Obsessive-compulsive disorder (OCD) is a prevalent neuropsychiatric disorder in children, adolescents and adults. Prevalence rates of 2% to 4% have been reported. Despite the presence of effective treatments, underdiagnosis in children and adolescents is frequent. In addition, OCD is frequently comorbid with externalizing disorders such as attention deficit hyperactivity disorder (ADHD) and anxiety disorders such as generalized anxiety disorder (GAD), panic disorder, social phobia and separation anxiety disorder. Comorbidity with affective disorders is frequent too. The OCD association with major depressive disorder (MDD) increases in adolescence, reaching similar rates in adults. The comorbidity and other clinical variables, such as the duration of illness, have been mentioned as predictors of severity and treatment response. An erratic family functioning has also been associated to the severity of the illness. The aim of this study was to determine if there were differences in demographic variables (sex and age), clinical variables (age onset of the illness and comorbid disorders) and family functioning between children and adolescents with mild/moderate vs. severe OCD. Methods In a comparative cross sectional design, 60 children and adolescents aged 6 to 17 years, who met criteria for OCD and were drug naive, were assessed. Subjects who were unable to complete the interviews or had severe neurological or medical comorbidities were excluded. Subjects were recruited at the inpatients and outpatients services of the Child Psychiatric Hospital Dr. Juan N. Navarro (HPIJNN). Trained physicians interviewed the subjects and their parents using the Schedule for Affective Disorders and Schizophrenia for School-Age Children Present and Lifetime version (K-SADS-PL), the Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS), the Global Assessment of Functioning Scale (GAS) and the family APGAR rating scale. Based on CY-BOCS scores, the sample was divided in mild/moderate OCD (CY-BOCS < 30 points) and severe OCD (CY-BOCS > 30 points). Statistical analyses: Descriptive statistics (frequencies, means, standard deviation) and comparative (t student and Chi square) were used. Univariate analysis and linear regression were conducted for assessing factors associated with OCD severity. Results The sample had a mean age of 12.57±2.91 years old, they were 71.7% male. The mean duration of their illness was 19±10.75 months. The average score of CY-BOCS was 22.76±8.66, family APGAR 14.68±6.02 and GAS 54.91±15.05. A relative with OCD symptoms was reported by 23.3% of the sample; in these relatives, the most common symptoms were those related to contamination fears and washing (16%), symmetry and order (6.6%), checking (3.3%) and hoarding (1.6%). The mean number of comorbid disorders was 3.16±1.68. The most frequent were MDD, ADHD and GAD. The internalizing disorders were reported as secondary to OCD. The most common obsessions in the sample were those related to contamination and germs 60% (n=37), followed by fears to harm 58% (n=35), somatic 26.7% (n=16) and symmetry 20% (n=12). The most common compulsions were washing 50% (n=30), repeating 43% (n=26) and checking 35% (n=21). Six subjects reported contamination thoughts or rituals-related enuresis and encopresis (fear of contamination when using the bathroom or doing it only at a specific time). The main reason for attention seeking was the presence of anxiety or depressive symptoms (48%), externalized symptoms (27%), OCD symptoms (20%) and psychotic symptoms (5%). The KSADS-PL interview showed that 26% (n=16), 33% (n=20) and 43% (n=26) of the sample reported academic, social and family dysfunction respectively. When the sample was divided according to their severity, 81.7% (n= 48) belonged to the mild/moderate OCD group and 18.3% (n=11) to the severe OCD group; this had a shorter duration of illness (16.54±5.3, vs. 19.73±11.5 months; t= 0.913, df=58, p=0.003). The comparison of demographic characteristics, OCD in family members, and family functioning showed no differences between severity groups. Significant differences on rituals between groups were found, since they were reported in 45% of the severe OCD group and in 4% of the mild/moderate OCD group (?²=14.9, df=1, p= 0.001). Although the severe OCD group had shorter duration of illness, some symptoms were present for a longer time than in the mild/moderate OCD group: checking (12.6±3 vs. 2.4±4.3 months, t=3.58, df=58, p=0.001), repeating (8.3±12 vs. 3.4±5.1 months, t=2.1, df=58, p=0.03) and the inclusion of others in their rituals (20.34±1.7 vs. 2.6±4.4 months, t=2, df=58, p=0.049). Two patients of the mild/ moderate OCD group and one of the severe OCD group were affected with schizophreniform disorder. Psychotic symptoms secondary to OCD were found only in the severe OCD group (45.5%). The main reason for attention seeking of the severe OCD group was the presence of obsessions and compulsions, and in the mild/ moderate OCD group the presence of internalized disorders. Univariate analysis showed that counting compulsions (F=7.27, p=0.01) and rituals (F=17.24, p=0.000) were related to OCD severity. However, the linear regression model showed that only the presence of rituals predicted the severity of OCD (B=0.591, t=4.1, p<0.001). Discussion The studied sample represents 2.7% of the patients who were evaluated during a seven-month period at the HPIJNN, the inclusion of a screening instrument for OCD would be helpful for the identification of obsessions and compulsions in clinical samples of children and adolescents. The demographic characteristics of the present sample were similar to those reported in previous studies of pediatric OCD. The duration of illness reported by this sample suggests an age of onset around 10 years old; the frequency of obsessions and compulsions reported were also similar to those in other samples. In particular, compulsions without obsessions and the rituals involving others. Almost all the studied patients had comorbid disorders, which in many cases lead their attention seeking. It has been observed that subjects with severe obsessions regarding contamination and aggression seek for help more frequently than patients with other symptoms. Rituals were more frequently seen in patients with severe OCD; these symptoms have been associated to alterations in executive functions and have been related with psychosocial dysfunction. Conclusions: Subjects of the mild/moderate OCD group sought attention mainly due to the symptoms of comorbid disorders. Subjects with severe OCD had a shorter duration of illness and more frequency of psychotic symptoms. In the present sample, rituals predicted the severity of OCD (B=0.591, t=4.1, p<0.001). [ABSTRACT FROM AUTHOR]