The prevalence of obsessive-compulsive disorder (OCD) has been recently estimated at 2% in the general population [1–3]. Traditionally OCD was believed to be a much more uncommon disorder. It has been hypothesised that patients were reluctant to disclose OCD symptoms in medical consultations and deliberately hid them from the clinicians [4]. This meant that only the most severe cases got to be evaluated by a specialist. The high comorbidity of OCD with other psychiatric disorders (mainly anxiety and depression) has also contributed to the underdiagnosis of OCD.

OCD may not be an extremely prevalent disorder [5], but it has high clinical relevance. Its importance lies in its deep impact on the patient's life and daily activities. This is particularly relevant in OCD because this disorder mainly affects young adults, who have a potentially high level of activity in all the spheres of life. The above mentioned high comorbidity of OCD with other psychiatric disorders [6, 7] also entails a significant worsening of quality of life, and has further consequences in the social and work spheres of the patient's life.

Once OCD has been diagnosed, measuring the financial costs associated with health service use would seem relatively easy. However, it is difficult to measure other factors that influence the financial impact of OCD such as impairment of quality of life and the effect of the disorder on the family, social and work environments. Since OCD is a chronic disease of relatively early onset, many of its repercussions cover the entire life span of the individual [8]. Regarding socio-economic variables, OCD patients have higher unemployment rates, lower average income, worse academic achievements and higher dependence from social security allowances and payments. According to data from the British National Survey of Psychiatric Morbidity, only 48% of OCD patients perceived a salary, 16% were unemployed and 36% were financially inactive [9]. 3% of non-institutionalised OCD patients have never had a job, suggesting an even earlier onset of the disorder. Data from North America are quite similar to these results [8]. Unfortunately there are few data from Spain.

One Spanish group [10] used the SF-36 scale and compared quality of life scores of OCD patients, general population and other groups of individuals (depressed patients, hemodialysis patients and kidney transplant receptors).

The aim of this study is to evaluate quality of life perception with the SF-36 scale in a sample of OCD outpatients, in relation to the severity of the disorder, obsessions and compulsions scales, socio-demographic data, comorbidity and age of onset of the disorder. We hypothesise that: 1) OCD patients have worse quality of life perception than the general population; 2) this is related to the severity of the clinical manifestations of the disorder; 3) the most affected areas are social functioning, emotional role and mental health.

Between November 2002 and November 2004, we recruited 64 adult patients (older than 18) with OCD diagnosis (according to DSM-IV criteria) at the psychiatric outpatient clinic at Ramon y Cajal General Hospital. All patients were receiving psychopharmacological treatment and/or psychotherapy at the time of the study. All of them agreed to participate in the study and signed a consent form. The research carried out is in compliance with the Helsinki Declaration. The name of the body which gave approval is the Ethical Committee for Clinical Research (Comité Ético de Investigación Clínica) of Ramon y Cajal Hospital (record of meeting number 118, Research Project 042/03).

The statistical analysis was performed with the computer program SPSS 11.5. We used 2-tailed Student's t-tests (for dichotomous variables) and Analysis of Variance (ANOVA, for variables with more than 2 categories) to compare SF-36 subscales scores among the categories of categorical sociodemographic variables. We used Pearson's correlation to evaluate the correlations among the SF-36 subscales scores and the test scores (global Y-BOCS, compulsions subscale, obsessions subscale and Hamilton's Depression Scale).

Previous studies [17] have shown that patients with anxiety and depressive disorders have higher levels of physical, social and emotional impairment than other medical or psychiatric patients.

The results of the present study are consistent with previous findings showing that OCD patients have significantly decreased mean QOL scores for every SF-36 subscale except those related to physical health and pain in comparison to the general population. However, in a recent study Moritz et al [18] observed that OCD also affected SF-36 subscales related to physical wellbeing.

We found a correlation among Hamilton scale scores and all SF-36 subscales. This is consistent with the results of other studies [18]. However, bad QOL perception in our sample may not be attributable to depressive symptoms, since the average Hamilton scale score was approximately 10 points. OCD had a clear negative repercussion on perceived QOL, independently of affective symptoms.

Sociodemographic factors, age of onset, and years of evolution of the disorder did not significantly affect QOL perception. This is consistent with the results of some studies [19]. However, other studies suggest that the delay in OCD diagnosis [20] and the length of illness [18] worsen QOL perception.

Other studies corroborate the impact of OCD on academic and work performance, social functioning and QOL [18, 20–22]. We found that being employed was related to better scores in the SF-36 subscale of physical role. This stresses the importance of employment status on QOL in OCD patients.

We only found one study that assessed QOL perception in Spanish OCD patients [10], and concluded that QOL perception was worse in OCD patients than in the general population. The most affected SF-36 subscales were social functioning, emotional role and mental health. Their results agree with ours and with other recent studies [18] in that not all SF-36 subscales were equally affected.

We found that patients with medical comorbidity scored lower in the subscales of general health and social functioning. Other researchers have reported similar findings in OCD populations without medical comorbidity [10, 22]. In our sample, patients with psychiatric comorbidity had worse scores in the subscales of pain, general health, vitality, social functioning and mental health. Koran et al [22] only found differences in mental health perception. The discordance in the results may be related to the low psychiatric comorbidity in Koran's sample (20%) and to the fact that all patients in our sample (but not in Koran's) [22] were receiving treatment at the time of the study. Psychiatric comorbidity may be a confounding factor since it may influence QOL perception. Previous studies have reported a negative effect of psychiatric disorders on QOL perception [23–26].

The severity of OCD was correlated with all SF-36 subscales and with the highest scores in Hamilton's scale. The obsessions subscale was correlated to all SF-36 subscales, while the compulsions subscale was correlated only to social functioning, emotional role, mental health and vitality. Compulsions were not related to general health perception. Other authors have pointed out the positive correlation between social impact and severity of the clinical manifestations of the disorder [22]. It is noteworthy that the scores in the compulsions subscale in our sample were not negatively correlated with all SF-36 subscales. Masellis et al [27] obtained similar results. This might be related to the fact that compulsions are strategies to reduce the anxiety generated by the obsessions [28]. Obsessions are perceived as intrusive and uncontrollable, generate marked uneasiness and have a greater impact on QOL than compulsions, which may be considered necessary for controlling the anxiety and discomfort. Evidence-based psychological therapies for OCD pay more attention to decreasing the compulsions than the obsessions [29], and may be less useful for patients with predominantly obsessive manifestations. Between 17% and 44% of OCD patients only experience obsessions [30]. This suggests that treatments specifically aimed at reducing the anxiety related to the obsessions may improve global QOL perception in OCD patients, particularly in those with predominant obsessive symptoms. Cognitive therapy [30–33] has been shown to be effective in patients with only obsessive symptoms, though no studies have reported an improvement in perceived QOL.

There are some limitations in our study. First, the sample size makes it difficult to control for gender and comorbidity. Second, the study design is transversal, meaning that perceived QOL may be determined by multiple punctual factors. This is magnified by the lack of control group. Third, the use of general scales (instead of specific scales for psychiatric patients) to measure perceived QOL may bias the results.

Further research is needed on perceived QOL in OCD patients, with bigger samples and control groups. Another line of investigation should be to determine the degree to which psychotherapies and pharmacological treatments make emphasis on health perception and how these treatments can improve the adaptation of OCD patients in family, social and work environments.

In this sample, OCD had a clear negative repercussion on perceived QOL except in the SF-36 subscales related to physical health and pain. This suggests that not all areas of the scale were altered by the disorder.

There was a correlation among Hamilton scale scores and all SF-36 items. The severity of OCD was correlated with all SF-36 areas and with the highest scores in Hamilton's scale. The obsessions subscale was correlated to all SF-36 items, while the compulsions subscale was correlated only to social functioning, emotional role, mental health and vitality. This suggests that treatments aimed at reducing the anxiety related to obsessions may improve global QOL perception in OCD patients, particularly in those with predominant obsessive symptoms.

Sociodemographic characteristics, age of onset and years of evolution of OCD did not affect QOL perception. However, being employed was related to better scores in the area of physical role.

Patients with medical comorbidity scored lower in general health and social functioning. Patients with psychiatric comorbididity had worse scores in the areas of pain, general health, social functioning and mental health.