: specific sample plans not viewed]

4.5 Ritual (response) prevention rules (agreed in Session 2; in force from Session 3)

• Purpose: block every escape route so that exposure can do its work. Meyer's prototype blocked rituals with staff supervision; outpatient versions rely on agreed rules, self-monitoring and support persons.
• Steps: (1) Re-state the rationale: each ritual done after exposure cancels the learning. (2) Write explicit, individualised rules together, for example: no washing or cleaning beyond an agreed, time-limited, non-ritualised minimum; checks limited to a single ordinary check, or none; no asking for reassurance; no mental undoing. The manuals give sample guidelines, which are not reproduced here. The original intensive protocol applied strict abstinence from washing, with only infrequent short showers. [uncertain: exact rules] (3) Agree what to do when an urge hits: notice it, record it, stay with the trigger, and if possible re-expose. (4) Agree what to do after a slip: record it, then immediately re-contaminate or re-expose to the same trigger to "undo the ritual" (an established ERP practice; [uncertain] that this exact term is the manual's). (5) Decide whether prevention is abrupt (all rituals stopped from session 3, as in the Penn intensive programme) or gradual (by ritual type or time window). This is the clinician's choice for clients who will not accept abrupt prevention. [synthesis] (6) Brief the support person on the rules (4.11). (7) Get the client's signature or written agreement (Kozak & Foa ch. 12, "explicating the contract").
• Example language (illustrative): "For the next week the rule is no hand-washing after touching anything on your list. If you slip, write it down and touch it again. A slip isn't failure; leaving it uncorrected is what matters."
• Watch for: substitute rituals (a new wipe, a quick mental phrase), "unforbidden" avoidance (avoiding instead of ritualising), and loopholes the rules failed to name. Kozak & Foa (ch. 17) list these as standard difficulties (see section 5).

4.6 In vivo exposure (therapist-assisted; from Session 3)

• Purpose: direct, prolonged contact with real feared stimuli while rituals are blocked.
• Before the first exposure (consent and safety): confirm informed consent to the specific task. Check the contraindications: active psychosis, acute suicidality, current mania, substance intoxication, or medical vulnerability relevant to the task (e.g., immunosuppression for contamination exposures, pregnancy, injury or illness). Adapt or defer the task where any of these apply. [synthesis/standard practice] Make sure the task carries no real danger, is legal, and causes no harm to others. Tell the client they can stop at any time; stopping is discussed, never punished. Agree the end point in advance.
• Steps: (1) Pick a moderate item (about 40–50 SUDS) for the first session. (2) Agree the exact task, including how long and how much contact ("hands on the bin lid, then face and hair"). (3) If useful, model it yourself first (4.13). (4) Ask the client to do it and to keep full attention on the trigger and its meaning. (5) Rate distress at the start and about every 5–10 minutes; in the IL variant, rate the expectancy of the feared outcome before and after instead. (6) Stay in contact until distress has clearly dropped (manual version, usually within the session; the Penn packet gives about 1–2 hours as an example of a long exposure), or until the expectancy has been tested (IL variant). Don't end at peak distress just because time runs out unless that was agreed in advance (IL variant). (7) Block rituals during and after the exposure, including "cleaning up". (8) Debrief (see 4.15). (9) Assign the same or a similar item for daily homework.
• Example language (illustrative): "Keep your hand on it. What's the thought right now? … Good, notice it and stay with it. Where's your number?" / "You don't need to like this; you just need to stay in it."
• Working: the client can start the item on later trials with less distress or less belief in the feared outcome, and generalises it to new versions. Not working: distraction, rushing, subtle protection (sleeve over the hand, holding the breath), or immediate covert neutralising. Ask "What are you doing in your head right now?"

4.7 Imaginal exposure (introduced in the session after in vivo starts)

• Purpose: confront feared consequences and obsessional images that can't or must not be produced in reality (catastrophes, harm to others, damnation), or that arise as spontaneous thoughts. Imaginal exposure also prepares the client for hard in vivo items. Kozak & Foa (ch. 9) give guidelines, a model introduction and a sample narrative; these are not reproduced.
• Steps: (1) Choose the scenario from the hierarchy: the feared situation plus its catastrophic consequence, for example having left the stove on and the house burning down, or hitting a pedestrian. (2) Write the script together in the first person, present tense, with sensory detail, thoughts, emotions and the consequence. Don't include reassurance, and don't write a neat resolution; leave the outcome uncertain or bad. (3) Record it as audio (manual practice is a recording replayed by the client). [uncertain: exact medium per edition] (4) Have the client listen with eyes closed and full attention, rating distress periodically. Many clinicians use a loop of about 30–45 minutes; agree the length in advance. [uncertain] (5) Discuss afterwards. (6) Assign daily listening as homework. (7) Revise the script as it loses impact, adding more feared detail. (8) For unknowable feared outcomes, such as going to hell or developing an illness in the future, target uncertainty and tolerability rather than disproof. Combine imagery with situational exposure, for example standing in the pesticide aisle while imagining not knowing whether one will get cancer (Jacoby & Abramowitz 2016).
• Example language (illustrative): "Write it as if it's happening now, and don't let the story rescue you at the end. The point is to practise living with the 'what if' without fixing it."
• Watch for: blunted or detached listening, mentally changing the story, adding reassuring endings, listening while doing something else, and overwhelming dissociation. Shorten or titrate the script if needed and use present-moment grounding. [synthesis] The client may fear that imagining harm makes it more likely (thought-action fusion); name this directly as the fear the exercise is testing.

4.8 Homework: daily self-directed exposure

• Purpose: carry the exposure dose into daily life. In the intensive format, homework follows every daily session. In the twice-weekly format, a brief phone check between sessions supports it (Abramowitz et al. 2003). Kozak & Foa ch. 13 covers "homework instructions".
• Steps: (1) At the end of each session, assign the in-session item, or one just below it, to be repeated for an agreed duration (often about an hour a day, or as long as needed for distress or expectancy to shift). [uncertain: duration varies by protocol] (2) Specify where, when, how and for how long, and what rituals are banned afterwards. (3) Add "carry-over" exposure, such as keeping a contaminated tissue in a pocket all day or leaving a light on when going out. (4) Assign listening to the imaginal recording. (5) Have the client record each exposure: rating at start, peak and end, or expectancy before and after. (6) Review the homework first thing next session.
• Example language (illustrative): "Same as today, but at home: an hour with the bin, then dinner without washing. Note your numbers before and after."
• Watch for: homework done briefly or with safety behaviours, done only on the day before the session, or not done at all (see 5.4).

4.9 Responding to reassurance seeking

• Purpose: remove reassurance as a ritual, whether it is sought from the therapist, family, doctors, the internet or religious authorities.
• Steps: (1) Name reassurance seeking as a ritual in the rules (4.5). (2) Agree a response script with the client while they are calm. (3) When asked, decline warmly, link the question to the OCD, and redirect to tolerating the uncertainty. (4) Teach family members the same response (4.11). Parents and partners learn to give non-ritualistic answers, decline to answer, or ignore the question; they replace reassurance with support (Abramowitz et al. 2018). (5) Treat online checking and confessing the same way.
• Example language (illustrative): "That sounds like the OCD asking for a guarantee. I'm not going to answer it, because we agreed answers keep it going. Can you sit with not knowing for now?"
• Watch for: indirect reassurance ("So most people don't get sick from that, right?"), reassurance through facial expression, and the therapist slipping into explanations of why something is safe. The IL literature cautions against cognitive reassurance before or during exposure because it lowers the expectancy being tested (Craske et al. 2014).

4.10 Home visits and in-situ exposure

• Purpose: transfer gains to the places where the OCD actually operates, and uncover home-specific triggers and rituals. The Penn intensive programme offers up to two therapist home visits when needed and practical.
• Steps: (1) Schedule the visit for after the client has made progress in the clinic, usually mid-to-late treatment. [uncertain: timing] (2) Walk through the home with the client and identify the remaining avoided areas and objects ("clean" rooms, unchecked drawers). (3) Run exposures there, for example "contaminating" the clean bedroom or leaving appliances unchecked while walking away. (4) Plan the remaining home-based homework. (5) Where a home visit isn't possible, telehealth video can be used for in-home exposure. [synthesis]
• Example language (illustrative): "Show me the places you keep 'safe'. Those are the ones we want to spend time in today."
• Watch for: pre-visit cleaning or preparation (ask the client not to), and family members who undo exposures after you leave.

4.11 Support person and family involvement: reducing accommodation

• Purpose: turn significant others from ritual partners into coaches. Kozak & Foa ch. 11 covers "patterns of collaboration" and "interpersonal conflicts", and the final session includes a meeting with significant others.
• Steps: (1) With consent, meet the support person or people early, in session 2 or soon after. (2) Explain the model and the rationale. (3) Identify current accommodation: doing rituals for the client, answering reassurance questions, changing household routines, extra laundry, excuses for missed school or work, avoiding "contaminated" items. (4) Agree a plan to reduce each form of accommodation, ideally in step with the client's hierarchy rather than all at once. [synthesis] (5) Coach the support person's role: remind the client of the rules without force; don't argue, criticise, physically prevent rituals or offer reassurance; praise effort. A couple-based model uses four coaching phases: plan the task and how the partner will help; start it and share thoughts and feelings; manage the anxiety peak; finish, evaluate and praise (Abramowitz et al. 2018, describing Abramowitz et al. 2013). (6) Agree what the support person will do if they see a ritual: a brief, neutral reminder, no policing. (7) Review family conflict that maintains symptoms.
• Example language (illustrative), to a partner: "When she asks if the door is locked, try: 'I know this is hard, and I'm not going to answer. I believe you can handle not knowing.' Then change the subject or offer to sit with her."
• Watch for: hostile, coercive or critical family responses, which raise dropout risk ([synthesis]), and families who escalate distress by abruptly withdrawing accommodation without the client's agreement.

4.12 Confronting the most difficult items; intermediate tasks

• Purpose: reach and repeat the top items, which the client most fears and which carry the most learning. Kozak & Foa ch. 14 covers this ("introducing the most difficult exposures", "tactics for helping", "alternatives to the planned exposure").
• Steps: (1) Announce top items a session or two ahead, so they're not a surprise. (2) Rehearse the task in imagination if needed. (3) Model the task (4.13). (4) If the client refuses, don't drop the item: break it into an intermediate task, for example touching the item with one finger rather than the whole hand, then progressing within the same session. (5) Offer a choice among top items to keep the client in control. (6) Once the client has done a top item, repeat it several times and generalise it.
• Example language (illustrative): "This is the one you've been dreading. We can take it in two steps today, but we're not skipping it. Which part do you want to start with?"
• Watch for: a pattern of endless intermediate steps, where avoidance is being negotiated. Return to the rationale and the client's goals (5.2).

4.13 Therapist modelling

• Purpose: show that the exposure is acceptable and how to do it fully, and reduce the client's sense of being asked to do something abnormal. Participant modelling was part of the Maudsley ERP work in the 1970s. [uncertain: the Rachman, Hodgson & Marks papers are not drawn on; this is stated from the secondary literature]
• Steps: (1) Do the exposure yourself first, calmly and fully (touch the floor, then your face). (2) Narrate briefly what you're doing. (3) Invite the client to copy you. (4) Fade modelling over time, so that the therapist's presence doesn't become a safety signal (4.21).
• Example language (illustrative): "I'll go first: floor, both hands, then my hair. Now your turn, same way."
• Watch for: the client reading your modelling as proof that the task is "safe". That works as reassurance; prefer framing it as "I'm willing to accept the uncertainty too". [synthesis]

4.14 Teaching "normal" behaviour; theme and variations (end phase)

• Purpose: replace strict ritual prevention with flexible, non-OCD habits, and generalise exposure across variants of each theme. This is covered in Kozak & Foa ch. 15: "repetitions and generalizations", "teaching normal patterns of behavior", "rules for 'normal' behavior".
• Steps: (1) Near the end, review which strict rules can relax. (2) Agree specific normal norms with the client, such as washing before meals and after the toilet, an ordinary shower length, or one check of the door on leaving. Base them on what most people do, not on what feels "right". (3) Keep exposing to variants of each feared item in new places. (4) Watch that the new norms don't drift back into ritual length or frequency.
• Example language (illustrative): "What would someone without OCD do here? Let's make that the rule. Not more, and also not total avoidance."
• Watch for: "normal" washing that is still done ritually (sequence, counting), and fear of relapse when ordinary washing resumes.

4.15 Post-exposure debrief and consolidation

• Purpose: make the learning explicit. The manual approach discusses distress decline and mistaken beliefs. In Abramowitz et al. 2003, therapists pointed out mistaken beliefs about catastrophic outcomes during exposure. The IL variant asks the client to state what was learned.
• Steps: (1) After each exposure ask: What did you predict? What actually happened? What surprised you? What did you learn about the feared outcome, and about your ability to tolerate the distress? (Craske et al. 2014; Jacoby & Abramowitz 2016) (2) Have the client write a one-line takeaway. (3) Link it to the next exposure.
• Example language (illustrative): "Before, you said 80% you'd be unable to bear it. You stayed 45 minutes. What does that tell you?"
• Watch for: debriefs that turn into reassurance ("so it's definitely safe") rather than learning to tolerate uncertainty. Debrief only after the exposure, not during it.

4.16 Final session: relapse prevention, self-exposure and follow-up

• Purpose: maintain gains and prepare for lapses. This is covered by Foa et al. 2012 "Final session" and Kozak & Foa ch. 16: "process of relapse", "rules for self-exposure", "stress-management techniques", "meeting with significant others", "new activities and interests", "scheduling follow-up contacts". Penn's intensive programme adds follow-up phone calls over the following weeks.
• Steps: (1) Review progress against the baseline, including a repeat of the clinician-rated severity measure. (2) Explain lapse versus relapse: occasional obsessions and urges are expected, and the response is planned exposure, not rituals. (3) Teach the client to design their own exposures for new or returning triggers, using a short rule set (identify the trigger → plan exposure → no rituals → repeat). (4) Plan for high-risk periods such as stress, illness, life transitions or the birth of a child. (5) Hold a final meeting with significant others about their role from now on. (6) Help the client fill the time freed from rituals with valued activities. (7) Schedule follow-up contacts or booster sessions. (8) Retrieval-cue practice (4.23) belongs here in the IL variant.
• Example language (illustrative): "If the checking urges come back in six months, what's your plan? … Right: write the hierarchy, do the exposure, no checking, and call if it's not shifting in two weeks."
• Watch for: the client stopping all exposure after treatment and starting to avoid again; residual subtle rituals not dealt with; and the loss of structure leading to drift.

Inhibitory-learning (IL) refinements. These come from Craske et al. (2014, general anxiety) and Jacoby & Abramowitz (2016, translated to OCD). Jacoby & Abramowitz call their OCD translation nascent and largely untested in OCD trials, so treat 4.17–4.25 as optional enhancements layered on 4.1–4.16, not as replacements. [uncertain: evidence base]

4.17 Expectancy tracking and violation

• Purpose: design each exposure to test the client's specific feared outcome as strongly as possible, and end it on learning rather than on calm.
• Steps: (1) Before the exposure, ask what exactly the client expects to happen. Include harm outcomes ("I'll get ill") and distress outcomes ("I'll lose control", "the feeling will never stop"). (2) Rate the likelihood 0–100. (3) Choose a task, duration or intensity that makes the outcome seem most likely to the client. For example, the client holds a knife while thinking about stabbing, with the therapist's back turned (Jacoby & Abramowitz 2016). (4) Continue until the expectancy has been violated, not until anxiety falls (Jacoby & Abramowitz 2016). Craske et al. cite a study that continued until expectancy fell below 5%. (5) Re-rate afterwards and debrief (4.15). (6) For unknowable outcomes, test the distress expectancies ("I can't stand not knowing") instead (4.7).
• Example language (illustrative): "What do you predict will happen if you hold the baby while having the thought? How likely, 0 to 100? … Let's set it up so it feels most likely, then find out."
• Watch for: vague predictions ("I'll feel bad"); pin them down. Watch also for pre-exposure cognitive challenge that lowers the expectancy beforehand, which Craske et al. advise saving for after the exposure.

4.18 Variability: varied and non-linear hierarchies

• Purpose: produce learning that is robust and closer to how triggers appear in daily life.
• Steps: (1) With prior agreement, pick items out of order: by importance to the client's values, or at random once the client has done a few moderate items (Craske et al. 2014; Jacoby & Abramowitz 2016). (2) Vary the stimuli, for example bins with and without lids, food waste versus office waste. (3) Vary the duration and intensity, and use exposure lengths that are agreed but not predictable. (4) Don't repeat an item until fear has gone before moving on.
• Example language (illustrative): "Real life doesn't hand you triggers in order. Today we'll draw from the 60s and the 80s, mixed."
• Watch for: refusal or dropout if variability is introduced too early. Craske et al. recommend still starting with the least distressing items.

4.19 Deepened extinction: combining fear cues

• Purpose: strengthen the learning by combining cues that have each been exposed separately.
• Steps: (1) Expose each cue on its own first. (2) Then combine them in a single exposure: external (contaminant), imaginal or verbal (doubt prompts), interoceptive (arousal, e.g., by brisk exercise or caffeine where medically safe). Example: leaving home without checking the stove, while aroused, while imagining not knowing whether it is off (Jacoby & Abramowitz 2016). (3) Another example: for harm obsessions about a baby, move from imaginal exposure, to holding a knife near the sleeping baby, to both together (Abramowitz et al. 2018).
• Example language (illustrative): "You've done the stove on its own and the fire script on its own. Now both: leave without checking and listen to the script on your way out."
• Watch for: combining cues before each one has been exposed alone. Jacoby & Abramowitz note that combining without prior separate exposure showed a strong return of fear in one study.

4.20 Multiple contexts

• Purpose: help the learning generalise beyond the therapy room.
• Steps: (1) Repeat exposures in different locations: clinic, home, work, public places. (2) Repeat them with and without the therapist, alone and with others, at different times of day, and in different moods or physiological states (Jacoby & Abramowitz 2016; Abramowitz et al. 2018). (3) Imaginal exposure can also be done in varied settings.
• Example language (illustrative): "You've done the bathroom door at clinic. This week: work, the gym and the supermarket."
• Watch for: gains that appear only at the clinic. The evidence for this strategy is mixed (Craske et al. 2014).

4.21 Removing safety signals and safety behaviours

• Purpose: prevent the client from crediting "safety" to something other than their own tolerance and the actual outcome.
• Steps: (1) List safety signals: the therapist's presence, a support person, a phone, medication carried "just in case", hand gel in a pocket, gloves, sleeves, distraction. (2) Remove them, at once if the client is willing, or gradually if that is needed to prevent dropout (Craske et al. 2014). (3) Fade the therapist's presence: do exposures with the therapist in another room, then alone.
• Example language (illustrative): "Leave the gel with me today. Let's see what happens without it."
• Watch for: new safety behaviours replacing old ones (ritual substitution, 5.6).

4.22 Occasional reinforced extinction

• Purpose: described by Craske et al. (2014) in general anxiety work. Occasional mild versions of the feared outcome are deliberately included so that the learning survives real-life setbacks. It is not discussed for OCD by Jacoby & Abramowitz (2016). [uncertain for OCD]
• Steps (only if adapted, [synthesis]): (1) Arrange exposures where mildly negative outcomes may really occur and are tolerable, for example social disapproval in a scrupulosity exposure, or a small, real mistake left uncorrected. (2) Never arrange real harm or genuine danger.
• Example language (illustrative): "Let's leave the email with the typo in it and see what happens, including if someone notices."
• Watch for: outcomes that become actually damaging (to job or relationships). Discuss and agree the limits in advance.

4.23 Retrieval cues (relapse prevention)

• Purpose: help the client recall the learning from exposure when facing triggers later.
• Steps: (1) Near the end of treatment, choose a cue that brings back successful exposures, for example a bracelet or mentally recalling a specific exposure (Craske et al. 2014). (2) Use it sparingly and not during early exposures, where it could reduce expectancy violation and act as a safety signal.
• Example language (illustrative): "When the urge comes back, picture the day you sat on the bathroom floor for an hour. What did you learn then?"
• Watch for: the cue becoming a ritual or talisman ("I'm safe if I wear it").

4.24 Affect labelling during exposure

• Purpose: name emotions without trying to change them. Craske et al. (2014) report it reduced arousal and increased approach in one study of spider fear. It is not discussed for OCD by Jacoby & Abramowitz. [uncertain for OCD]
• Steps: (1) During exposure, ask the client to describe the feeling and the thought in simple words ("I feel disgusted; I'm thinking my hands are filthy"). (2) Don't follow it with reassurance or reappraisal.
• Example language (illustrative): "Just say it out loud, what you're feeling and thinking, without arguing with it."
• Watch for: labelling that turns into reviewing, analysing or reassurance seeking.

4.25 Expanding session spacing

• Purpose: schedule practice so that retrieval is effortful, which may strengthen long-term retention (Jacoby & Abramowitz 2016).
• Steps: (1) For clients who are doing exposure and ritual prevention well between sessions, move from twice weekly to weekly to fortnightly. (2) Keep massed or intensive scheduling for poor insight, difficulty with the rationale or response prevention, or heavy family accommodation (Jacoby & Abramowitz 2016).
• Example language (illustrative): "You're running this yourself now, so let's space sessions out and see how the learning holds."
• Watch for: spacing that leads to drift; bring sessions closer again if symptoms rise.

Format adaptations

4.26 Couple-based ERP

• Purpose: for clients whose partner is heavily involved in rituals or accommodation (Abramowitz et al. 2013, described in Abramowitz et al. 2018).
• Steps: (1) Assess OCD and how the couple has arranged their life around it. (2) Present the model and rationale to both partners. (3) Train communication: sharing thoughts and feelings rather than solving problems, and deciding on exposures as a team. (4) Use partner-assisted exposure with the four coaching phases (4.11). (5) Reduce accommodation and replace reassurance with support. (6) Build exposure into daily couple life, such as reusing rooms that have been avoided.
• Example language (illustrative): "Let's plan this one as a team. What would be helpful from you, and what would be accommodation?"
• Watch for: the partner as enforcer, or the partner's burnout. Couple conflict may need separate attention.

4.27 Children and adolescents: family-based ERP

• Purpose: ERP adapted for developmental stage, with the family central. Kozak & Foa ch. 19 covers teaching children and families about OCD and the pace of exposure for children. Parent roles may be consultant, coach or co-client (Kendall 2011, cited in Abramowitz et al. 2018).
• Steps: (1) Psychoeducation for child and parents in age-appropriate language. Externalising OCD as a separate "bully" is common in paediatric protocols such as March & Mulle (1998). [uncertain] (2) Build a hierarchy with the child and give them choice of steps. (3) Use rewards for effort (contingency management) rather than arguing (Abramowitz et al. 2018). (4) Teach parents how to supervise exposure and reduce accommodation. Parent-only accommodation reduction (the SPACE programme, Lebowitz et al. 2014) is an option where the child won't engage. (5) Pace exposures more slowly where needed (Kozak & Foa ch. 19).
• Example language (illustrative): "OCD is being bossy again. What's one small way you can boss it back this week?"
• Watch for: parents forcing exposures; school accommodation; developmental limits on understanding the rationale.

4.28 ACT-informed ERP

• Purpose: an alternative frame for exposure that emphasises willingness, defusion and values rather than anxiety reduction (Twohig et al. 2015, described in Abramowitz et al. 2018). Use it when the client is stuck on "getting rid of" anxiety.
• Steps: (1) Frame exposure as practising willingness to have unwanted experiences. (2) Use defusion, treating thoughts as experiences rather than rules. (3) Choose "behavioural commitments" based on values, without anxiety-reduction goals. Abramowitz et al. (2018) describe metaphors used in this approach; they are not reproduced here.
• Example language (illustrative): "What would you be doing this week if the OCD weren't calling the shots? Let's make that the exposure."
• Watch for: acceptance used as a new ritual ("I'll accept it so it goes away").

4.29 Intensive, group and telehealth delivery

• Intensive: daily sessions for 3 weeks (see section 2). Arrange a support person and local accommodation for clients who travel. Plan home visits and follow-up calls (Penn CTSA 2015).
• Group: the same components delivered in a group. Individual hierarchies are needed, and in-group exposure time per client is shorter. [synthesis]
• Telehealth: exposures in the client's real environment by video, which is useful for home-based triggers. Confirm the client's location and an emergency contact at each session. [synthesis] Phone support between sessions improves adherence in technology-delivered CBT (Kenwright et al. 2005, cited in Abramowitz et al. 2018).

5. Handling problems

[source: problem categories follow Kozak & Foa 1997 ch. 14 ("tactics for helping the clients", "crises unrelated to exposure") and ch. 17 ("concealment of symptoms", "symptom (ritual) substitution", "unforbidden avoidance", "incomplete abstinence from rituals", "handling arguments", "emphasizing the client's control over treatment", "intermediate tasks", "emotional obstacles"), and Foa et al. 2012 ch. 10 "Problems commonly encountered". I saw only these headings, not the text, so the responses under them are my own synthesis consistent with the principles in section 3, unless another source is cited. [synthesis throughout unless cited]]

5.1 Refusal or low readiness to start

• Signs: agrees in principle, but cancels session 3 or bargains to keep rituals. Known risk: high distress, refusal and dropout.
• Responses: (1) Go back to "understanding before persuading": does the client actually own the model? (2) Do a cost-benefit review of OCD versus treatment, tied to the client's values and goals. (3) Use motivational-interviewing style: reflect ambivalence and elicit the client's own reasons. [synthesis; MI-ERP and readiness interventions exist in the literature but are not drawn on] [uncertain] (4) Offer more control: they choose the first item and a gentler start. (5) Consider medication referral or deferring treatment rather than starting half-heartedly. (6) Agree on a short trial period, e.g., two weeks of exposure, then review.
• Illustrative: "It makes sense you'd want to keep the shower routine; it's been your safety for years. What has it cost you? What would you lose by trying it our way for two weeks?"

5.2 Stuck: stalling on low items, endless intermediate steps, or no change

• Responses: (1) Check for hidden rituals, mental neutralising or safety behaviours. This is the most common cause of no change. [synthesis] (2) Check that the exposure matches the core feared consequence; it may target a surface fear. (3) Check the dose: duration, daily frequency, whether homework is real. (4) Raise intensity, combine cues (4.19) or vary contexts (4.20). (5) In the IL frame, stop waiting for distress to fall. If expectancies are shifting and rituals are absent, keep going up. (6) Set a clear agreement to reach the top items by a set session (4.12). (7) Consider intensive scheduling or home visits (Jacoby & Abramowitz 2016).

5.3 Overwhelm: distress far above plan, panic, dissociation, shutdown

• Responses: (1) Don't end the exposure abruptly at peak if it can be avoided. Step down to an intermediate version of the same item and keep the client in contact (Kozak & Foa ch. 14 "alternatives to the planned exposure"). (2) Ground in the present: feet on the floor, describe the room, label the affect (4.24). Don't use relaxation as an escape. (3) For dissociation during imaginal exposure, open the eyes, use shorter segments, and speak in the present tense. (4) Afterwards, review what made it overwhelming and recalibrate the hierarchy. (5) If the client is repeatedly overwhelmed, re-check for comorbid PTSD, depression or other conditions that may need treatment first. [synthesis]
• Illustrative: "This is a lot. Let's not run from it; let's step down a notch. Keep one hand on it and tell me what you see in the room."

5.4 Non-adherence to homework or ritual prevention

• Responses: (1) Review the self-monitoring log non-judgementally. (2) Find the barrier: unclear instructions, too high a step, no time, a family member undoing the work, or the belief that "it only counts with you". (3) Make the task smaller and more specific, with a set time and place. (4) Add between-session phone check-ins; the twice-weekly protocol used a brief call (Abramowitz et al. 2003). (5) Involve the support person (4.11). (6) Do more in-session exposure, and consider the intensive format or home visits. (7) Reinforce partial completion while keeping the standard.
• Illustrative: "Let's work out what got in the way, not whether you tried hard enough. When exactly did it fall apart?"

5.5 Concealment of symptoms

• Responses: (1) Normalise taboo content with matter-of-fact examples. (2) Ask directly about mental rituals, reassurance seeking, and rituals done out of the therapist's sight. (3) Make honest reporting the measure of success, not looking well. (4) Remember that concealment of harm, sexual or religious obsessions often reflects shame and fear of being judged dangerous. Distinguish ego-dystonic obsessions from intent or risk through standard assessment, and say so explicitly to the client when that is appropriate. [synthesis]

5.6 Ritual substitution, "unforbidden" avoidance, incomplete abstinence

• Responses: (1) Add each new ritual or avoidance to the rules as soon as you notice it (hand wipes replacing washing, glancing replacing checking, avoiding replacing ritualising). (2) Cover mental rituals explicitly: counting, praying, reviewing, "good" thoughts, analysing. Counter them with imaginal exposure and by deliberately re-evoking the obsession after a mental ritual. (3) For slips, re-expose immediately (4.5). (4) If abstinence is incomplete, decide together whether to target a stricter rule or one ritual type at a time. [synthesis]

5.7 Arguments, demands for guarantees, "what if this time it's real?"

• Responses: (1) Don't debate the content of the obsession or offer probability proofs, since these become reassurance (4.9). (2) Return to the process: "That's the doubt OCD always brings. What do we do with doubt in this treatment?" (3) Emphasise the client's control over treatment and choices (Kozak & Foa ch. 17), not the therapist's authority. (4) For religious scrupulosity, consider involving a religious authority the client trusts to clarify normal practice, once, not as ongoing reassurance. [synthesis]

5.8 Alliance ruptures

• Signs: the client sees the therapist as cruel or pushy, is angry after a hard session, or feels shame after a failed exposure.
• Responses: (1) Name it early and ask directly. (2) Validate how hard the work is, and acknowledge any of your own missteps (e.g., pushing too fast or poor timing). (3) Re-affirm consent and the client's control. Renegotiate the pace, not the principle. (4) Review the agreed contract together (Kozak & Foa ch. 12). (5) Repair before asking for the next exposure. [synthesis]
• Illustrative: "Last time I pushed harder than we'd agreed, and I'm sorry. Let's look at the plan again together."

5.9 Depression, suicidality, and crises unrelated to exposure

• Responses: (1) Screen for depression and suicide risk at assessment and periodically during treatment, using standard risk-assessment practice. Severe depression can blunt engagement in exposure and may need treatment first or alongside ERP. [synthesis] (2) If suicidal ideation emerges, pause exposure and assess. Collaborate on a safety plan; counsel on restricting access to lethal means (safe storage or removal by family); involve crisis services as indicated. Resume ERP when the client is stable. (3) Take care with exposures involving harm themes when depression and real ideation are both present. Separate ego-dystonic intrusions from genuine suicidal thoughts through careful assessment before building any related exposure. [synthesis] (4) For life crises (bereavement, job loss), Kozak & Foa (ch. 14) list "crises unrelated to exposure". Address the crisis and keep ritual prevention going if possible, then resume exposure. [synthesis]

5.10 Poor insight and overvalued ideas

• Responses: (1) Don't aim to convince the client first. Frame exposures as experiments whose result will be the client's own. (2) Favour massed or intensive scheduling (Jacoby & Abramowitz 2016). (3) Rate belief in the feared outcome and track it. (4) If insight is absent or delusional, assess for psychotic features and coordinate psychiatric care. [synthesis]

5.11 Disgust and "not just right" experiences that don't fade

• Responses: (1) Expect slower decline: disgust and not-just-right feelings habituate less readily than anxiety. Target tolerance, not reduction (Jacoby & Abramowitz 2016). (2) Test the expectancy "I can't stand this feeling". (3) Leave things deliberately asymmetrical, "wrong" or incomplete, and carry the feeling into everyday activities.

5.12 Harm, sexual and religious obsessions: safety and therapist discomfort

• Responses: (1) Exposures use imagery, words, scripts and in vivo proximity (e.g., holding a knife near others with the therapist present, or visiting a playground). Never use acts that harm or break the law. (2) Make sure the clinical assessment clearly distinguishes OCD from paraphilia, violent intent or psychosis before exposures begin, and document it. (3) Seek supervision for the therapist's own reluctance; a hesitant therapist teaches the client that the thought is dangerous. [synthesis; exposure examples from Jacoby & Abramowitz 2016 and Abramowitz et al. 2018]

5.13 Family hostility or over-accommodation

• Responses: (1) Hold a joint session to re-explain the model. (2) Replace criticism with coaching (4.11). (3) Plan accommodation reduction as a gradual, agreed schedule. (4) For children, parent-focused accommodation reduction (SPACE) can be used if the child won't engage. (5) Refer for couple or family therapy when conflict is wider than OCD. [synthesis; SPACE cited in Abramowitz et al. 2018]

5.14 Real-world risk in contamination fears (e.g., healthcare workers, immunocompromised clients)

• Responses: (1) Base exposures on ordinary community norms and on the client's workplace or medical guidance, not on the fears. (2) Never ask clients to breach infection-control rules or medical advice. (3) Coordinate with the client's physician where health is a factor. [synthesis]

5.15 Return of fear or relapse after treatment

• Responses: (1) Normalise lapses as expected. (2) Reinstate the self-exposure plan (4.16). (3) Offer booster sessions. (4) Look for new themes; OCD often shifts content. (5) Use retrieval cues (4.23) and practice in new contexts. Craske et al. (2014) note that return of fear is influenced by time, change of context, stressful events and relearning.

5.16 Dropout during intensive treatment

• Responses: (1) Make sure the support person and logistics are in place before starting (Penn CTSA 2015). (2) Pace the first week carefully. (3) Check alliance and fatigue daily. (4) Have a fallback to twice-weekly sessions rather than stopping altogether. [synthesis]

Sources

Sources

• Exposure and response prevention — Wikipedia
• Exposure and Response (Ritual) Prevention for OCD: Therapist Guide, 2nd ed. — Oxford Academic
• Appendix (Information-Gathering Form) — Oxford Academic
• Therapist Guide product page — Ovid
• Therapist Guide contents — Psychology Tools
• Mastery of OCD: Therapist Guide — table of contents (MCPHS library)
• Penn CTSA OCD Intensive Information Packet (2015)
• Penn CTSA OCD treatment page
• Abramowitz, Foa & Franklin (2003), intensive vs twice-weekly EX/RP
• Craske et al. (2014), Maximizing exposure therapy: an inhibitory learning approach — PMC
• Jacoby & Abramowitz (2016), Inhibitory learning approaches to exposure therapy: translation to OCD — UNC repository
• Abramowitz, Blakey, Reuman & Buchholz (2018), New directions in the CBT of OCD
• Behavior therapy for OCD — ocdhistory.net

Not drawn on, so claims citing them are marked [uncertain]: Hezel & Simpson (2019), Indian J Psychiatry; Foa (2010), Dialogues in Clinical Neuroscience; NICE CG31.

Books and papers cited

• Foa, E. B., Yadin, E., & Lichner, T. K. (2012). Exposure and Response (Ritual) Prevention for Obsessive-Compulsive Disorder: Therapist Guide (2nd ed.). Oxford University Press. (Companion client workbook: Treating Your OCD with Exposure and Response (Ritual) Prevention Therapy.)
• Kozak, M. J., & Foa, E. B. (1997). Mastery of Obsessive-Compulsive Disorder: A Cognitive-Behavioral Approach. Therapist Guide. Graywind/Psychological Corporation.
• Meyer, V. (1966). Modification of expectations in cases with obsessional rituals. Behaviour Research and Therapy, 4, 273–280.
• Foa, E. B., & Kozak, M. J. (1986). Emotional processing of fear: Exposure to corrective information. Psychological Bulletin, 99, 20–35. [uncertain]
• Abramowitz, J. S., Foa, E. B., & Franklin, M. E. (2003). Journal of Consulting and Clinical Psychology, 71, 394–398.
• Craske, M. G., Treanor, M., Conway, C. C., Zbozinek, T., & Vervliet, B. (2014). Behaviour Research and Therapy, 58, 10–23.
• Jacoby, R. J., & Abramowitz, J. S. (2016). Clinical Psychology Review, 49, 28–40.
• Abramowitz, J. S., Blakey, S. M., Reuman, L., & Buchholz, J. L. (2018). Behavior Therapy, 49, 311–322.
• Abramowitz, J. S., Baucom, D. H., et al. (2013). Couple-based CBT for OCD. Behavior Therapy, 44, 395–407. [via Abramowitz et al. 2018]
• Lebowitz, E. R., Omer, H., Hermes, H., & Scahill, L. (2014). The SPACE program. Cognitive and Behavioral Practice, 21, 456–469. [via Abramowitz et al. 2018]
• Twohig, M. P., et al. (2015). Exposure therapy for OCD from an ACT framework. J Obsessive-Compulsive and Related Disorders, 6, 167–173. [via Abramowitz et al. 2018]
• March, J. S., & Mulle, K. (1998). OCD in Children and Adolescents: A Cognitive-Behavioral Treatment Manual. Guilford. [uncertain]
• Rachman, S., & Hodgson, R. (1980). Obsessions and Compulsions. Prentice-Hall. [via ocdhistory.net]
• Marks, I. M. (1981). Cure and Care of Neuroses. Wiley. [via Wikipedia]