1 Final MBBS, King’s College Hospital, Denmark Hill, London.
2 Consultant Psychiatrist, South London and Maudsley Foundation NHS Trust, Denmark Hill, London.
3 Professor of Mental Health and Cultural Diversity, Health Service and Population Research Department, Institute of Psychiatry, King’s College, London.
Summary: Obsessive Compulsive Disorder (OCD) has similar features comparable with excessive religiosity. Religious rituals in particular have been compared with obsessive rituals. There are differences in frequency and intent. Furthermore, the individuals may not resist religious thoughts or rituals. In this paper we compare these two types of behaviours and draw some conclusions regarding the differences that clinicians may need to be aware of rather than musicalizing normal culturally appropriate behaviour. Literature findings are mixed but it appears that organized religion predisposes to OCD is not upheld. It would appear from studies that specific rituals in organized religion may well have some influence on OCD in terms of symptoms, possible explanations, help-seeking and therapeutic alliance once treatment has been initiated. The relationship between OCD and organized religion may need to be investigated further especially in the context of various religions. This may not only improve our understanding of development of OCD, it may help us to establish how these findings can be used to develop and deliver culturally appropriate and sensitive services which patients may find acceptable with better outcomes.
Religion and psychiatric disorders have had a very mixed history. The use of possession states, speaking in tongues, seeing visions and explanations related to magico-religious beliefs have been present for millennia. Psychiatric conditions including hysteria, anorexia, trances and rituals have played an important role in explaining the aetiology as well as management through the use of religion. However, organized religion, rituals and behaviour vary across monotheistic and polytheistic religions. Thus differences between eastern religions where rituals may be more integrated in daily living and western religions may play a role in seeing some behaviours as pathological. In this paper we review the relationship between religion and obsessive compulsive disorder.
Obsessive Compulsive Disorder
Obsessive Compulsive Disorder (OCD) constitutes one of the anxiety disorders and has a prevalence of 1.5% in the general UK population (Davies, 2009). There is a slight female preponderance in sufferers with a 3:2 female: male ratio and onset occurs most commonly in the third decade (Davies, 2009).
OCD is defined in ICD-10 as two weeks of obsessional thoughts and/or compulsive actions with associated anxiety. The obsessional thoughts of OCD are often distressing and can be in the form of urges, ideas and images (ICD-10). These disturbing thoughts are recognized as the patient’s own regardless of their unpleasant nature. The patient will often try initially to resist these thoughts and this is accompanied by an increase in anxiety levels. As the illness progresses, thoughts will be resisted less and less as the associated anxiety becomes unbearable. The second component of OCD, compulsive actions, is described as “stereotyped behaviours” (ICD-10) which are repeatedly carried out by the patient and are often ritualistic. These may be performed in isolation or may be linked to the patient’s obsessional thoughts. For example, if a patient’s main obsession is that he is unclean; his compulsive action may be hand washing. As with obsessions, an initial attempt to resist carrying out compulsions is usual, but, as above, with illness progression, compulsions are performed more frequently in an attempt to find some relief from symptoms. OCD is often exceptionally distressing for the patient as they become consumed with the demands of their condition to the exclusion of all other commitments.
The aetiology of OCD remains unknown and much research is ongoing. In particular, there is currently a great deal of interest surrounding religion and its contribution, if any, to the development of OCD. There are countless religions across the globe and to discuss them all would be impractical. Thus, this discussion will focus on the three major monotheistic religions - Islam, Judaism and Christianity.
Religion and its link to OCD
Religion is defined as “the belief in and worship of a superhuman controlling power, especially a personal God or Gods” (Oxford English Dictionary, 2006). The putative link between religion and OCD was first put forth by Freud (1961, originally 1907 cited in Tek and Ulug, 2001) who described obsessional neurosis as an “individual religion” and religion as “a universal obsessional neurosis”. This could be argued to be a simplistic assessment of religion but some aspects of religious practice do resemble the common OCD symptoms described above.
Muslims, for example, are required to wash five times per day before prayer. This washing is known as Wudu and is a form of ritual ablution. Parts of the body must be washed in a specific order and in a specific way. If a mistake is made, then the whole process must be begun again (Islam, 2010). This procedure could be said to display remarkable similarity to the ritualistic compulsive acts which characterize OCD. Judaism, especially ultra-orthodox sects, requires that the faithful live by an extensive list of laws which govern all aspects of life as laid out in the Shulchan Aruch (Karo, 1565 cited in Hermesh et al., 2003). If laws are not adhered to or not followed completely then obligations must be repeated until the law is fulfilled. Similarities with clinical OCD are obvious.
Links have been also made between Christianity and OCD. Christianity is not as prescribed a religion as Islam or Judaism. However, there is considerable emphasis placed on the importance of thoughts and their moral significance. Christians may believe that his own private thoughts are equivalent to physical actions. This concept is known as moral thought-action fusion (moral TAF) and is thought to have originated from the Sermon on the Mount where Jesus said:
“Ye have heard it was said by them of old time, Thou shalt not commit adultery: But I say unto you, that whosoever looketh on a woman to lust after her hath committed adultery with her already in his heart”
(Matthew 5: 27-28, King James Version)
Since thoughts are considered equal to actions, they must be atoned for in an equivalent manner. This is similar to OCD where unusual importance is attached to thoughts and where there is often an overwhelming desire to somehow neutralize repugnant thoughts. This theory is in keeping with the work of the Obsessive-Compulsive Cognitions Working Group 1997-2001 (OCCWG) which suggested that there were a number of cognitions that could be linked to OCD including “over importance of thoughts, an inflated sense of responsibility and excessive concern regarding the importance of controlling thoughts” (the OCCWG 1997-2001 cited in Abramowitz et al. 2004). It is possible therefore that people who demonstrate moral TAF- i.e. religious Christians - could be more prone to OCD.
As such, there are several aspects of the three main monotheistic religions which would be said to demonstrate remarkable similarity to the symptoms of OCD. It could even be argued that religion represents a culturally sanctioned form of OCD which is encouraged in order to gain favour with God. If true then it begs the question of whether there is an increased prevalence of patients with strong religious beliefs amongst OCD sufferers and subsequently whether religious people are more at risk of developing OCD in their lifetime.
Evidence supporting a link between religiosity and OCD
Upon examination of the literature pertaining to OCD and religion, it became clear that this is not an extensively researched area. However, there are a number of studies which support a link between religiosity and OCD.
Sica et al. (2002) investigated the relationship between religiosity and obsessive-compulsive symptoms in three groups of Italian Catholics following the work of Rassin and Koster (2003) who found that degree of religiosity correlated with TAF scores. On the basis of these findings, Sica et al. (2002) investigated the link between Catholic religiosity (a religion known to sanction TAF) and obsessive-compulsive symptoms. Three subject groups were used-one “highly religious” (HR) group of 54 nuns and friars, one “moderately religious” (MR) group of 47 individuals involved with their churches and lastly one “low religious” (LR) group of 64 students who scored lowest on a religious beliefs questionnaire. All subjects also completed a number of other questionnaires pertaining to obsessive compulsive symptoms including the Obsessive Beliefs Questionnaire (OBQ). Upon analysis of results, the HR group scored higher than the LR group on key features of moral TAF - over importance of thoughts and responsibility. In addition, OCD symptoms were found to be related to control of thoughts and over importance of thoughts in religious subjects. As such, it seems that religious individuals are more at risk of the cognitive biases proposed by the OCCWG to predispose to OCD and indeed were found to have higher levels of obsessionality. This implies that there may be a putative link between religiosity and OCD in Italian Catholics.
Similarly, Abramowitz et al. (2004) investigated the link between obsessive-compulsive symptoms and Christianity, specifically Protestant religiosity. The subjects consisted of Protestants who were classified into three groups - high, moderate and low religiosity - by means of a three item questionnaire determining religious beliefs (Abramowitz et al., 2004). The HR group were identified to have increased obsessional symptoms and compulsive washing rituals as compared with the LR group. In addition, the HR group was found to demonstrate the cognitive biases identified by the OCCWG (cited in Abramowitz et al., 2004) stated above that are believed to predispose to the development of OCD. Therefore, Abramowitz et al. (2004) reported a link both between religiosity and cognitive biases thought to increase the risk of developing OCD and between obsessional symptoms and compulsive rituals.
The work of Yorulmaz et al. (2009) also supports these findings. In this study, Turkish Muslim subjects were compared with Canadian Christian subjects on religiosity and obsessive-compulsive symptoms. Both groups were found to be equally religious. No difference was found between highly religious subjects from either group and incidence of obsessional thoughts and compulsive actions, implying that “religiosity” broadly predisposes to obsessive-compulsive symptoms. However, it was reported that highly religious Turkish Muslims reported increased intensity of OCD symptoms and more concerns surrounding thoughts (Yorulmaz et al., 2009). This work therefore suggests that there may be a link between religiosity and OCD regardless of religious denomination but also that there may be some differences between groups.
Evidence against a link between religiosity and OCD
Whilst there is some work that supports a link between religiosity and OCD as outlined above, there is also considerable evidence to refute this. There is even some work which directly contradicts the findings discussed in the previous section. For example, like Sica et al. (2002), Siev et al. (2010) also investigated the link between Christian students. In keeping with the results of Sica et al. (2002), Siev et al. (2010) reported that TAF scores were found to be related to religiosity in the Christian group. As expected, no such relationship was found in the Jewish group. Interestingly though, TAF was only found to correlate with obsessive-compulsive symptoms in Jewish subjects and not in Christian subjects. This contradicts the principle findings of Sica et al. (2002) since it implies that TAF is only pathological when it is not culturally sanctioned rather than in groups where it is endorsed and represents a cultural norm.
Further evidence is provided by Steketee et al. (1991). In this study, relevance of religious beliefs was measured in a clinical sample of subjects - 33 with OCD and 24 with another anxiety disorder - and compared to the incidence of obsessive-compulsive symptoms. Religiosity was measured by means of a questionnaire and obsessive-compulsive symptoms were measured by a number of checklists. Following analysis of results, it was found that no particular religion was more common in OCD patients than in other groups and, most crucially, that religiosity was not significantly greater in OCD patients as compared with other subjects.
Hermesh et al. (2003) generated similar findings in their study which looked specifically at the connection between Jewish religiosity and OCD. Three subject groups (OCD patients, panic disorder and a control group of general surgical inpatients) were recruited and their religiosity and OCD symptoms were measured. Subsequently, it was found that the OCD group did not contain significantly more religious subjects compared to either of the other groups, rejecting the notion of a link between religion and OCD.
These findings were correlated by studies using non-clinical samples. Assarian et al. (2006) measured religiosity and obsessive-compulsive symptoms in a group of Iranian students and found no link between religiosity and OCD. Similarly, Lewis (1994) investigated the link between obsessional behaviour and religious practices in a non-clinical sample of students. Whilst no evidence was found to support a link between religiosity and OCD, a significant relationship was found between religiosity and obsessive-compulsive type traits. Further work is obviously required to determine if traits predispose to symptoms but since this remains unsubstantiated, it must be concluded that there is no link to OCD.
Difficulties in analyzing the results of studies
There is evidence both for and against the argument that increased religiosity predisposes individuals to developing OCD. However, it is challenging to draw conclusions from the work outlined above due to several problems with the studies themselves which make the results hard to interpret.
When investigating the link between OCD and religiosity, it is crucial to have a universally agreed definition of religiosity. “Religiosity” could be defined as strong religious beliefs or strict adherence to religious practices, or, a combination of the two. The importance of this point is highlighted by the disparity of results seen with the work of Abramowitz et al. (2004) and that of Hermesh et al. (2003). Abramowitz et al. (2004) defined religiosity as strength of religious attitude and measured this by means of a three item questionnaire. Conversely, Hermesh et al. (2003) defined religiosity as encompassing both religious attitudes and religious practices. A suitable explanation for these results may be that Protestant religiosity is linked to OCD whereas Jewish religiosity is not. However it is hard to tell whether the subject groups are comparable in terms of religiosity despite their different faiths. It may be that one study used subjects of a low religiosity whereas the other used highly religious subjects. It is therefore difficult to separate the effects of different religions from those of religiosity. Whilst a universal definition of religiosity that could be applied across religious groups would be ideal, it is likely that the meaning of religiosity will vary across groups. For example, Smart (1999) commented that Jewish religiosity “is characterized more by orthopraxy than orthodoxy” (Smart, 1999 cited in Siev and Cohen, 2007). This implies that religious practice is more important in Judaism than actual belief (Siev and Cohen, 2007). On the other hand, Christianity places much more emphasis on the importance of religious beliefs (Siev and Cohen, 2007). Consequently, it may be more accurate to measure religious attitudes when trying to determine levels of religiosity in Christians but religious practices in Jewish people. Equally there may be other religions in which both aspects are as important as the other-e.g. Islam. It may also be worth considering whether definitions apply uniformly across religions with a number of subgroups. In short, definitions of religiosity specific to each religious group should be determined and applied universally to obtain uniformity across studies. This will be especially important for studies which compare different religious groups.
Similarly, following establishment of agreed definitions, it will be crucial to develop a uniform method of measuring religiosity. In the studies cited above, religiosity was both defined and measured in different ways. Many of the tools used were developed specifically for the study in question and so were subsequently not used in any other study. For example, whilst Steketee et al. (1991) measured religiosity by means of a three item questionnaire, Yorulmaz et al. (2009) measured religiosity using the Religiousness Screening Questionnaire, a different tool developed by the group. Whilst the validity of these tools is not in doubt, it is difficult to determine if they are measuring similar parameters. A tool to measure religiosity has been developed-the Penn Inventory of Scrupulosity (PIOS) and this may be appropriate for use. However, more research is needed to determine its applicability across religions. If the PIOS is deemed to be inappropriate, an alternative tool should be developed. In addition, it is also important to consider the effectiveness of self-reporting. Whilst it may be difficult to rate the religiosity of a subject, studies may then be more objective and more suitable for comparison.
An additional problem encountered concerned the subjects used by different groups. Specifically, some studies recruited clinical participants whilst others used a non-clinical sample. It remains debatable whether findings from studies using these two subject groups are comparable. In addition, the validity of a study which aims to investigate the link between religiosity and OCD but uses non clinical, supposedly healthy, subjects must be questioned. In reality, such studies are more likely to be measuring obsessive-compulsive traits rather than OCD. It is interesting that all of the studies which reported positive findings used non-clinical samples whereas those reporting negative findings were a mix of clinical and non-clinical samples. This may suggest that whilst OCD is not related to religiosity, religiosity may be related to obsessive-compulsive traits, in keeping with the conclusions of Assarian et al. (2006). Further work is needed to determine if obsessive compulsive traits ever decompensate to OCD and if so, the effect religion has on this. It would also be helpful to repeat studies reporting a link with traits using clinical subjects and vice versa.
Sampling of clinical subjects
However, whilst clinical subjects would be preferred, obtaining a representative sample of patients with OCD is difficult. Sampling from secondary care means that only those patients who have sought help from medical practitioners are represented. This may be especially relevant as highly religious people may be more likely to turn to their religious leaders for help rather than to healthcare services. Secondly, sampling from secondary care may mean that only patients with moderately severe OCD are studied and those with milder illness who are treated in primary care or inpatients with the most severe disease are not taken into consideration. Therefore, without further work all that can be included currently is that any perceived link found between religiosity and OCD applies only to OCD warranting secondary outpatient care.
Summary of difficulties encountered
In summary then, there are a number of problems associated with the studies investigated. This makes findings difficult to interpret. Whilst it seems there may be a slightly stronger case for the argument that there is no link between OCD incidence and religiosity, it is impossible to reach a firm conclusion until the above problems have been addressed. If a link is found to be supported by the literature, it would also be interesting to distinguish cause and effect – i.e. to determine if religion leads to obsessive compulsive traits or if those with inherent obsessive-compulsive traits are more likely to be religious to gain a complete understanding of this phenomenon.
Is there a link between religiosity and OCD symptomatology?
Whilst the question of whether religiosity causes or predisposes to OCD is still unresolved, there does seem to be considerable evidence in favour of a more subtle effect of religiosity on OCD. It has been reported that religious patients are more likely to complain of religious symptoms of OCD. This implies that religiosity may have an impact on OCD via symptomatology.
Evidence proposing a link
This was one of the main findings of the work of Steketee et al. (1991) who noted that the more religious a patient, the more likely he or she was to complain of religious obsessions. These findings are echoed by Mahgoub and Abdel-Hafeiz (1991) who found that by far the commonest form of obsession (50% of all subjects) involved religious prayers and/or ritual ablution and that compulsions were most likely to take the form of religious repeating (50% of all subjects) in their subject group of Saudi Arabian Muslims. Similar results have also been obtained in studies looking at the Jewish population. Greenberg and Witztum (1994) compared symptomatology in ultra-orthodox versus non ultra-orthodox outpatients with OCD. 13/19 ultra-orthodox patients were found to have religious symptoms compared with 1/15 in the non-ultra-orthodox group. Likewise, Greenberg and Shefler (2002) found that 26/28 ultra-orthodox OCD patients had religious symptoms and that, on average, patients reported three times as many religious as non-religious symptoms.
Therefore it seems there is a strong argument in favour of religion influencing OCD symptomatology. Whilst the studies cited above are still subject to the limitations mentioned previously, the use of clinical subjects and the similar results obtained are encouraging signs. These findings are in keeping with current opinion on OCD that whilst the broad categories of OCD symptoms, namely obsessions and compulsions, appear consistently around the world (Silva and Bhugra, 2007), the nature of obsessions and compulsions vary widely between different cultural groups. A possible explanation for this, put forth by Silva and Bhugra (2007), is that obsessions and compulsions reflect the concerns of the individual which is are influenced by the collective concerns of that individual’s cultural group. For example, thirty years ago obsessions in the UK commonly related to a fear of asbestos exposure, whereas nowadays they are more likely to involve fears about contracting HIV (Silva and Bhugra, 2007). Since religion forms a considerable part of culture, it has been postulated that religious beliefs too could also have an impact on OCD by influencing symptomatology in keeping with the findings cited above.
Implications for clinical practice
Whilst the claim that religion predisposes to OCD remains unsubstantiated, it appears that religion does have some influence on OCD in terms of symptomatology and therefore its presentation. Whilst this information is useful in terms of adding to general understanding of OCD, it is important to establish how these findings might help improve clinical practice and management of OCD patients.
Help-seeking in religious patients
The findings that religious people may develop more religious symptoms of OCD and that OCD may be more severe in patients who are highly religious (Steketee et al., 1991) raises interesting questions with regards to help-seeking in this group. It may be that religious patients are more likely to go to religious leaders before presenting to a GP/other medical practitioner thus giving the impression of worsening severity. If this is the case then healthcare providers might like to address any barriers to care that can be identified. Patients may, for example, feel that staffs do not understand their religion or the symptoms they are experiencing. To deal with this, practices might consider courses to educate medical staff or GPs may consider contacting religious leaders, with patient’s consent, in such cases in order to improve his/her understanding of the case (as suggested in section 10.1.4 of the 2007 NICE guidelines for OCD). Patients may also be unaware of the services that are available to them, especially if they live in close-knit communities (e.g. ultra-orthodox Jews). Practices could therefore try to engage religious leaders and provide them information about the services available so that they can then raise awareness within their communities. Bonchek and Greenberg (2009) also highlight the importance of involving rabbis and other religious leaders in treatment if possible, as they can both advise on what is and is not appropriate in religious terms for patients and could help with concordance by effectively giving a patient permission to treat their problem.
On a similar note, it may also be important to consider the best treatment option for religious patients with OCD. Cognitive Behavioural Therapy (CBT) has been shown to be effective in the treatment of OCD (Bonchek and Greenberg, 2009) but it may be important to consider the appropriateness and usefulness of current techniques in this patient group. For example, it is clear that instructing a religious patient to resist the urge to pray entirely is likely to be met with resistance. In addition, it is also harder to monitor the progress of symptoms involving prayer given its personal nature (Bonchek and Greenberg, 2009). A variant of CBT known as “guided prayer repetition” has been proposed by Bonchek and Greenberg (2009) which, it is claimed, is a more acceptable and successful form of therapy for religious patients. It will be interesting to determine its use and effectiveness on a wider scale.
In summary, the findings reported above have a number of important implications for clinical practice. Firstly, it will be crucial to identify barriers to healthcare for religious patients and to deal with these accordingly. Secondly, it will be important to tailor treatment to the religious patient in order to achieve a favourable outcome, including, if appropriate, involving religious leaders in care. Lastly, there should be a move to increase education and awareness of this problem amongst medical practitioners in order to improve patient management.
To conclude, research carried out thus far can neither confirm nor deny the theory that religious devotion predisposes to mental illness. It seems likely that religiosity does not predispose to OCD but may lead to an increased propensity to develop obsessive-compulsive traits. Further work is however required to elaborate on these findings and their implications. Regardless, there is a growing body of evidence to support the proposition that increased religiosity may lead to religious-type OCD symptomatology. It will be important to consider this finding both in clinical practice and in the consideration of the new 2011 NICE guidelines for OCD in order to enable better diagnosis, treatment and outcome of this condition.
ASSARIAN, F., BIQAM, H. & ASQARNEJAD, A. 2006. An epidemiological study of obsessive-compulsive disorder among high school students and its relationship with religious attitudes. Arch Iran Med, 9, 104-7.
ISLAM, F. E. 2010. How to perform Wudu [Online]. Available: http://www.faizaneislam.org/islam/how_to_perform_wudu.shtml [Accessed 20th November 2010].
THE BRITISH PSYCHOLOGICAL SOCIETY AND THE ROYAL COLLEGE OF PSYCHIATRISTS 2007. Obsessive-Compulsive Disorder: core interventions in the treatment of obsessive-compulsive disorder and body dysmorphic disorder. . National Institute for Health and Clinical Excellence (NICE) Guidelines. .