Clinical profile of obsessive-compulsive disorder in children and adolescents: A multicentric study from India.

CONTEXT: Data from the Western countries suggest that obsessive-compulsive disorder (OCD) in children and adolescents is associated with male preponderance, comorbid neurodevelopmental disorders, and high family loading. However, data are limited from the developing countries with respect to the demographic and clinical characteristics of OCD in children and adolescents. AIMS: To study the demographic and clinical characteristics of children and adolescents (age ≤18 years) with OCD. SETTING AND
DESIGN: This was a cross-sectional study, conducted in outpatient treatment setting, across six centers in India.
MATERIALS AND METHODS: Participants were assessed using a semi-structured pro forma for sociodemographic information, clinical characteristics, the Children's Yale Brown Obsessive Compulsive Scale (CYBOCS), Structured Clinical Interview for Diagnostic and Statistical Manual, 5th Edition Research Version, Children's Depression Rating Scale, and Family Interview for Genetic Studies.
RESULTS: The sample was largely male with a moderate illness severity. Nearly 75% of the sample had illness onset before the age of 14 years. Aggressive, contamination-related obsessions and washing, checking, and repeating compulsions were the most common symptoms. CYBOCS assessment revealed that >2/3rd of children and adolescents endorsed avoidance, pathological doubting, overvalued sense of responsibility, pervasive slowness, and indecisiveness. Family history and comorbidity rates were low. OC-related disorders were present in about 10% of the sample.
CONCLUSIONS: This study suggests that the clinical characteristics of OCD in children and adolescents in developing countries differ on certain aspects as reported from developed countries. Copyright:

INTRODUCTION

Obsessive-compulsive disorder (OCD), a debilitating condition common across childhood, adolescence, and adulthood, has a lifetime prevalence of 1%–4%.[12] It varies in its clinical presentations and comorbidity profile. It typically runs a chronic, fluctuating course and has a significant negative impact on the individual and caregivers.[3] OCD has a bimodal age of onset, the first peak at age 11 and a 2nd peak in early adulthood.[4] About 20% of all affected persons develop symptoms of OCD by the age of 10 years.[5] Childhood/adolescent OCD commonly presents with contamination, aggressive, somatic, symmetry, and hoarding obsessions, and cleaning, checking ordering and repeating compulsions.[678] Sexual and religious obsessions are more prevalent in adolescents in comparison to children or adults with OCD.[6] Childhood/adolescent-onset OCD is also characterized by a male preponderance with a male: female ratio of 2–3:1,[26] whereas adults or later-onset (age >40 years) cases have almost equal or more females, respectively.[9] Comorbid disorders are present in almost 2/3rd of childhood/adolescent;[10] common comorbidities include anxiety disorders, mood disorders, tic disorders, attention deficit/hyperactivity disorder, and conduct disorders.[11] Family studies with childhood/adolescent probands show higher rates of family history of OCD as compared to studies with adult probands.[12]

Studies from the developing world, on adults with OCD, have shown sociocultural differences in phenomenology, comorbidity, and treatment response patterns.[13] It is unclear how these differences manifest in childhood/adolescent OCD. Moreover, the challenges in diagnosing OCD in children, given the secrecy and poor insight in this group, make this disorder especially relevant for studies across sociocultural contexts. Further, while symptom dimensions and comorbidity have received considerable attention in children and adolescents with OCD, the new diagnostic category of OC-related disorders in Diagnostic and Statistical Manual, 5th edition (DSM-5)[14] and phenomena related to obsessive-belief patterns such as avoidance, pathological doubting, pervasive slowness, indecisiveness and overvalued sense of responsibility have received little attention.

This study aimed to examine the demographic and clinical profile (i.e., phenomenology, comorbidity, and family history).

MATERIALS AND METHODS

Participants were treatment-seeking children and adolescents with OCD, recruited at 6 centers in India. All the participating centers were government-run teaching institutions. Data was collected using a uniform research protocol. Institutional Ethics Committees at all participating centers approved the study. Psychiatrists experienced in diagnosing and treating OCD in this age group were involved in the supervision of data collection at each center. Children and Adolescents aged 5–17 years, with a primary diagnosis of OCD, were screened for participation. A DSM-5[14] diagnosis of OCD was confirmed using the Structured Clinical Interview for DSM-5-Research Version (SCID-5-RV).[15] Assent was sought from the children/adolescents, and one/both parents gave written informed consent. Patients were excluded if they had a history of psychosis or bipolar disorder, antedating the onset of OCD, neurodevelopmental disorders, organic brain syndrome, obsessive-compulsive symptoms occurring as part of a general medical or neurological disorder, or medication/substance-induced OCD symptoms.

Tools

Semi-structured pro forma for sociodemographic and clinical details

This proforma, specifically designed for the study recorded age, marital status, religion, education, occupation of the child/adolescent and the informant (typically a parent); family income, nuclear/joint family, from rural/urban background. Clinical details on age at onset, age at consultation, comorbidities, duration of illness, course, precipitating factor, duration of treatment, current treatment details, adequate treatment trials, treatment with psychotherapeutic methods.

Structured clinical interview for Diagnostic and Statistical Manual, 5th edition Research Version[15]

SCID-5-RV is a semi-structured interview for making diagnoses of psychiatric disorders as per the DSM-5. Psychiatrists and trained mental health professionals can use it. “Research” version, as compared to the “clinical” version, is a more comprehensive tool for generating all current and lifetime DSM-5 diagnoses, including relevant subtypes, severity, and course specifiers. Designed for the purpose of research, the SCID-5-RV permits customization for a particular study. While the SCID-5-RV is primarily tailored for adult subjects, it can be used with adolescent subjects with word modifications. In the present study, we posed the questions primarily to parents/guardians to get information about the child's/adolescent's behavior.

Diagnostic and Statistical Manual, 5th Edition[14]

The DSM-5 is the latest edition of the American Psychiatric Association's classification system for psychiatric disorders. It incorporates the widening understanding of psychiatric disorders, in terms of better definitions of clinical features, biological and developmental underpinnings.

Children's Yale Brown Obsessive Compulsive Scale

Children's Yale Brown Obsessive Compulsive Scale (CYBOCS)[16] is a structured, reliable and widely used measure of OCD symptoms in children/adolescents. It relies on information from both child and parent. It can be used from age 5 to 18 years. The severity score on the CYBOCS can be interpreted as: 0–7: Subclinical, 8–15: Mild, 16–23: Moderate, 24–31: Severe, 32–40: Extreme; change in severity over time can be studied. Items 11–17 on the CYBOCS rate insight, avoidance, obsessional beliefs-indecisiveness, overvalued sense of responsibility, pervasive slowness/disturbance of inertia, pathological doubting, and global severity. These items capture commonly seen clinical phenomena in children/adolescents with OCD. They may influence symptom severity, and have an impact on treatment outcomes.

Children's Depression Rating Scale-Revised

Children's Depression Rating Scale-Revised (CDRS-R)[17] is a brief rating scale based on a semi-structured interview with the child (or an adult informant who knows the child well). Designed for 6 to 12-year-olds, and successfully used with adolescents, it can be administered in just 15–20 min and easily scored. The interviewer rates 17 symptom areas on a 7-point Likert scale so the CDRS-R can capture slight but notable changes in a child's symptoms. The CDRS-R gives a single summary score.

Family Interview for Genetic Studies

The Family Interview for Genetic Studies (FIGS)[18] gathers data on the history of psychiatric disorders in family members. It is a clinician-administered tool that comprises of three parts – general screening questions, face sheet, and the symptom checklist. Symptoms checklists help establish the best estimate diagnosis in family members. The utility of FIGS lies in that family members do not have to be interviewed in person, and the patient/caregiver can provide information about others.

Data analysis

Data were summarized using means and standard deviations (SDs) for continuous variables and frequencies and percentages for categorical variables. Where needed, Student's t-test or ANOVA, and Chi-square test were used to test group differences for continuous and categorical variables, respectively.

RESULTS

Sociodemographic features

A total of 173 children were included from 6 centers. Five of the six centers contributed between 15% and 23% of the sample each; only one center had a contribution <10%. Males (n-113; 65.3%) comprised 2/3rd of the study sample. Mean age of the sample was 14.4 ± 2.5 years, with a range of 6–17 years. Almost all (97.7%) children/adolescents were currently studying. A majority of the sample was Hindu (87.3%), and almost 2/3rd were from urban (64.2%), nuclear (62.4%) families [Table 1].

Table 1

Sociodemographic and clinical characteristics of sample (n=173)

Variable	Mean (SD)/frequency (%)
Gender distribution (males), n (%)	113 (65.3)
Age (years)	14.4 (2.5) (range=6-17)
Education (number of years)	8.1 (3.1)
Nuclear family	108 (62.4)
Urban residence	111 (64.2)
Age at onset of OCD (years)	12.3 (2.7) (range=5-17)
<5	0 (0)
5-7	11 (6.4)
8-10	34 (19.6)
11-14	86 (49.7)
>14	42 (24.3)
Age at first consultation (years)	13.5 (2.8) (range=6-17)
Lag duration in seeking treatment (years)
Mean (SD)	1.18 (1.34)
Range	0-6
Median	0.8
<1	55 (52.7)
1-3	63 (37.2)
3-5	15 (8.9)
>5	2 (1.2)
Duration of illness (years)	2.2 (1.8) (range=0.17-10)
Duration of treatment (years)	1.1 (1.6) (range=0-10)
Adequate treatment with at least 1 SRI	68 (39.3)
Adequate treatment with ≥2 SRIs	12 (6.9)
History of past psychotherapy	62 (35.8)
Episodic course	20 (11.6)
Presence of precipitating factor	21 (12.1)
Comorbid psychiatric illnesses, n (%)	57 (32.9)
Depressive disorder	33 (19.1)
Substance use disorder	5 (2.9)
Anxiety disorder	20 (11.6)
OCRDs	17 (9.8)
Hoarding disorder	8 (4.6)
Body dysmorphic disorder	4 (2.3)
Trichotillomania	6 (3.5)
Skin picking	2 (1.2)
Feeding and eating disorder	4 (2.3)
Somatic symptom disorder	3 (1.7)
Externalizing disorders	10 (5.8)
ADHD	11 (6.4)
Intermittent explosive disorder	8 (4.6)
Trauma and stress related disorder	2 (1.2)
Tic disorders	6 (3.5)
Childhood depression rating scale scores, mean (SD)	31.6 (14.9) (range=17-75)
Score <20 (no depression), n (%)	61 (35.3)
Score 20-30 (borderline), n (%)	31 (21.4)
Score >30 (depression), n (%)	75 (43.4)
Family history of psychiatric illnesses in first degree relatives, n (%)	55 (31.8)
OCD	32 (18.5)
Depressive disorders	10 (5.8)
Tic disorders	9 (5.2)
Bipolar disorder	8 (4.6)
Anxiety disorders	6 (3.5)
Psychotic disorders	6 (3.5)
Substance use disorders	4 (2.3)
Feeding/eating disorders	1 (0.6)

OCD – Obsessive-compulsive disorder; SD – Standard deviation; OCRDs – Obsessive-compulsive related disorders; ADHD – Attention deficit/hyperactivity disorder

Clinical characteristics

The mean age of onset of OCD in our sample was 12.3 ± 2.76 years [Table 1]. The youngest age at onset was 5 years. Around 1/4th of the sample had an onset of illness on or before the age of 10 years; however, >75% of the participants had the age of onset ≤14 years. The mean lag duration (duration of untreated illness) in seeking treatment was 1.18 ± 1.34 years. Episodic course (11.6%) and presence of precipitating factors (12.1%) were seen in a minority. Comorbid psychiatric disorders were present in about a third of the sample. The most common comorbidity was depression, seen in around 15% of the sample followed by anxiety disorders in 11% of the sample. The mean depression scores were only in the borderline range, as per the CDRS-R. Obsessive-compulsive related disorders (OCRDs) were seen in 9.8% of the sample, with hoarding disorder, trichotillomania and body dysmorphic disorder were present in 4.6%, 3.5%, and 2.3% cases, respectively. A family history of psychiatric illnesses was found in 28.9% of the sample with most commonly morbidity in family being OCD (16.8%) followed by tic disorders (5.2%)

Assessment on the Children's Yale Brown Obsessive Compulsive Scale

The commonest lifetime and current symptom dimensions among obsessions were doubts about contamination and aggressive thoughts, while the most common compulsions were those involving washing, checking, and repeating [Table 2]. Proxy compulsions by family members were reported in about 1/5th cases at the time of assessment and by 1/4th of the cases in the lifetime. Almost a fifth of the patients reported obsessions and compulsions from the miscellaneous list on CYBOCS. Current miscellaneous obsessions included – need to know or remember (n = 14), intrusive images/music (n = 7), fear of not saying just the right thing (n = 6), fear of saying certain things (n = 2), and obsessive doubts about day to day events (n = 2). Current miscellaneous compulsions included – need to tell, ask, or confess (n = 15), measures to prevent harm (n = 15), mental rituals (n = 7), need to touch, tap, rub (n = 12), self-damaging behaviours (n = 10), excessive list-making (n = 1), and rituals involving blinking or staring (n = 1). Insight assessment on CYBOCS showed that the spread from excellent to poor insight was almost equally distributed in the sample. Insight was absent in only 5 cases (2.9%). An interesting aspect of the CYBOCS assessment was the evaluation of avoidance, indecisiveness, overvalued sense of responsibility, pervasive slowness, and pathological doubt. A majority of the sample endorsed at least mild avoidance (74%), indecisiveness (70.4%), pervasive slowness (68.8%), pathological doubts (67.7%) and overvalued sense of responsibility (63.6%).

Table 2

Assessment of the sample on the Children’s Yale Brown Obsessive Compulsive Scale

Variable	Mean (SD)/frequency (%)
YBOCS total score	17.8 (8.3) (range in sample=0-35)
YBOCS obsessions score	8.9 (4.3) (range in sample=0-18)
YBOCS compulsions score	8.8 (4.6) (range in sample=0-18)
YBOCS score category
Subclinical (0-8)	20 (11.6)
Mild (9-15)	48 (22.7)
Moderate (16-23)	51 (29.5)
Severe (24-32)	47 (27.2)
Extreme (33-40)	7 (4.05)
Variable	Current	Lifetime
CYBOCS symptom dimensions - Obsessions
Contamination, n (%)	75 (43.4)	94 (54.3)
Aggressive, n (%)	68 (39.3)	83 (47.9)
Magical thinking/superstitious, n (%)	51 (29.5)	63 (36.4)
Sexual, n (%)	39 (22.5)	44 (25.4)
Religious, n (%)	31 (17.9)	42 (24.3)
Somatic, n (%)	24 (13.9)	34 (19.7)
Hoarding/saving, n (%)	22 (12.7)	31 (17.9)
Miscellaneous, n (%)	29 (16.8)	35 (20.2)
Need to know or remember	13 (7.5)	16 (9.2)
Fear of saying certain things	3 (1.7)	4 (2.3)
Fear of not saying just the right thing	6 (3.5)	6 (3.5)
Intrusive images	5 (2.9)	7 (4.05)
Intrusive music, sounds, words, etc.	10 (5.8)	3 (1.7)
Other miscellaneous obsessions (doubts)	3 (1.7)	4 (2.3)
CYBOCS symptom dimensions - Compulsions
Washing/cleaning, n (%)	81 (46.8)	96 (55.5)
Checking, n (%)	69 (39.9)	82 (47.4)
Repeating, n (%)	67 (38.7)	83 (47.9)
Counting, n (%)	40 (23.1)	52 (30.1)
Ordering/arranging, n (%)	32 (18.5)	44 (25.4)
Hoarding/saving, n (%)	26 (15.0)	32 (18.5)
Excessive games/superstitious behaviors, n (%)	23 (13.3)	33 (19.1)
Rituals involving other persons (proxy), n (%)	35 (20.2)	45 (26.0)
Miscellaneous, n (%)	51 (29.5)	56 (32.4)
Mental rituals	8 (4.6)	10 (5.8)
Need to tell, ask, confess	10 (5.8)	11 (6.4)
Measures to prevent harm	24 (13.9)	27 (15.6)
Excessive list making	1 (0.6)	3 (1.7)
Need to touch, tap, rub	9 (5.2)	10 (5.8)
Need to do till just right	1 (0.6)	2 (1.2)
Trichotillomania	1 (0.6)	2 (1.2)
Others	1 (0.6)	1 (0.6)
Variable	Frequency (%)
CYBOCS - Insight
Excellent	44 (25.4)
Good	44 (25.4)
Fair	39 (22.5)
Poor	41 (23.7)
Absent	5 (2.9)
CYBOCS - Avoidance
No deliberate avoidance	45 (26)
Minimal	42 (24.3)
Some avoidance clearly present	41 (23.7)
Avoidance prominent	39 (22.5)
Extensive avoidance	6 (3.5)
CYBOCS - Indecisiveness
None	51 (29.5)
Mild	48 (27.7)
Moderate	48 (27.7)
Severe	22 (12.7)
Extreme	4 (2.3)
CYBOCS - Overvalued sense of responsibility
None	63 (36.4)
Slight	40 (23.1)
Significant	41 (23.7)
Severe	27 (15.6)
Extreme	2 (1.2)
CYBOCS - Pervasive slowness
None	54 (31.2)
Mild	41 (23.7)
Moderate	44 (25.4)
Severe	33 (19.1)
Extreme	1 (0.6)
CYBOCS - Pathological doubt
None	56 (32.4)
Mild	38 (22.0)
Clearly present	40 (23.1)
Prominent	37 (21.4)
Constant	2 (1.2)

YBOCS – Yale Brown Obsessive Compulsive Scale; CYBOCS – Children’s YBOCS; SD – Standard deviation

Severity of illness

The Mean CYBOCS total score of the sample was 17.8 ± 8.28. Most of the patients (60.7%) had at least moderate or higher severity of illness (YBOCS score >15). Those with longer duration of treatment lag had significantly higher severity of illness on CYBOCS (r = 0.238, P = 0.002). Further, those with higher severity of illness had significantly poorer insight into their illness (r = 0.421, P < 0.001). Avoidance, indecisiveness, overvalued sense of responsibility, pervasive slowness and pathological doubt correlated significantly with the total CYBOCS severity (r = 0.33–0.52, P < 0.001), with avoidance showing the highest correlation (r = 0.52).

Gender differences Table 3

Males and females in the sample differed on only a few parameters [Table 3]. The age at onset and seeking help for the first time were both higher for males. Females had higher prevalence of comorbid anxiety disorders and somatic obsessions. There were no statistically significant differences seen in between the two genders in terms of other sociodemographic or clinical parameters [all variables from Tables 1 and 2].

Table 3

Differences between males and females in the sample

Variable	Males (n=113)	Females (n=60)	t/χ2
Age (years), mean (SD)	14.92 (2.40)	13.57 (2.60)	3.346**
Education (years), mean (SD)	8.77 (2.99)	6.95 (2.97)	3.821***
Age at onset of OCD (years), mean (SD)	12.82 (2.65)	11.42 (2.74)	3.251**
Age at first consultation (years), mean (SD)	14.05 (2.74)	12.5 (2.76)	3.509**
Comorbid anxiety disorders, n (%)	6 (5.3)	13 (21.7)	10.72**
Somatic obsessions, n (%)	17 (15)	17 (28.3)	4.38*

*P<0.05; **P<0.01; ***P<0.001. SD – Standard deviation; OCD – Obsessive-compulsive disorder

DISCUSSION

The primary aim of this paper is to describe the clinical characteristics, comorbidity and family history in a multi-centric sample of treatment-seeking children and adolescents with OCD. To the best of our knowledge, this is the largest and only multi-centric study on childhood OCD from India. Sociodemographic characteristics in studies from the developing world are often a reflection of both neurobiological underpinnings of the illness and treatment-seeking behaviors, which are in turn affected by education, awareness and sociocultural practices. The sample in the present study largely comprised of students, who were Hindu, from nuclear, urban families. Despite being a sample from tertiary care centers in India, a significant proportion of the study sample comprised of those seeking treatment for the first time. This could possibly be due to a shortage of specialized child mental health services in the country. Our sample had a 2:1 male-female ratio. Earlier studies from developed involving children and adolescents from diagnosed with OCD have reported similar gender distribution.[261920] However, some of the studies have reported significant variations in gender distribution.[21] Following from this, sociocultural influences may have a role to play in gender distribution. In the developing world, males are often given preferential care and receive prompt health care attention in comparison to female counterparts at every age.

The Age of onset of OCD in our sample was 12.3 ± 2.7 years with age range of 5–17 years. Delorme et al.,[4] reported the first peak age of onset in OCD is around 11 years of age. In our sample, 25% patients had developed OCD by the age of 10 years, 50% by the age of 13 years, and 75% by the age of 14 years. Thus, early adolescence (between 10 and 14 years of age) is when the largest proportion of patients developed clinically salient symptoms. Kessler et al.,[5] in a large community sample from the USA reported about 20% cases of OCD develop OCD before 10 years, which is <25% onset before 10 years in our sample. Early onset cases are significant from the point of view of early identification and intervention, especially since an early onset of symptoms has been associated with poor outcome and prognosis,[2223] given the significant impact on their socio-occupational abilities in this crucial developmental phase.

Symptom dimensions in our study were consistent with earlier studies[24] with contamination and aggression-related obsessions, and washing and checking related compulsions being the most common. Patients typically had a polysymptomatic illness with multiple obsessions (mean ± S.D. = 2.0 ± 1.32) and compulsions (mean ± S.D. = 2.5 ± 1.76). One-fifth of the patients had proxy compulsions, highlighting the role of family accommodation in childhood OCD. Symptom dimensions were not different between males and females, except somatic obsessions that were higher in females. A significant proportion of children endorsed obsessions and compulsions in the miscellaneous category. During clinical assessment it is important that these symptoms are specifically asked for.

Nearly 50% of the sample had fair or lower level of insight, a finding in keeping with descriptions of childhood OCD.[25] A poorer insight was associated with higher illness severity on CYBOCS, and with lag in treatment seeking, consistent with the role of insight in outcome of illness.[26] Poor insight complicates assessment and adversely influences compliance to pharmacological and nonpharmacological treatment.

We used the CYBOCS to assess associated phenomenon in OCD. In adults, dysfunctional beliefs like over-estimation of threat and inflated responsibility, importance and control of thoughts, and perfectionism and intolerance of uncertainty were considered to be of key importance in the development and maintenance of OCD.[27] There is no available instrument to assess these in children; however, the CYBOCS has items that assess these. Avoidance includes behaviors wherein a patient with OCD avoids doing certain activities, or being in certain places and with certain people because of the possibility that obsessions and compulsions may be triggered. It is considered one of the key factors in maintaining OCD. Almost 75% of our sample reported avoidance. Indecisiveness (a pervasive difficulty in making decisions about day-to-day activities), pervasive slowness (a difficulty in starting or finishing tasks, and taking longer that typically needed in routine activities), pathological doubting (a constant tendency to doubt the completion or correctness of day-to-day activities) and an overvalued sense of responsibility (feeling responsible for the consequences of one's actions, including significant self-blame for events that may even be out of one's control) are other features associated with OCD. Incorporating an assessment of these dimensions in a child aids a comprehensive understanding about the illness, the child's experiences, and may contribute to effective cognitive-behavior therapy based treatment. In our sample 65%–70% of children endorsed these. An overvalued sense of responsibility has been most consistently associated with OCD.[28] Salkovskis et al.[29] proposed that experiencing heightened responsibility, overprotective parents and rigid rules, and thinking that one influenced or caused a negative life event act as “pathways” to the development of inflated responsibility beliefs, thereby increasing risk for OCD. In fact, overvalued sense of responsibility has been proposed as a target for both prevention and management in OCD;[28] cognitive restructuring techniques to address this phenomenon would also help facilitate insight into illness. These phenomena are understudied in children. Avoidance, indecisiveness, pervasive slowness, and overvalued sense of responsibility are related to functional impairment in childhood OCD and they reduce with treatment.[30] As is evident from our dataset, these phenomena are prevalent in children and adolescents and can be elucidated from patients and their caregivers. It is imperative that clinicians ask for these during the evaluation of a child/adolescent with OCD.

Comorbidity rates of our sample were much lower than western studies that have reported comorbidity in up to 80% cases.[3132] The most common comorbidities included depressive, and anxiety disorders. Anxiety disorders were commoner in females compared to males. A part of the lower comorbidity rate could be due to the exclusion of children with comorbid severe or organic mental illnesses. A study from India,[33] on comorbidities in juvenile-onset OCD also showed a low overall comorbidity rate compared to Western literature. However, they did find a higher rate of disruptive behavior disorders and tic disorders compared to our study. Use of an older diagnostic system (DSM-III-R) may partly explain the difference in rates. It may be that since our sample had a relatively later onset of symptoms in comparison to some other studies,[2134] a typical comorbidity pattern of tics and externalizing disorders may not be seen. Previous studies have suggested that cases with onset of illness after 12 years of age resemble adult OCD cases.[35] Around 10% of our sample had comorbid OCRDs-hoarding disorder (4.6%), trichotillomania (3.5%) and body dysmorphic disorder (2.3%). The prevalence of these conditions in children and adolescents is not well ascertained. They typically start in middle-late adolescence[14] and may be missed if not specifically enquired for in a child presenting primarily with OCD. The implications on treatment, and outcome and their long-term course need further systematic study. Rates of family history for psychiatric illnesses in our sample were low, similar to another study from India.[36] The commonest illness in first-degree relatives was OCD (16.8%) followed by tic disorders (5.2%). The family history data in this study relied on report from the primary caregiver of the patient. This method is known to underestimate the rate of family history;[12] moreover, we focussed on assessing family history in only first-degree relatives.

Strengths and limitations

Our study sample comprised of treatment-seeking children from tertiary care teaching institutions in north India with specialized child psychiatry units. This may not be representative of the community or other clinical centers. However, due to lack of specialized child psychiatry in India, these centers do see a lot of first contact or early referral cases and not just difficult to treat or treatment resistance cases. Cross-sectional design of the study may have led to some recall biases, however, care was taken to interview both child and at least one parent to reduce the risk of incorrect information. The absence of functionality assessments, more detailed family accommodation measures, and medical evaluations for possible autoimmune pathogenesis are limitations of our study. Strengths of our study include data collection from a large sample, multi-centric, using a uniform protocol and standardized tools.

CONCLUSIONS

In a first multicentric study on childhood OCD from India, in 173 children/adolescents with a DSM-5 diagnosis of OCD, we found early adolescence (10–14 years) as the most common age of onset, a male predominance, delay in treatment seeking of about a year. Rates of comorbidity and family history were low. OCRDs were found in about 10% of the sample. On the CYBOCS, avoidance, indecisiveness, pervasive slowness, pathological doubting, and an overvalued sense of responsibility were reported in 2/3rd of the sample. OCD in children and adolescents has been touted as a distinct phenotype.[37] Our study adds to this understanding, especially with its report on the ancillary CYBOCS symptoms and OCRDs.

Financial support and sponsorship

Indian Psychiatric Society: Research Grant to Committee for Research, Education and Training Foundation.

Conflicts of interest

There are no conflicts of interest.