OCD is Not a Phobia: An Alternative Conceptualization of OCD

Abstract

While the Diagnostic and Statistical Manual Version 5 (DSM-5) has separated obsessive-compulsive disorder (OCD) from other anxiety disorders, conceptualization and treatment of OCD is still identical to the conceptualization and treatment of phobias. Many differences exist between phobias and OCD justifying a different conceptualization and treatment modality for OCD. Implications of the phobic model for OCD are discussed and its limitations are shown. Ethical and counterproductive forms of treatment for OCD derived from the phobic model are also presented. An alternative conceptualization of OCD, the inference-based approach, is presented to show that the phobic model is not the only one which can be applied to OCD. Advantages of this non-phobic way of conceptualizing OCD is illustrated and discussed.

Until the 1960s, obsessive-compulsive disorder (OCD) was considered untreatable by almost all psychiatrists and psychologists. The first truly effective treatment for OCD developed at that time was exposure. How this came about was summarized by Rachman in his 2005 address to the OCD foundation, “Dr. Meyer applied to humans what studies had shown applied to frightened animals: if they were exposed to what scared them for a prolonged time and prevented from leaving the situation, they became less scared”. As such, exposure, which evolved to exposure and response prevention (ERP), stems from the idea that people with OCD suffer from a phobia, similar to a spider phobic. Much like a phobia, people suffering from OCD face their fears and tolerate the anxiety until the object of fear does not arouse anxiety anymore. The primary implications of the phobic model derived from conditioning models are that phobia occurs through learned association between a stimulus and a source of fear. This, in itself, gives us the required parameters for exposure, i.e. presenting the feared conditioned stimuli so that the person learns that it’s not frightening. Rachman and collaborators (1976) showed in an experiment that exposure of OCD was not effective in and of itself, but instead, that response prevention was the effective component of behaviour therapy.

This treatment (ERP) became widely applied to disorders in which the primary emotion was anxiety and fear, as in anxiety disorders, including until recently OCD. Over the years, exposure was refined to become ERP and the theory behind exposure evolved. This evolution was very well summarized by Abramowitz and Jacoby (2014): Foa and Kozak (1986) proposed emotional processing theory and adopted Lang’s concept of the fear structure to create a model for understanding the mechanisms involved in exposure therapy for anxiety disorders such as OCD. Craske et al. (2008) noted inconsistencies in emotional processing theory and specified that during exposure new non-threat associations are formed that compete with existing threat associations. The inclusion of fear structure to explain ERP brought forth the perceived need for patients to feel anxiety during their exposure, a tenet also supported by inhibition learning theory: “Nevertheless, an important tenet of expectancy violation is that extinction learning is strengthened when the discrepancy between what patients expect and what actually occurs is maximized” (Jacoby and Abramowitz 2016, p. 32). However, as mentioned by Jacoby and Abramowitz (2016) “[…] no empirical studies have addressed this question […]” and to our knowledge this remains the case. Also, a review of the evidence regarding the necessity of clients to feel high levels of anxiety during therapy found that this notion was largely unsupported by empirical investigation (Jacoby and Abramowitz 2016, p. 33). This taps into the crucial issue of the lack of knowledge regarding the process behind ERP. Differently put, it is still unclear why ERP works. Both Foa and Craske (2008) note mixed evidence for the ERP requirement for high initial anxiety as a predictor of success (Asnaani et al. 2016).

Much like ERP, cognitive behavioural therapy (CBT) also addresses OCD as a phobia. It postulates that intrusions are normal experiences and that it is the associations and/or interpretations made of this normal phenomenon which are problematic. As such, it disregards the content of the intrusion just like we treat the nature of the feared stimulus (the spiders) as a natural occurrence in spider phobia. The CBT focus on problematic appraisals of intrusions implies that the fear has a basis in reality but is largely exaggerated (e.g. the house could burn down if the stove was left on, a person could catch HIV if the hands were not washed). As such, it is still unclear whether CBT (in which ERP is a large component) is more effective than ERP alone and a recent mega-analysis supports the notion that CBT and ERP do not differ significantly in effectiveness (Steketee et al. 2018). Finally, CBT focuses on the anticipated consequences in OCD (e.g. the house could burn down) in the same way that we focus on the anticipated consequence in spider phobia (e.g. the spider could crawl on you). So, the target of CBT for OCD (and of the phobic approach to treating OCD) is to learn to tolerate the anxiety rather than eliminating the obsession (which might or might not occur as a consequence of tolerating the anxiety).

OCD as a phobia

Are we right to consider OCD as a phobia? According to the Oxford dictionary a phobia is “an extreme or irrational fear of or aversion to something”. While this can apply in some cases (e.g. fear of dirt on tables), it is not applicable in all cases (e.g. obsessions concerning symmetry). Phobia and OCD differ according to how stimuli are processed. In phobia, the complete category of object is feared (e.g. big and small spiders are feared), while in OCD specific types of objects are feared, mostly for their symbolic meaning (e.g. germs may be feared on glue and mud, but not on door handles). When asked, people suffering from OCD have different reasons for fearing different stimuli (e.g. “germs on sticky surfaces, such as glue, are more likely to stay on the skin than germs on non-sticky surfaces, such as door handles”) and it is rare that two people with the same obsession share the same reason for their fear. In phobia, the severity is graded according to the physical threat posed by the feared object (e.g. bigger spiders are more feared than small spiders), while in OCD, severity is not graded according to physical threat (e.g. errors in a text may be more anxiety-provoking than errors in a bank account). These selective compulsions take self-sabotaging proportions where the person is, for example, so concerned with having the hands clean that skin infections result from excessive washing. In phobia, the dominant emotions are fear and anxiety, while in OCD other emotions are also present, such as disgust and shame. Distress in OCD sometimes occurs as a result of the shame people feel regarding their obsessions and who they think they might be, demonstrating that OCD might generate other emotions more painful than anxiety. In phobia, selective perception is heightened (e.g. specific parts of spiders may be focused upon such as legs and fangs), while in OCD perception is often ignored or distrusted and replaced with perseverative staring (van den Hout et al. 2008) or other invented rules (e.g. my hands are clean after ten consecutive washings). This distrust encourages the endless repetition of compulsions and may explain why the obsession is never resolved since the more the person rehearses the obsession, the more they are removed from reality and the senses. Finally, while the danger is often exaggerated in phobias (e.g. very few people actually get killed by a spider, but some species of spiders are still very dangerous), the danger in OCD is very often imagined (e.g. even though I washed it 25 times in a row, germs could still remain on the door knob). This explains the often-perceived senselessness and ego-dystonicity of obsessions where people with OCD themselves “can’t believe that they’re having such thoughts’’. In addition, in some cases although the primary obsessional doubt is imagined the consequences could be realistic if the initial doubt were grounded (e.g. leaving the scene of a supposed hit and run accident is illegal).

From this phenomenological analysis, it is clear that the experience of OCD and the experience of phobias differ. Beyond feeling anxiety and fear, every other aspect of the experience of both disorders is different. So why does the same treatment work for both disorders.

Exposure works… but how?

A meta-analysis made by Fisher and Wells (2005) show that while 15% of OCD clients receiving ERP become asymptomatic following treatment, 15% do not show any change and the remaining 70% still show symptoms. These results were replicated by a more recent meta-analysis by Öst and collaborators (2015). While there is disagreement on how satisfactory these numbers are, by most criteria ERP is an effective treatment (Tolin 2010, Gunter and Whittal 2010). If OCD is not a phobia, then why do some people improve using a treatment for phobia? Traditional learning theories do not appear to be implicated in the effectiveness of ERP. There is little to no consistent evidence of associative learning between the development of obsessions and exposure to fearful events (Emmelkamp 1982). Although there is evidence of trauma triggering OCD (Dykshoorn 2014). Operant learning may explain the maintenance of OCD only (the compulsion successfully decreases anxiety and is thus reinforced). However, accounts of obsessions being reinforced by family and the environment are not sufficient to include operant learning as a developmental mechanism (Emmelkamp 1982). Other types of learning not implicated in phobias are relevant to OCD, such as: vicarious; modelling; or anecdotal. People with OCD often do use their own inductive experience to construct reasons justifying their obsessions.

Cognitive explanations for exposure were hinted at early on its development. The title of Meyer’s original article (1966) “Modification of expectations in cases with obsessional rituals” implies that the aim of exposure is the modification of expectation, a cognitive change. Meyer never attributed the therapeutic gains to exposure itself, but rather to the patient’s reality testing and realization that “anxiety did not reach unchecked levels; the disastrous consequences were not fulfilled” (Meyers 1966). Rachman and Hodgson (1980) also noted that ERP induced secondary changes in the form of cognitive restructuring, raising question as to the mechanism creating change in ERP. Theories underlying ERP have also hinted at cognitions. Emotional processing theory involves change in meaning and inhibition learning change in appraisals (threat expectancies). Finally, successful therapy involves change in cognitions (Wilhelm et al. 2015).

Exposure and ethics

In an article by Olatunji and collaborators (2009) published as a riposte to a New York Times article, describing ERP as a form of torture, the authors argue convincingly that ERP is not a form of torture for a number of reasons and even quote articles from the United Nations convention against torture in support. An argument however it should be noted unnecessary for other forms of psychotherapy. While clearly ERP is not torture, ethical issues need to be addressed with ERP.

There are three principal ethical problems: the first relates to the maxim: ‘’do no harm”. Risk can be present to patients under certain exposure conditions. As Olatunji et al. (2009) admit the therapist cannot control all the consequences of exposure. Advocating, for example, that patients with contamination concerns put their hands in a bucket of fecal matters during exposure is not only nauseating, but can also compromise the patient’s health and confirm his obsessive idea that dirt and germs can lead to unforeseen consequences. A client exposing to a hit and run obsession whilst driving could have an accident. Similarly, a client asked to refrain from superstitious rituals to prevent harm could experience an adverse event by chance. The CBT recommended response to this unforeseen is a kind of soft shoe shuffle where the therapist has to deftly change the target. (e.g.’ but the accident/ bad event/ shame wasn’t as bad as you thought it would be’) (p. 176).

While most psychologists would not deliberately engage in demeaning or counterproductive forms of treatment, it is not possible to foresee all adverse effects of ERP. A young man encouraged to stop his mental rituals supposedly preventing his pet rodent from dying was assured that his pet rodent would survive and that no harm would come to him. Unfortunately, the next day the pet rodent died, and the therapist was put in the difficult position of trying to explain to the young man that his ritual stopping had nothing to do with it.

The second ethical dilemma relates to the common meaning of Kant’s categorical imperative to act universally, namely: “Do unto others as you would have them do unto you” Kant (1785). How many therapists would willingly place their hands in a bucket of waste, or watch hours of pornography?

The third ethical consideration is boundary crossing. In many cases, ERP requires exposure to be done outside of the office, which entails a boundary crossing. This can occur when the therapist accompanies the patient on an exposure and interacts in a dual manner as a companion and professional. Tales of therapists finding themselves in difficult or compromising situations abound: for example, accused of stalking while observing a client, or callousness when standing by as a client shows distress.

The following two cases further illustrate how ERP can be unbeneficial and unethical. The first is the case of a lady who feared putting garbage in the wrong container (i.e. in the recycling container). When probed as to why she had that fear, she said that recycling and saving the earth was amongst her values and that she would not tolerate doing something that could damage the environment. In her case, ERP required that she intentionally placed garbage in the wrong recycling bin and tolerated the ensuing anxiety of misplacing until it decreased. However, is the feared consequence of misplacing rubbish really the problem? The focus on consequences misses the primary obsession that it was her doubt that he she had not placed the items correctly which was worrisome. In other words, is the problem the fact that garbage is placed incorrectly in the recycling or the fact that she could be a careless person disregarding the environment if she had placed the garbage in the wrong container? The ERP exercise not only went against her personal values and caused her distress. Placing garbage in the wrong container was against the law in her country. Again, Kant’s imperative: Would the therapist willingly do this and break the law? The second case is a man who feared becoming homosexual. As such, his ERP therapist instructed him to expose himself to homosexual pornographic content, which he did. Unfortunately, this later became a testing behaviour in which he would watch the homosexual pornographic content to convince himself that he was not becoming homosexual. In the year previous to starting another treatment, he had watched 1600 hours of such videos. Again, this exercise raises all the previous ethical issues, but in addition misses the target: the man was not homophobic but rather doubted his own sexuality and identity, and watching large amounts of pornography distressed him and in a different context could itself be considered an abnormal behaviour. Would the therapist watch such pornography over long periods of time? This exercise also resembles testing behaviour which is a form of OCD ritual.

Testing behaviour

Testing behaviours are mental or physical attempts to disconfirm obsessions arising in OCD. But rather than discrediting the obsession, testing paradoxically gives credibility to the obsession since it usually makes the person doubt more and brings forth a confirmation bias where the person tries every possible test in order to conclude that the obsession is false but thereby confirms that the douibt could be true. Examples include: exposing oneself to situations likely to evoke the obsession, enacting part of the obsessions to see if they’re valid or implementing tests (either physical or mental) to partially test the truth of the obsession. Testing behaviours can also include conjuring up scenes (either in concordance with the obsession or not), rerunning the scene in one’s head, replaying memories to examine what may have happened, conjuring up images, arguing with the obsession, mental or physical avoidance, revisiting a site either in the mind or physically, turning over an element of the obsession to see if any resolution can be had from it and trying to find certainties in the doubt.

All testing behaviours lead to self-sabotage, because they give credibility to the doubt exactly because the person engages in it. A testing behaviour can become the subject of an obsession. For example, a client with OCD who doubts harmful rays can come from stores’ anti-theft detectors, puts his head next to the detector to test if there are harmful rays. His obsessional behaviour then becomes putting his head next to the store detectors, but his main doubt does not concern store detectors, but rather the possibility of harmful rays. Another example is a client with OCD who doubts that the stove may be left turned on based on the doubting premise that he might not have turned it off properly. As a testing behaviour, he checks whether the stove is properly turned off by twiddling the stove’s button. As the client leaves, he now thinks that while the stove was previously properly turned off, maybe now it is not. Testing behaviours encourage more obsessions to occur by giving validity to the reliance on the imagination rather than the senses and encourage obsessions about obsessions, leading to an endless spiral. The importance of differentiating between testing behaviour and the obsessional doubt is that the person often becomes obsessed with an aspect of the testing behaviour stemming from the obsessional doubt. People suffering from OCD often conjure up images which they are apparently unable to dismiss. This in turn then becomes an obsession.

For example, a person who doubts that they could experience a sacrilegious thought conjures up a sacrilegious image which then haunts them as an obsession. As such, people suffering from OCD often complain about the testing behaviour (e.g. an image or another consequence from the obsessional doubt) instead of complaining about the obsessional doubt. However, the target of therapy needs to address the original obsessional doubt.

Disentangling: doubt; testing behaviour, appraisals; and mental compulsions

In general, there is no clear demarcation in the ERP sequence to the occurrence of obsessions, neutralisations, testing behaviours and mental compulsions and appraisals. Is preoccupation with a sinful act an appraisal of a compulsion; the obsession of the result of a testing behavior which conjured up a sinful image, or the result of an unsuccessful attempt to control the thoughts. The products of testing behavior can frequently be confused with the obsessions. An image conjured up to test the possibility of an obsession itself becomes the obsession; as does replaying a memory.

The confusion in an obsession between an obsessional doubts and consequences might also come from a mental compulsion designed to neutralize an obsession. For example, a client with OCD may try replacing a “bad” thought with an image, which will lead the same client to report that his obsession is the image (e.g. replacing doubt about blasphemy with an image of Jesus and reporting that the obsession is the image of Jesus). The same confusion could also arise between an obsessional doubt and a subsequent appraisal. For example, a client with OCD could appraise an obsessional doubt (such as “I might strangle my child”) as meaning that she is “bad” and report the appraisal as the obsession (“I am a bad mother”).

In the phobic model, there is no firm criteria for distinguishing between obsession, appraisal, testing behaviour and mental compulsion, except by questioning; a subjective enterprise. The success in disentangling these processes thus rely on the skills, intuition and training of the therapist as well as client insight.

Looped tapes and scripts

Also, of ethical dubiousness is the unvalidated adaptation of exposure techniques such as loop tapes and scripts. Loop tapes were initially intended by Salkovskis (1983) to illicit mental neutralization in OCD patients so that they could become aware and target the neutralisations. Afterwards they (patients with OCD) were instructed to listen to the loop tape of their obsessions so that they could practice not neutralizing. This technique was developed in response to the lack of ERP effectiveness for those performing mental neutralization. Later, Salkovskis and Westbrook (1989) warned against using loop tapes for habituation or for simple exposure without response prevention. Loop tapes have never been validated alone in a randomised clinical trial (although numerous case studies have been reported using loop tapes; and the technique is recommended in a number of text books). Freeston and collaborators (1997) published an open trial in which loop tapes were used amongst other techniques, such as in vivo exposure. However, this fails to validate whether loop tapes are effective in treatment or whether the effect is due to in vivo exposure. No randomized controlled trials comparing ERP with and without loop tapes have been published.

Scripts were originally intended to expose clients to feared consequences that could not be reproduced in vivo (e.g. the house burning down). The only trials who validated the use of scripts were run by Foa and collaborators (e.g. Foa and Goldstein 1978, Foa et al. 1980) on OCD patients with checking and washing compulsions only as part of imagined exposure used along with in vivo exposure. Since scripts were recorded and used as loop tapes, researchers relied on the trial by Freeston and collaborators (1997) to justify its use. However, this study did not empirically justify the use of loop tapes and scripts.

Furthermore, no examples of scripts were given in the early literature. This is problematic because the only relevant part of the scripts is the content, which was not available. Also, no criteria were defined as to what the scripts needed to include beyond the feared consequence. One of the first examples was found in Abramowitz (2001): “…You’re wondering whether or not Kim is really going to tell these dirty jokes to her children and the uncertainty is making you very anxious. You desperately want to explain that you don’t usually tell such crude jokes, and ask her not to repeat them to any children. God will send you to hell for being such a poor influence unless you stop her. Luckily, as you’re walking to your car, you see Kim. She’s laughing with several other people and you fear she may have told them your jokes. You have to tell them all not to repeat the jokes. But as you get closer, they enter their cars and drive away. Now you’ll never know whether or not other people head these jokes. The uncertainty is unbearable, and you can’t think about or do anything else but worry about going to hell…” (Abramowitz 2001, p. 82). However, the scripts seem to have changed both in content and target over the years. Gillihan and collaborators (2012) give the following example: “I am determined to overcome my OCD and so I decide to stop my OCD-related checking rituals. Each night before I lock the front and back doors of my house and then walk away without making sure I’ve really locked them by turning the handle and tugging on the door; I resist my urges to check multiple times that they are indeed locked. I also refrain from making sure every night that the windows are actually closed and locked. One night, I see on the news that a burglar broke into a house in our town and I have a strong urge to double check that everything is secure, but I resist this urge because the most important thing for me is to get rid of my OCD. So, I tell myself that I have to face the anxiety, live with the possibility that someone could break in, and not check the doors and the windows in order to avoid relapsing into OCD. My wife wakes me up in the early hours of the morning saying that she heard a noise and thinks there is someone in the house. As I’m getting my slippers on to go downstairs, she walks into the hallway to check on our 3-year-old son in his bedroom. She runs into a burglar and screams, waking up our son. He sees my wife struggling with the burglar and starts to cry. The burglar pushes my wife into our son’s bedroom and she falls and hits her head on a chair. The burglar runs down the stairs and out of the house. My wife, who is bleeding from the fall, looks at me through her tears and says, “You didn’t check to make sure that the doors were locked, did you? How could you be so irresponsible and selfish? You are so focused on getting rid of your OCD that you neglect your responsibility to the family”. I feel terrible, and even worse when my son begins to have recurring nightmares about intruders coming into the house. My family counted on me to protect them and I let them down, all because I selfishly tried to tackle my OCD. Now I doubt my family will ever trust me again and I will have to live with the guilt and shame of what I have done” (Gillihan et al. 2012, p. 14). Another striking example is given in Bruce, Ching and Williams (2018): “I have been having the fear of harming my daughter for the last few months. Things got so bad that I came to treatment to help me get over my worries. My therapist told me I had to stop avoiding my daughter because this only made my fears worse. The one fear that haunts me the most is feeling attracted towards her private parts and losing control of myself during diaper changes. I was sure that exposing myself to my daughter’s genitals would only make my fears come true and that I would inevitably hurt my daughter, but I went ahead and tried anyway. Eventually it got easier and I thought I had gotten over the fear. I was able to touch her and wipe her diaper area without looking away and I had no problem with this at all, but then one day I realized while I was doing this that I was becoming sexual aroused. I was shocked and couldn’t believe this was happening. I thought that my therapy was supposed to keep this from happening, but I realized that my therapist did not realize she was working with a true pedophile. The anxiety has come back but I know that since I am getting aroused that this means I am sexually attracted to my daughter! She is getting older and her vagina looks more womanly so it makes sense that I am attracted to her in this way. All of this feels very real but there is nothing I can do about it. I know my wife will not be able to handle the news that I am attracted to my daughter, so I keep it a secret and I keep changing her diapers and I keep getting sexually aroused by this. The last few times, I became so aroused that I was about to lose all control. I was going to jump on her but fortunately my wife walked in and prevented the tragedy. My daughter will definitely be harmed by this, but it is not my fault I am a pedophile. I know that I must try to hold back from jumping on my daughter, but I know any moment it can happen. I will lose control of my body and mind and it will happen. I only hope my wife does not find out. My daughter is too young to realize that I am a pedophile, but now I know I do not really have OCD, I have to accept that I am a pedophile who is on the verge of harming my daughter. There is no turning back now. I actually like the feelings” (Bruce et al. 2018, p. 399).

Comparing the scripts, it is noticeable that the first script is about exposing a person to the possibility that feared consequences could (or not) occur, while the following ones describe detailed consequences that are highly unlikely to occur. The two more recent scripts also make inferences about the person’s character, which, given recent advances in vulnerable self-theme, and feared self, might be a major therapeutic error (Aardema et al. 2017). Both scripts also blame the client for trying to resolve his OCD. Successful exposure to such scripts depends on the person habituating to the sequence and creating enough distance through repetition of the absurdity of the script to dismiss it as unrealistic or as is the current aim of CBT to tolerate the uncertainty with minimal distress.

But one can question the ethical relevance of having a person listen several times a day to a script saying explicitly that he is a paedophile. Also asking someone with OCD who is already invested in the likelihood of remote possibilities, that he or she should just tolerate the uncertainty of becoming a paedophile seems more likely to intensify the obsession. However, there are no reports of adverse events to this procedure. Either there are none or more likely as witnessed in workshops therapists are reluctant to report such events, tending to blame themselves for ineptitude rather than questioning the procedure. Psychoanalysts tended to blame the patients for lack of progress; behavior therapy blames the therapist.

A parallel can be drawn between this and confessional scripts where the hapless accused are forced to repeatedly listen to tapes accusing them of crimes. The scripts can also resemble testing behaviour, noted above, which people suffering from OCD frequently carry out to confirm their fears.

Alternative accounts of OCD onset

If the phobic approach is a poor fit to OCD, are there alternatives to this approach? The following section illustrates how OCD can be conceptualized differently with an approach, focusing on reasoning, which contradicts all the counsels of the phobic model to treatment. The inference-based approach (IBA; see Julien et al. 2016 for a review of the evidence on IBA) conceptualizes OCD not as set of feared outcomes resulting from maladapted beliefs or unlucky conditioning, but as a chain of reasoning errors resulting from maladapted reasoning processes. The emphasis on reasoning processes prevents ethical problems such as asking people to change their values or to act against their values. It also reduces the risk of confirming the feared outcome to zero. Finally, it concords better with OCD phenomenology.

Described briefly, IBA postulates that a perceptual event (such as seeing a doorknob) triggers an obsessional doubt not based in reality (such as “the door might be unlocked”) which leads to feared consequences and maladapted appraisals (such as “maybe I’ll get robbed and this will be terrible”) which brings forth emotions (usually anxiety, but also others such as shame or disgust) that are neutralized with compulsions (such as checking the door). This neutralization reinforces the conviction that the obsessional doubt is real (“why would I have checked the door if it is locked?”), which keeps the OCD cycle going. This identification of the origin of obsession avoids the confusion between initial doubt, consequences, interpretation, neutralization and testing behaviours. In IBA, the obsessional sequence is very clear so that the obsessional doubt is distinguished from the trigger, from the anticipated consequences and from the compulsion and testing behaviour in a straightforward chain (see figure 1).

Figure 1.

Sequence of obsessional events within the IBA model

The IBA has several differences from the phobic model. IBA explains the idiosyncrasy of OCD by looking at the content of cognition instead of ignoring it. This focus on content and idiosyncrasy is manifest by the use of the personal narrative where the person gives the reasons behind their OCD. This emphasizes the importance of rhetoric and language, but also the importance of the imagination. Emotions are also person centred, departing with the traditional “anxiety-fits-all” approach. Instead of trying to overcome “weaknesses and biases”, IBA builds on the existing strengths of the person, such as the ability to reason in non-OCD ways in everyday situations. Both onset and maintenance of OCD is also explained in this reasoning model whereas the phobic model only successfully explains maintenance. The reliance of IBA on reasoning processes to resolve the obsessions also eliminates the requirement of clients to feel anxiety in order to improve.

A crucial difference between IBA and the phobic model is the assertion that obsessions are not normal thoughts coming out of nowhere. While the content between obsessions and ‘normal’ intrusions is often similar, subtle but key differences exist. Obsessions trigger greater emotions than normal thoughts. The modality in which obsessions are experienced include greater visual content and more vivid bodily sensations (Moritz et al. 2018). However, obsessions and OCD related intrusions occur more often without direct evidence justifying their occurrence (Julien et al. 2009, Audet et al. 2016) and are generally referred to as occurring out of context (Fradkin and Huppert 2018). Obsessions occur without a justifying context because they are based on imagined possibilities while normal thoughts are based on reality. Obsessional doubt is pathologically created doubt not normal reality-based doubt. This distinction is key in the treatment of OCD because it is unnecessary to argue with an imagined reality, one must simply stop relying on it and rely on the information already given by reality sensing.

IBA and ERP in the treatment of OCD

The reasoning process posited by IBA can also explain the successes of ERP. According to IBA, whenever the OCD cycle (doubt, consequences, emotions, and compulsions) is broken, improvement should occur. As such, ERP promotes reality sensing (e.g. reliance on the senses and common sense) by forcing the person to feel the anticipated consequences of OCD. In other words, the person faces and accepts reality as is. Reality hits us in the face, without the need for any other special effort and avoidance to perceive it; as it does to the person with OCD in other non-OCD walks of life. Contrary to ERP, IBA proposes to break the OCD cycle by addressing the doubt, which is economical and generates less anxiety. Research has found that a decrease in obsessional doubt leads to a decrease in anticipated consequences, while a decrease in anticipated consequences does not lead to a decrease in obsessional doubt (Grenier et al. 2010, Béland and O’Connor 2014, Perreault and O’Connor 2014). Also of importance to treatment of OCD is the lack of evidence that improvement in OCD symptoms can occur when there is no decrease in obsessional doubt (Aardema et al. 2010).

Inference based therapy (IBT), the treatment derived from the IBA, consist of 9 steps leading to a non-anxious way of changing obsession : (1) distinction between obsessional and normal doubt; (2) explanation of the logic behind OCD; (3) discussion of how the OCD doubt is 100% irrelevant in the here and now; (4) building the narrative; (5) explaining how the participants cross over from reality to the imagination; (6) showing the reasoning devices found in OCD; (7) explaining the selective nature of doubt; (8) discussing the vulnerable self-theme found in obsessional stories; and (9) “learning” reality sensing and how to tolerate the void.

IBT leads to a radically different treatment than ERP, as will be illustrated by returning to our two case examples. Concerning the lady fearing throwing out both her garbage and recycling in the wrong container, the conceptualization would lead to a much different treatment target. The obsessional doubt was “Maybe I could throw garbage in the recycling bin”, leading to the anticipated consequence of “which would do terrible damage to the environment”, generating anxiety, guilt and shame, motivating the client to check to see if the garbage was properly discarded. Instead of asking the person to deliberately throw garbage in the recycling bin, an IBT therapist would examine how the doubt is generated and whether her senses said she properly placed the garbage and therefore, what narrative motivates her need to check. This examination of how the doubt is generated is done without breaking the laws applicable to her country and avoids a counterproductive challenging of her environmental values or of the consequences of polluting the environment. Concerning the man who feared becoming homosexual, his obsessional doubt was “Maybe I could become homosexual”, leading to the anticipated consequence of “and then people would reject me”, generating anxiety, motivating the client to masturbate to homosexual pornography to test whether the obsessional doubt is real or not. Instead of exposing the client to a script describing his supposed homosexual tendencies and watching hours of pornography to overcome his supposed homophobia, an IBT therapist would examine the basis of the doubt that he is homosexual and whether in reality he prefers heterosexual relationships and if so, what story convinces him to doubt his identity. This resolution of the obsessional doubt is made without any potential non-ethical use of exposure and does not risk an exposure exercise becoming a new testing behaviour. An important element of IBT is to underline that the OCD doubt is pathological not normal doubt and is created subjectively not based in reality. Hence going into the doubt removes the person from reality.

IBT has also been found effective in an open trial (Aardema et al. 2017) and two randomized controlled trials comparing IBT with ERP and CBT (O’Connor et al. 2005, Visser et al. 2015). Results have shown that IBT compares favourably with effect sizes reported in meta-analysis of CBT/ERP. The two randomized-controlled trials found IBT to have larger albeit non-significant, effects sizes than either CBT or ERP. IBT is also as effective regardless of level of overvalued ideation (Aardema et al. 2017, Visser et al. 2015), which is not the case with CBT or ERP.

Finally, IBT has also been combined effectively with CBT and ERP in a case series (van Niekerk et al. 2014). The therapy was based on the framework from IBA with IBT techniques used in priority, but beliefs could be challenged when appropriate and exposure was performed when needed. All three participants improved significantly and two showed minimal symptoms after therapy. So, while differing from the phobic model, IBA is compatible with current CBT and ERP and both types of therapy can be used in conjunction successfully.

Conclusion

This article sought to expose the limitation of applying the phobic model to OCD and the implication of this application. While effective, concerns include ethical challenges as well as the imposition of undue suffering on participants. Another model, the inference-based approach, has been presented with the purpose of showing that it is possible to conceptualize OCD successfully without resorting to the phobic model. Successes of the phobic model were explained in light of this new model to illustrate to clinicians the possibility of breaking from the phobic model without compromising patient recovery.