Background
The existence of comorbid OCD in patients with psychotic disorders has been recognised for over a century, but the reports have been less consistent with regard to the extent of the comorbidity and the influence on the clinical expression. Comorbidity rates are in general reported to be higher than the prevalence in the general population. In a recent meta-study Achim et al. estimated comorbidity rates for OCD in patients with schizophrenia at 12% [
1].
Several studies have proposed that comorbid OCD affect the clinical expression of patients with psychotic disorders. Comorbid OCD is related to more impairmed social functioning [
2,
3] and more neurological abnormalies [
4‐
6]. Patients with schizophrenia and comorbid OCD have been reported to have earlier onset of psychosis, and an increased rate of depressive symptoms and suicidal attempts [
7‐
9], compared to their non-OCD counterparts. However, these patterns have not been constantly reported [
10,
11].
Most studies of comorbid OCD in psychotic disorders have been conducted on patients with schizophrenia in large psychiatric institutions or day treatment centres. The samples often consist of chronic patients, which introduce several limitations related to the effects of illness duration, institutionalisation, and aggravation or induction of second-onset OCD during antiserotonergic antipsychotics [
12‐
14]. To increase the understanding of comorbid OCD in psychosis, there is a need for studies where these methodological challenges are met.
The present report is from a representative and relatively large sample of patients with first-episode psychosis (FEP), recruited in a population-based catchment area in an early phase of the psychotic disorder. The primary aim was to investigate the prevalence of OCD in the FEP-patients, with strict diagnostic methodology. The secondary aim was to investigate whether the patients with comorbid OCD differed from the patients without OCD on socio-demographic characteristics, type of psychotic disorder, general functioning, and severity of positive and negative symptoms at the start of treatment.
Discussion
In this study we found the prevalence rate of comorbid OCD among patients with FEP to be 10.6%. The reported prevalence is higher than the prevalence of OCD in FEP in studies by Craig et al. [
13], Sim et al. [
29], and Sterk et al. [
30]. The independent reassessment of the OCD-diagnostics raised the prevalence from 6.5% to 10.6%; this was a procedure that, to our knowledge, was not presented in other studies of OCD in FEP-patients. Our sample included affective psychosis with mood-incongruent delusions in the FEP-sample. Since this was the diagnostic group with the highest number of OCD-patients in the sample, the inclusion of this patient group in the sample may have raised the OCD-prevalence.
Our study indicated a significantly lower comorbidity of OCD in patients with schizophrenia spectrum disorders. Most noteworthy is the finding that among the patients diagnosed with schizophrenia as their primary diagnosis, none were diagnosed with comorbid OCD. The prevalence of OCD for patients diagnosed with schizophrenia spectrum disorders was in line with the studies from Craig et al. [
13], Sterk et al. [
30] and Sim et al. [
29], but was
Table 1
Baseline comparison of demographic characteristics of OCD and non-OCD first episode psychosis
Age (SD) | 22.96 (8,007) | 27.05 (10,321) | .053* |
DUP in weeks mean (SD)/median | 61.3(113.9)/14 | 80.5 (204.9)/20 | .497 |
Male (%) | 43.2 | 58.6 | .112 |
Single marital status (%) | 87.5 | 74.5 | .116 |
Drug abuse % | 15.4 | 29.8 | .122 |
Years of education mean (SD) | 11.4 (1.7) | 11.8 (2.6) | .501 |
Table 2
Comparison of diagnostics between OCD and non-OCD in first episode psychosis
Schizophrenia spectrum disorders | 5 (19) | 87 (40) | .043* |
Schizophrenia | 0 (0) | 52 (24) | |
Schizophreniform disorder | 2 (8) | 10 (5) | |
Schizoaffective disorder | 3 (12) | 25 (11) | |
Brief psychosis | 0 | 13 (6) | |
Delusional disorder | 2 (8) | 19 (9) | |
Affective psychosis MID | 10 (38) | 45 (20) | |
Psychosis NOS | 8 (31) | 34 (15) | |
Drug induced psychotic disorders | 1 (4) | 19 (9) | |
Psychotic disorder NOS | 0 | 3 (1) | |
lower than the prevalence reported in a study of first-episode schizophrenia spectrum disorders by Poyurovsky and colleagues [
12] and in a study of first episode schizophrenia and related disorders by de Haan [
31]. In our study, OCD was most prevalent among patients with affective psychosis and psychosis NOS.
The second purpose of the study was to compare demographic and clinical characteristics between the OCD and the non-OCD group. The FEP-patients with OCD were found to have a significantly younger age at the start of treatment than their non-OCD counterparts. This
Table 3
Comparison of clinical characteristics of OCD and non-OCD in first episode psychosis
GAF Symptoms (SD) | 33.1 (5.7) | 31.2 (7.4) | .138 |
GAF Function (SD) | 40.8 (10.4) | 39.5 (9.5) | .524 |
PANSS Total | 65.4 (14.1) | 64.61 (13.8) | .699 |
Positive component (SD) | 14.5 (3.9) | 14.5 (3.9) | .898 |
Negative component (SD) | 19.7 (8.3) | 18.7 (7.3) | .542 |
Excitement component (SD) | 8.9 (2.7) | 7.9 (3.3) | .174 |
Depressive component (SD) | 14.7 (3.1) | 12.3 (3.6) | .002** |
Cognitive symptoms (SD) | 5.5 (2.8) | 5.8 (2.9) | .634 |
Suicidal behaviour; plans or attempts (index) (%) | 26.9 | 9.6 | .009** |
Suicidal behaviour; plans or attempts (lifetime) (%) | 34.6 | 27.3 | .431| |
is in line with the findings from Craig and colleagues [
13], who reported that a significantly larger portions of the FEP-patients with obsessive compulsive symptoms (OCS) was younger than 21 years compared to FEP- patients without comorbid OCS. Poyurovsky and colleagues [
32] reported that adolescent schizophrenia patients with OCD had an earlier onset of psychosis compared with their non-OCD counterparts, and Üçok and colleagues [
7,
11] reported that among schizophrenia patients, comorbid OCD was related to earlier debut of schizophrenia. No age difference in debut of psychosis were found between OCD and non-OCD patients in hospitalised first-episode schizophrenia patients [
12] or in a sample of patients with first-episode schizophrenia and another related disorder [
10]. A possible earlier age of debut, however, is noteworthy since it may indicate an accentuated neuro-developmental origin of OCD in patients with psychosis [
32].
Patients with co-morbid OCD reported a higher rate of suicidal plans or attempts at index point. This finding is in line with the findings from Üçok et al. [
7] and Sevincok et al. [
8]. However, this pattern was not reported in a more recent study by Üçok et al. [
11]. A heightened rate of suicidal plans or attempts in the OCD-group is an important finding, especially given that previous studies have indicated that suicidality at index point is a predictor for continued suicidality in patients with FEP [
23].
The patients with co-morbid OCD scored higher on the depression component of the PANSS. This in line with de Haan et al. [
31], who reported that, among patients with recent onset schizophrenia and related disorders, patients with co-morbid OCD were more depressed, as measured by the Montgomery-Asberg Depression Rating Scale (MADRS). The increased prevalence of depressive symptoms and suicidal plans and attempts may be a result of living with OCD, which is supported by findings from Rasmussen & Tsuang [
33], who reported that up to 75% of patients with an OCD diagnosis also fulfilled the criteria of a depressive disorder. A possible confounding variable may however be the higher number of patients with affective psychosis among the patients with comorbid OCD. Fontenelle et al. [
9] argued that depression may serve as a vulnerability marker for OCD in psychosis since patients with OCD and psychosis reported a higher rate of depressive symptoms both before and after conversion from ultra-high risk for psychosis. The increased depressive symptoms in patients with co-morbid OCD may be clinically important, given findings that people with psychotic disorders and concurrent depressive symptoms have poorer long-term functional outcomes compared to the non-depressed [
34], and depressive symptoms have been found to be related to related to an increased risk of suicidality in FEP-patients [
35].
In our study, we found no differences between the OCD patients and their non-OCD counterparts in total PANSS score, or in components other than the depressive factors on the five component PANSS analysis [
28]. This findings are in line with the results from de Haan and colleagues [
10], who did not find any significant relationship between the presence of OCD or OCS and psychotic symptoms in a sample of patients with first episode schizophrenia and related disorders. In a recent meta-study of Cunill et al. [
36] they did not report significant differences in PANSS scores between schizophrenia patients with and without OCD. The results from our study may support the hypothesis from Berman and colleagues [
37] that obsessive-compulsive symptoms are independent of psychotic symptoms.
It is difficult to compare studies since the prevalence of OCD varies significantly, and many studies have selected samples. A common bias in these estimates of comorbidity has also been the enrolment of patients with OCS and not strictly OCD. This has resulted in a gross overestimaton of co-morbid OCD in many studies [
38]. The diagnostics of OCD in patients in our study were based on strict evaluation of the criteria for OCD, based on the guidelines suggested by Bottas and colleagues [
39]. The diagnostics were made in three independent steps, to ensure the quality of the assessment. The reassessment of the diagnostics increased the number of OCD-cases from 16 to 26. Interestingly, the re-examination did not reveal any false-positive OCD cases, which may indicate that false negative results may be a more problematic issue in diagnostics of OCD based on SCID-I in FEP-patients than false positive. These findings indicate that an increased focus on OCD in the diagnostics of FEP-patients is clinically important. This may reduce the rate of false negatives, as the diagnostics with SCID-I may underestimate the prevalence of OCD in FEP-patients, even when well-trained assessors carry out the interview. Correct diagnostics are central for optimal treatment [
29,
38,
40].
The study has several strengths. First, the sample is large and from a well-defined catchment area. We assume that nearly every first episode case has been identified and we regard our sample as being highly representative. Second, we independently reassessed all the patients for OCD to ensure the validity of the OCD-diagnostics.
A limitation of the study is that since the primary focus of the parent study was psychosis, there were no assessment instruments specific to OCD included, such as the Yale-Brown Obsessive Compulsive Scale [
41,
42]. Another limitation is that depressive symptoms were quantified with a second-order method, the PANSS subscale. A third limitation is lack of information about prior treatment (SSRI/CBT) for the patients. It should also be noted that 40% of eligible patients declined to participate, so it is unknown what their diagnoses were and therefore to what extent this may affect the representative nature of the sample.
Competing interests
The authors declare that they have no conflict of interest.
Authors’ contributions
KH, BH, IJ, and TKL contributed to the study design. KH, BH, IJ, and TKL contributed to data collection. KH and IJ conducted the statistical analysis. KH, BH, IJ, and TKL interpreted the data and drafted the manuscript. All authors participated in critical revision of manuscript drafts and approved the final version.