Religion and OCD, young Muslim women, Saudi

Fifteen young women suffering from OCD in Saudi Arabia were interviewed about their experience of the illness and the roles played by religion. Religion was not perceived as a cause of the illness, but the illness can show itself in religious symptoms – notably with respect to prayer, and in a phase in which the young women were very strict and literal-minded with themselves and their families, with respect to religious observance. Religious symptoms were reported as more upsetting than other symptoms – being seen as damaging to the sufferer’s piety. Other facets of the importance of religion in the experience of OCD were shown in help-seeking, in choosing to go first to religious healers for treatment, and only when these were unsuccessful were mental health professionals consulted. At this point, it was very important that the professional should be seen as trustworthy, and the criterion of trustworthiness was religiosity, specifically the use of pious, Qura’anic quotations, and a covered face (for a woman)/long beard (for a man). The accounts of the roles of religion in the experiences of the young women interviewed suggest that religion is an arena – but not the sole arena – for the expression of OCD symptoms. Religion plays an important role in determining the acceptability of treatments and treatment providers.

Introduction

Religion and obsessive-compulsive disorder (OCD) have been linked together in the psychological literature for a long time, yet the relationship between the two is still far from being understood. In his work on obsessive actions and religious practices, Freud (1961/ 1907) highlighted the similarities between the two, particularly in the rituality of the behaviours involved, as well as the guilt involved when those behaviours are abandoned or not carried out correctly. However, Freud also noted the clear differences between the two where religious practices are meaningful in every detail and are carried out in accordance with religious norms in contrast to neurotic symptoms which are pointless even to the person himself let alone to other people. The interest in this relationship has grown and some interesting studies have gradually clarified the relationship between religion and OCD. However, this area of research remains complicated by several factors. These include: (1) The nature of the religion studied, whether it puts more emphasis on personal versus

communal worship and whether it commands intrinsic versus extrinsic expression of religiosity. (2) Related to the first point is the measurement of religiosity. Developing valid and reliable measurements of religiosity that are culturally sensitive can be an extremely difficult task. (3) There are varied aspects of religious activities involving thoughts, beliefs, and feelings, which should be taken into account when assessing religious behaviour. (4) Creating a unified definition for the constructs in question, hence judging both religiosity as well as obsessionality can be quite subjective, as Yossifova and Loewenthal (1999) have shown in a study where people described as religiously active were judged as being more obsessional than those not described as religiously active. Therefore, drawing valid, reliable conclusions can be difficult, as is the comparison between different studies. Lewis (1998) reviewed a series of studies examining religion and obsessionality. He pointed out other problems which also apply to most of the research examining the relationship between religiosity and mental health: (a) Most of the studies deal with correlations, and therefore do not allow us to infer causal relationships. (b) The correlation patterns between measures of religiosity and obsessionality are inconsistent and sometimes contradictory. (c) Most of the studies were on university students for whom religion did not always seem to have special salience. Nevertheless, findings suggest that a high level of religiosity is associated with obsessionality. Maltby (1999) examined the relationship between regular church attendance and obsessional symptoms among 574 adults form the USA, Northern Ireland, and England. The findings suggested that individuals who attended church at least once a week reported more obsessional symptoms than those who attended church less frequently. The author explains the failure of finding such an association in previous studies (Lewis & Joseph, 1994; Lewis & Maltby, 1995; Maltby, 1995) by the fact that the effect of the relationship between regular church attendance and obsessional symptoms can only be detected in large samples and that the majority of the studies which failed to detect this effect used a sample of less than 150 respondents. Therefore, the jury is still out with regards to the empirical support of Freud’s views that religion and OCD are intertwined and that religiosity is a socially acceptable form of obsessionality. Yet Freud also noted that the difference between the two is also very clear since religious activities serve important goals both to the individual and society, while OCD rituals are considered absurd and aimless even by the sufferers. Thus, one could conclude that this contradiction reflects a balance between the similarities and the differences between the two constructs as noted by Freud himself (Lewis, 1998). Lewis (2003) reviewed the studies exploring the relationship between religious attitudes, practices, orientation, and obsessional traits and symptoms. Conclusions were similar to the previous review (Lewis, 1998) since findings suggested a relationship between religiosity and obsessionality as a personality trait. This has an important implication, since it increases the possibility of generalising this finding across different measurements which have been employed by the nine studies identified by the author. However, meta-analytical procedures were suggested as a better alternative to the vote-counting method employed by the author; since a majority of the studies exploring the relationship between religiosity and obsessionality involve a small sample size usually less than 150 participants which prohibits detecting small effects by using vote-counting method alone. Sica, Novara, and Savanio (2002) asked a group of Italians who were identified as either high (Catholic nuns or friars), medium (a typical person who attends church and is well rounded on religious matters), or low on religiosity (people not interested at all in religious practices). Results have shown that high and medium religiosity was associated with higher obsessionality, obsessive-compulsive cognitions, and perfectionism. A further shortcoming of the studies reviewed above is that they are mainly carried out on individuals from the

Christian faith; therefore, they do not necessarily further our understanding on the relationship between religiosity and obsessionality in individuals from different religious groups. Studies carried out in non-Western societies give us some insight on the phenomenology of OCD and how it relates to religion. Greenberg and Witztum (1994) have studied this relationship among ultra-orthodox Jews in Jerusalem. They concluded that religion provides the setting for the expression of OCD symptoms rather than being a distinctive theme or cause of OCD. In another study carried out by Greenberg and Shefler (2002), 28 psychiatric referrals between the ages of 17 and 43 years old diagnosed with OCD were interviewed. Twenty-six of them had religious symptoms, 18 of them also had non-religious symptoms, but religious symptoms were three times more prevalent than non-religious symptoms. The majority of interviewees viewed their religious symptoms as most disturbing. Only nine of those interviewed thought their non-religious symptoms were their main complaint. It is worth noting that all nine were ultra-orthodox from birth. Religious and non-religious symptoms did not differ in terms of the level of distress, resistance, sense of irrationality, and hours spent daily, and in all cases the sufferer was more likely to turn to a religious healer for religious symptoms and to a mental health professional (MHP) for a non-religious symptom. The findings of this study are limited by two factors, the sample size and the selection bias; all those interviewed had already sought professional help from mental health workers and that of itself gives an indication of their views on the nature of the disorder. Also in Israel, Zohar, Goldman, Calamary, and Mashiah (2005) conducted two studies exploring the relationship between religiosity and obsessional symptoms. The study is novel as it looks at individuals who had experienced a religious change that is from observance to non-observance or the other way round. In the first study, 256 undergraduate volunteers between the ages of 18 and 32 completed four five questionnaires: (a) the Maudsley Obsessive-Compulsive Inventory (MOCI); (b) the obsessional thought checklist, (c) the child adolescent perfectionism scale; (d) the student religiosity questionnaire. In the second study, there were 61 participants between the ages of 18 and 59, 30 of them had been more religious and relaxed their religious observance, while the other 31 had become more religious. In both studies, participants described their religious status as being (a) secular, (b) traditional, (c) orthodox, and ultra-orthodox. Findings indicated that there was no relationship between religiosity as measured by the student religiosity questionnaire and OC behaviour as measured by MOCI. Perfectionism correlated significantly with OC behaviour, and only a weak correlation was found between perfectionism and religiosity. The group described as more religious scored higher on the MOCI than those who underwent a change from being more religious to less religious. The authors speculate that OC tendencies were characteristic of the individual and possibly played a role in the process which brought about the change in their level of religious commitment (i.e., from being less religious to being more religious). Studies carried out in different Muslim countries which have yielded some interesting results and indicate certain patterns which are worth our attention. A phenomenological study published in 1994 by Okasha, Saad, Khalil, El Dawla, and Yehia, reported on 90 patients diagnosed with OCD using the International Classification of Diseases 10th edition [ICD-10] and assessed by the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) for classification and severity of symptoms. Sixty-nine percent of the patients were males, and 32% were females. The mean age of the sample was 23.7 years. Forty percent of the patients presented with a mixture of obsessions and compulsions, whereas 29% presented solely with obsessions and 31% solely with compulsions. Sixty percent of the obsessional thoughts were related to areas of religion, making religious compulsions the most common

compulsive behaviour. Repeating rituals and cleaning were the most common among the compulsions. Muslim patients were more likely to have symptoms related to the last two areas, while those from the Christian faith were more likely to have symptoms related to orderliness. This was a particularly interesting, since it is one of the few studies which compared Muslims with those of other faiths living in the same culture, and hence the findings give us some indications that religion can have a distinctive effect in shaping the manifestation of OCD symptoms, which may not necessarily be the same as the cultural factor. Another study was carried out in a different Muslim country also showing that religious symptoms were among the distinctive features of the phenomenology of the disorder. Mahgoub and Abdul-Hafez (1991) interviewed 32 OCD sufferers from the eastern province in Saudi Arabia and found that religious symptoms dominated both the obsessions as well as the compulsions, with doubts being the most common feature. However, the sample is biased since interviewees were recruited from the eastern province where Muslims from the Shia’ sect, who constitute only 5% of the Saudi population, are over represented. Another study was carried out in Bahrain by Shooka, Al-Haddad, and Raees (1998) and showed that religious symptoms were experienced by the majority of the 50 patients investigated in this study. Subjects aged 18–65 years were interviewed using the Arabic version of the Preset State Examination and were screened by the Arabic version of the Maudsley Obsessive-Compulsive Inventory and the Global Obsessive-Compulsive Scale. Forty percent showed religious and blasphemous obsessional thoughts and doubts. Had further qualitative data been presented, one would have gained a better insight into the thoughts as such and hence been able to compare them to religious obsessions experienced by sufferers from different religious and cultural backgrounds. Ghassemzadeh et al. (2002) examined 138 Muslim OCD sufferers in Iran. Self-rating instruments were translated to Farsi and used for the purpose of the study; those were the Maudsley Obsessional-Compulsive Inventory (MOCI), Compulsive Activity Checklist (CAC), and the Beck Depression Inventory (BDI). After analysing the data, the researchers concluded that the overall pattern of symptoms was similar to the patterns reported in Western settings. However, religious symptoms were identified, but unlike the previous studies, they were not viewed as a distinctive feature of OCD among Iranians. However, gender differences were noted, as obsessions with self-impurity were more common among females, while blasphemous thoughts were more common among men. The researchers argue that the variations in OCD symptoms across cultures were more apparent than real, and that the findings of this study support this view. Yet, one might argue that this study while valuable in many different aspects has not succeeded in capturing the distinctive features of the sufferers’ experiences and that this was mainly due to: (a) The use of instruments developed to be used in a Western society, and that may not be sensitive enough to pick on the specific symptoms which are experienced by those of different culture. (b) Although some of the instruments used touch on religious symptoms, none of them elaborate enough on this dimension of the OCD experience. (c) The researchers depended completely on the data emerging from the self-rating instruments mentioned above. Had they complemented this with qualitative methods, they could have developed a more detailed picture that may lead to different conclusions. In general, studies indicate that OCD symptoms of religious nature are common in different cultures. However, a majority of the studies are correlational and do not attempt to explain this phenomenon, or capture the role of religion in shaping the presentation of OCD. The study to be described here was carried out in Saudi Arabia, where there are significant gender-related hurdles in research. Saudi society is a very conservative society,

particularly in the central part of the country where this study was carried out. There are extremely strict rules governing the way people, and particularly women, behave in public. These rules include their dress code; they have to wear a black cloth covering all their body including the face. Women considered liberal cover their body and hair, but not their face. Women consider liberal remain a minority and are frowned upon in most public places. In hospitals, however, many professional women, physicians, and otherwise tend not to cover their faces: it may be more acceptable for women working in the medical field not to cover their face. In all interviews, the first author (LAS) covered body, hair, wearing the traditional Saudi costume (Abaya), but kept face unveiled. This would have affected the way the interviewer was perceived, and the way questions were answered. Effects could not be quantified, but interviewees were comfortable, if not appreciative, for having someone who is interested in their experiences. In a pilot interviewer with a male, however, the interviewee was uncomfortable, and it was decided not to include males in this study. Also, some sufferers who were interested in taking part in the study were unable to do so because an accompanying male family member was not agreeable either because he did not see the point of taking part in the study or simply because it was not possible to wait the extra time due to other commitments. This study aims to explore the role of religion in the experience of OCD among young, female, Saudi sufferers attending psychiatric clinics in Saudi Arabia. Semi-structured interviews covered the development of religious symptoms; their effect on the person’s functioning, and the steps taken to overcome the disorder. It should be born in mind that participants in this study were therefore solely female.

Method

Participants were: (1) Female between the age of 14 and 30 years old. (2) Onset of OCD before the age of 21 years old. (3) Confirmed diagnosis of OCD with symptoms in the religious domain; no cases without religious symptoms were present. Confirmation of diagnosis was obtained in two ways: (a) thorough check of prospective participants’ medical records, (b) brief verbal report from mental health professional involved in the treatment of prospective participants. (4) All interviewees had to be Muslims, Arabic speaking, from Saudi Arabia. (5) Fifteen female sufferers took part in the study.

Data collection Since the interviews were semi-structured, it was difficult to estimate how long each interview would take; 60–90 minutes were aimed for. In addition to basic demographic data, the interview schedule consisted of open-ended questions which covered the following areas: (a) history of presenting problem(s); (b) subjective account of religious symptoms; (c) religious coping; (d) religious development; and (e) the role of religion in the development, maintenance, and treatment of presenting symptoms. Video and audio recording were not considered appropriate in a conservative society such as the Saudi, let alone in a psychiatric setting which is treated in the highest level of secrecy. For this reason, only written records were made of interviews.

Data analysis Thematic analysis, an established form of Grounded Theory, was used to analyse the data emerging from the semi-structured interviews (Willig, 2001). After completing each interview, translated from Arabic to English, answers were read, reread, and theoretical codes emerged from the data. At first, codes were given to low-level categories but as cyclical process of reading the interview progressed, linkage between low-level categories was established, and they were integrated into higher-level categories using a coding paradigm. Finally, linkage was established between high-level categories in all interviews and overarching themes emerged. In order to check the reliability of data analysis, 20 quotes were extracted from the interviews and presented in a random way to a judge (KL) along with six themes which emerged from the data. The judge was able to match quotes with corresponding themes except in two cases where she gave more than a single possible answer. In both cases, quotes corresponded to the same theme and the formulation of this particular theme was revised and clarified.

Results The following themes emerged.

Help-seeking behaviour All interviewees who took part in this study said that their first attempt to seek outside help was going to a faith-based healer. In all cases, interviewees took this step willingly and did think of faith-based healers as an appropriate outlet for help. The main methods used by faith-based healers were (a) reading verses from the Qura’an, (b) giving the sufferer holy water to drink (zamzam), (c) giving the sufferer blessed oil to spread all over their body. In most cases all three methods were used. R., 23 years old, explained:

‘‘When my symptoms started two or three years ago, going to a female faith healer was the first thing I did; she told me those were obsessions and what you need is zamzam, and blessed oil; she also read verses from the Qura’an on me; I used to feel very comfortable when she read the Qura’an.’’ Only in one case, D., 24 years old, was a different method used:

‘‘The faith healer I went to, who I know now is a shaman, asked me to get a goose, with specific features, slaughter it, and bathe in its blood. I can’t believe I did that, it was torture.’’ The interviewee, however, acknowledged that the person she went to was more of a shaman than a faith-based healer. She later went to a faith-based healer who used the traditional methods mentioned by other interviewees. And like other interviewees, such methods provided a temporary relief rather than a lasting alleviation of the obsessional symptoms. The most common explanation given by faith healers on what caused the obsessional behaviour was that it was caused by Satan, and that it was triggered by an evil eye. Only in one case, M., 22 years old, a different explanation was given:

‘‘I went to a faith healer with my parents, he read verses from the Qura’an and said it was evil eye that caused the symptoms, but recently he said that there is wind in my head that is causing some noise.’’ Based on that, the interviewee was suspected to have some psychotic symptoms, but this was not confirmed on a closer examination.

In general, for interviewees, psychiatric service was perceived as the last resort: (1) An average of six months to one year between the patient’s first seeking of a faith-based healer to the time when she reaches an MHP. (2) Patients went to the MHP after acknowledging that benefits of faith healing were limited in terms of lifting OCD symptoms. N., a 15-year-old interviewee, described her journey to seek help for her obsessional symptoms:

‘‘The obsessions started last year, I started having thoughts about my faith, about the prophet, about God’s creation. Then I started having doubts about making mistakes in my prayers or when I read verses of the Qura’an. I started repeating my prayers so many times that I had to miss school on many occasions. My parents decided to take me to a faith healer who started reading verses of the Qura’an to me, this gave me a temporary relief, but the symptoms persisted. This is my third visit to the doctor and I already feel better.’’ L., 17 years old, was also taken to a faith healer when her symptoms started:

‘‘A year ago I started taking long time in performing wodoo, when I told my mother she took me to a motawaa (faith healer), he read verses of the Qura’an on me, I was relieved. He told me don’t repeat wodoo, you will be listening to Satan if you repeat your wodoo. After a while my father brought me to the hospital. He gave me some pills, symptoms are much better but I don’t know what those pills are doing to me. I stopped going to the motawaa four months ago.’’ That all interviewees went to faith-based healers when deciding to seek help for their obsessional symptoms might be explained by this being the initial and spontaneous reaction of Saudi families. Interviewees came from different social and economical backgrounds; yet when faced with their daughters’ problem, they all responded in the same way; this is seen as an indication of the deeply ingrained belief in traditional methods of healing. It is important to know that as young women, the interviewees do not have the power to seek help themselves, thus the decision of how to seek help must involve other members of the family, leading to the question of whether the interviewees would have taken the same steps towards treatment had they been able to make the decisions, and act on them independently.

Perception of causality of obsessional symptoms Evil eye is caused by a person, not necessarily ill-intentioned, admiring something one possesses. This can be a materialistic thing or a personal attribute, and is caused by not citing the name of God in the moment of admiration which would have prevented the evil eye. This was the first and most influential explanation of the cause of symptoms. After realising that symptoms were caused by a psychiatric disorder which has well-known biological causes, patients accepted the biological aspect of the illness but still believed that evil eye had triggered the biological imbalance. So the two explanations are not mutually exclusive, and thus most patients continued to go to the faith healer as well as the MHP. This was the case with L., 17 years old:

‘‘My family understands, they know that [my] obsessionality is caused by Satan, but they still believe I should take the pill, and also, read verses of the Qura’an on myself.’’ So the explanation is shared by the sufferer and her family. G., 21 years old, gave a simple yet firm understanding what caused her OCD:

‘‘I am convinced that what I am going through is a psychiatric disorder, but any psychiatric disorder is caused by evil eye.’’

Obsessional symptoms were explained the same way any negative event or problem was explained, by evil eye. This explanation can have a number of implications: (a) obsessional symptoms were not particularly stigmatised; (b) those struck by evil eye are usually perceived as having positive attributes attracting envy form others; interviewees certainly conveyed this belief, and drew esteem form that; (c) interviewees sometimes felt responsible for attracting evil eye which lead to the development of their symptoms. L., 17, makes this point clear:

‘‘I have very beautiful hair. My mom always told me to put it up so I wouldn’t attract [an] evil eye, but once I put it down at a wedding party and perhaps I was struck by evil eye, and developed those obsessions.’’ Beliefs about causality certainly played an important role in interviewees’ understanding and dealing with their symptoms. Traditional beliefs about causality remained powerful in spite of the success of modern medicine. Interviewees did not dispute the biological mechanisms underlying their symptoms but rather believed that those mechanisms were affected by metaphysical power.

The importance of religiosity and religious knowledge in mental health professionals The level of a professional’s religiosity was determined mainly by they way s/he appears: having a long beard in the case of a male professional and covering the face in the case of a female professional. This first impression is emphasised by the way the professional talks. Professionals, who use a religious rhetoric, citing verses from the Qura’an or teachings of the prophet, were perceived as more religious than those who used non-religious rhetoric. The MHPs’ level of religiosity seemed to play a role in shaping the therapeutic relationship at many different levels: (a) A religious professional would be more conscientious and thus would neither harm nor manipulate their patients (general trust in religious people). As M., 27 years old puts it:

‘‘I don’t really care about the level of religiosity of the psychiatrist who is treating me; he is not a family member or a friend so why should I feel protective of him. But it would be better if he was religious, since he will have fear of God, and thus be more dedicated in doing his job.’’ G., 21 years old, gave a further a more idealistic view of religious professionals:

‘‘I feel that the level of the professional’s religiosity correlates with how much I trust her/him. I have more admiration if I think the person who is treating me is religious, I listen to her/him, I accept what they say, I become much more receptive of what they say than if she/he was not religious.’’

(b) A Muslim MHP would be in a better position to understand religious symptoms the patient is going through, and thus patients are more willing to discuss those symptoms with a Muslim MHP. This was the opinion of C., a 17 years old interviewee:

‘‘I certainly prefer a Muslim psychiatrist, a non-Muslim would not understand what I am going through, would not understand my problem.’’

(c) A religious MHP would have a better knowledge of the religious view of certain behaviours related to the symptoms (fatwa), and thus he/she will be in a better position to give advice. H., 17 years old, said:

‘‘I’ve never thought about the level of religiosity of mental health professionals, but I don’t think I would still be with my psychiatrist had she not been a religious woman. It assures me that her opinions are correct because she is knowledgeable about how Islam views my illness and what I should or should not do. I don’t think I would have trusted her had she been not religious.’’ In one personal observation during this study, when attending sessions with a MPH who appeared religious, patients asked him to read verses from the Qura’an and pray for them to get better. They related to him the way they related to a religious figure believing that he had the power to heal them not only by prescribing medicine but also by extending his blessings.

Symptoms in the religious domain are more disturbing than in other domains Religion is seen as the most important element of the patient’s life, and failure to carry out religious rituals in the proper manner is a punishable sin. A., a 14-year-old interviewee, compared her compulsive tendency to repeating prayers and homework:

‘‘I am more disturbed by the fact that I repeat my prayers than the fact that I check my homework. I hear all those stories about people who don’t pray well or don’t do ablution going to hell and it scares me. Those thoughts always occur to me when I am doing wodoo, it disturbs me psychologically.’’ Disruption in performing religious duties was also the major cause of anxiety for R., 23 years old:

‘‘I used to have symptoms in my prayers, I repeat them, shout during my prayer, and repeat wodoo; my family told me this was the work of Satan. I feared my prayer have not been accepted, it caused me anxiety that destroyed my life. I still have other symptoms now but I can manage.’’ F., 17 years old, made this point very clear:

‘‘Of course religious symptoms are more distressing, I feel more responsible, and have more fears when I am performing a religious duty. I have symptoms in the area of cleanliness in general, but when I am about to perform my prayer, I check the cleanliness of my clothes unlike any other time.’’ For some interviewees, religious symptoms seemed to be the reason for actively seeking help. This was the case with H., 17 years old:

‘‘All symptoms were distressing, but religious symptoms were the ones that made me come to the psychiatrist.’’ There was a single exception, where the interviewee thought that symptoms not related to religion were more disturbing. J., 21 years old, expressed a more forgiving view of God:

‘‘Non-religious symptoms, thoughts relating to cleanliness, my relationship with other people make me more anxious since the case of religious symptoms I can declare my sins to God, and express my remorse, and God might forgive me, in other areas this is not possible.’’ This interviewee seemed to hold an image of God that was more forgiving than punitive compared to the rest of the interviewees. Performing religious rituals is considered highly important in the interviewees lives. The disruption caused by their symptoms did not only trigger high levels of guilt but also was an indication that the sufferer was not functioning well.

The most prevalent symptoms were related to daily prayers The five daily prayers, and their preceding ablution rituals (wodoo), followed by repeating the mandatory prayers five times a day. All interviewees thought that these symptoms were most distressing, time-consuming, and physically exhausting. As a result, they were perceived as most interfering with normal life activities and the main reason for seeking treatment. In rare cases, interviewees said that they had suffered from obsessional rumination. Thoughts about God and the beginning of creation were relatively rare. M., 22 years old, is one of three interviewees who experienced symptoms as such:

‘‘I get those thoughts about who created God, how is God able to create everything. I know those thoughts are from Satan, so I say to myself, my God is Allah, I do not have a God but him.’’ The rarity of covert obsessional doubting was particularly interesting. There are two possible reasons for that:

(1) Interviewees may have experienced doubts about the creation of God, the existence of God, etc. but since thoughts as such were perceived as extremely threatening and provoked high levels of anxiety, interviewees were not able to express them during the interview, especially since it was their first encounter with the interviewer. S., 24 years old, gave a hint but was unable to elaborate on her thoughts:

‘‘Questions that dominate my mind, there is no way of answering them, I feel guilty for having such questions.’’ (2) The other possible explanation is that since overt expressions of worship is a central aspect of the Islamic religion, and since this aspect is particularly emphasised in the religious socialisation in Saudi society, members of society become more concerned about the ritualistic aspect rather than the introspective aspect of religion in the way they construct their relationship with God. If we assume that those who suffer from obsessive-compulsive symptoms are exhibiting extreme forms of behavioural patterns common in their society, this might explain why thoughts regarding the existence of God were not the one of the main causes of symptoms described in the interviews. G., 21 years old, said:

‘‘I don’t think that a person is responsible for his thoughts, behaviuors are the most important.’’

Symptoms in the religious domain were mostly overt, relating to rituals most frequent in the Muslim’s daily life. Obsessional ruminations with religious content were relatively rare.

Role of religion in coping with symptoms Religion is a main source of coping with OCD symptoms; personal prayer is used increasingly even after patients stop going to the faith healer. Also, prayer is one of the main sources of coping with general stress; family and social network were among the other sources. L., 17 years old, said:

‘‘I really feel sad when bad things happen, but I sit, listen to the Qura’an, and feel better. My family is the same. Religion helps you adapt, prayer distances Satan from me and the more I pray the more God protects me.’’

G., 21 years old also acknowledged the protective role of religion:

‘‘Had I not been a Muslim I would have gone! Religion is the foundation; it even alleviates the distress of illness, since God will commend our endurance. Those who have no religion commit suicide. The Muslim knows that fate is from God. One has to look at the stories of the prophets and what they have been through, it helps.’’ T., 30 years old, found comfort in performing religious rituals:

‘‘Whenever I have a problem, I pray and fast.’’ Religion not only helped interviewees cope with their illness, but also with difficulties in all aspects in their lives. Interviewees turned to religion to make sense of their illness, and for inspiration to accept it and cope with it.

Religion and self-esteem Interviews revealed an association between the patient’s perception of their level of religiosity and their self-esteem: (a) Committing acts that are perceived as religiously unacceptable triggered feelings of guilt and a sense of failure, which had a negative effect on the patient’s self-esteem; (b) The more the patient perceived herself as being compliant with the teachings of Islam, the more proud she became of herself. R., a 23-year-old interviewee expressed her desire to be a preacher; three other interviewees shared this desire with R.:

‘‘I am quite religious, but I want to be even more religious, I want to strictly adhere to the teachings of Islam and become a preacher. I want to stop listening to music and all things which are sinful.’’ Although this might appear to be extreme for some people, this is acceptable in the general society as an indication of an admirable level of religiosity. While H, 17 years old, said with a shy smile:

‘‘I consider myself religious compared to girls around me, I pray, I fast, don’t listen to music, I cover my face, but I have to do more, sometimes I feel good about myself at other times I don’t but when I adhere to the Islamic code I feel a greater sense of satisfaction.’’ Interviewees were aware of the distinction between the obsessional nature of their symptoms and the acceptable adherence to Islamic teachings. The later was viewed in a positive way and being a good Muslim was something interviewees aspired to attain.

Going through a period of religious extremism while experiencing OCD Most interviewees described a rigid understanding of the teachings of Islam, which affected their relationship with friends and family at an early stage of their illness. Not only did interviewees expect themselves to be perfect Muslims, but also expected the same of their friends and especially members of their family. They felt that they had a responsibility towards those close to them, to show them the righteous way and stop them from committing sins in every way they can. Because of the strict interpretation of the teaching of Islam so many common behaviours were perceived by sufferers as sinful. This led to the development of social strains which added to the distress experienced by the sufferer. M., 22 years old, talked about her experience in this respect:

‘‘I feel so frustrated when I see my sisters not covering their faces properly, I suffer psychologically, I advise them all the time. During the final holy nights of Ramadan, I found them

watching television, I shouted Allah Akbar (God the greatest), they did not respond. I broke the television. Recently, I stopped advising them, I feel better psychologically.’’ The rigid understanding of religious teachings and the extreme measures taken to apply them was seen as one of the early signs of OCD. To them religion was a set of fixed rules which had to be applied precisely and failing to do so will lead to dreadful consequences. Their personal responsibility for following these rules extended beyond their own behaviour to that of people around them.

Discussion

The qualitative data collected from OCD sufferers give us an insight into the experience of illness in a religious, conservative society like Saudi Arabia. Interviewees believed their symptoms were caused by evil eye; this can have a number of implications on the way people understood and dealt with their illness: (a) the evil eye explanation implies that the sufferer was not responsible, and thus is unable to control the symptoms; (b) this explanation implies that the person had some positive attributes which attracted the evil eye, this means that sufferers did not view themselves as bad people and thus the illness was not viewed as a punishment for being sinful; (c) as a result the person should feel less stigmatised. Sufferers’ understanding of their illness shaped their help-seeking behaviour; they turned to faith-based healers for help to alleviate their symptoms. This is consistent with the trend observed in neighbouring Kuwait by Darwish (2004) who found that people with mental health illness visited faith-based healers, and 35% of them preferred visiting faith-based healers than MHPs. The question that needs to be asked, is whether people are more likely to visit faith-based healers if their symptoms were of a religious nature, or do people turn to faith-based healers regardless of their symptoms? One can speculate that people turn to faith-based healers regardless of the nature of their symptoms, Darwish did not specify religion-related symptoms in his study, but in the case of religion-related symptoms, it might take suffers more time, if ever, to recognise their symptoms as a mental illness which needs psychiatric attention. Similar to the findings of Greenberg and Witztum (2001), this sample thought that religious knowledge was important for MHPs. But contrary to the Jewish study, this sample found that their religion-related symptoms were most disturbing. In fact, a number of them cited their religion-related symptoms as the main reason for them to seek medical help. This contrasts with the ultra-orthodox Jews who were less bothered by their religion-related symptoms compared to OCD symptoms affecting other areas of their lives. This could be due to the fact that the sample in the Jewish study was identified as highly religious and for people who are highly religious more can mean better. But it is impossible to compare the two samples because the level of religiosity of participants was not determined in this study. Interviewees found religion-related symptoms most disturbing since they disrupt their relationship with God, which was difficult to tolerate. In addition, one would speculate than in a religious society like the Saudi, where a lot of activities evolve around religious rituals, suffers’ lives must have been severely disrupted by their illness not only because they were not able to fulfil God’s demands, but also, because they were failing to fulfil their function in their society. The experience of dramatic change in the sufferers’ level of religiosity at the beginning of the illness is quite significant. It is consistent with the findings of

Zohar et al. (2005): who found that individuals who changed from being less religious to being more religious scored highly on measurements of OCD. Further studies are needed to understand this change better, whether it was sudden or gradual, or was it brought about by certain life experiences, etc. It is important to note that we might have a very different picture had males been included in the study. First, in Saudi Arabia males lead more independent lives compared to females, thus they are more able to make decisions and act on them independently. This can have a profound effect on the sufferer’s experience, since he can choose where to go for help without necessarily consulting with other family members. Second, the five daily prayers are a communal ritual for men who go to the mosque for prayer, which must have an effect on the experience of symptoms related to daily prayers. Although it is not possible to generalise the findings from this study because of the small sample size, gender bias in recruiting participants, and the because of the qualitative nature of the data collected; the findings, nevertheless, generate interesting hypotheses which can be tested in further investigations. Finally, there is a selection bias; participants who took part in the study were all recruited from mental health clinics; this indicates that they had some trust in the medical establishment even when using traditional methods. Those who do not seek medical help at all might have views different than those presented in this study. Since all interviewees sought the help of faith-based healer in the beginning of their illness for a period of time before arriving at the psychiatric clinic, one would speculate that there might be a large number of sufferers who never make it to the psychiatric clinic. There might be a number of reasons for that, (a) either sufferers experience some benefit from visiting faith-based healers and thus do not feel the need to search for other sources for help; (b) sufferers might not be aware that their condition is a psychiatric disorder, and that there is a specialised service provided for such condition; (c) sufferers might be aware of the benefits they might get from psychiatric services, but are discouraged from seeing the psychiatrist by either their own misconceptions about psychiatry or the misconceptions of their family members and are unable to act independently. Among the young women who were interviewed, we can see the importance of religious factors in determining the acceptability of treatments and treatment providers. We can also see that religion is not experienced as a causal factor in OCD, but religion is seen as an arena for the expression of OCD, much to the discomfort of the sufferers.

Acknowledgement This article is based on part of a thesis awarded a PhD to the first author, from London University in 2005. The second author acted as supervisor.