Pica (disorder)

Background & objectives: There are limited data on child mental health needs in our country. Therefore, an epidemiological study to determine the prevalence rates of child and adolescent psychiatric disorders was initiated as a two-centre (Bangalore and Lucknow) study by the Indian Council of Medical Research. It also aimed to study the psychosocial correlates of the psychiatric disorders. We present here the findings of Bangalore Centre.

Methods: In Bangalore, 2064 children aged 0-16 yr, were selected by stratified random sampling from urban middle-class, urban slum and rural areas. The screening stage was followed by a detailed evaluation stage. The ICD-10 DCR criteria were used to reach a penta-axial diagnosis.

Results: The results indicated a prevalence rate of 12.5 per cent among children aged 0-16 yr. There were no significant differences among prevalence rates in urban middle class, slum and rural areas. The psychiatric morbidity among 0-3 yr old children was 13.8 per cent with the most common diagnoses being breath holding spells, pica, behaviour disorder NOS, expressive language disorder and mental retardation. The prevalence rate in the 4-16 yr old children was 12.0 per cent. Enuresis, specific phobia, hyperkinetic disorders, stuttering and oppositional defiant disorder were the most frequent diagnoses. When impairment associated with the disorder was assessed, significant disability was found in 5.3 per cent of the 4-16 yr group. Assessment of felt treatment needs indicated that only 37.5 per cent of the families perceived that their children had any problem. Physical abuse and parental mental disorder were significantly associated with psychiatric disorders.

Interpretation & conclusions: Prevalence rates of psychiatric morbidity in 0-16 yr old children in India were found to be lower than Western figures. Middle class urban areas had highest and urban slum areas had lowest prevalence rates. The implications for clinical training, practice and policy initiatives are discussed.

Key words Childhood mental health problems - epidemiology - health policies for children - prevalence - psychiatric disorders

Children under 16 yr of age constitute over 40 per cent of India's population and information about their mental health needs is a national imperative. From the early 1960s, there have been efforts at conducting epidemiological studies in community, clinic and school settings. Community surveys have the advantage of being more representative; they include children and adolescents who do not attend school and those who do not access mental health services.

Early Indian studies reported prevalence rates of psychiatric disorders among children ranging from 2.6 to 35.6 per cent1-5. A comparatively recent and methodologically superior study6 reported a rate of 9.4 per cent in a sample of 1403 rural children aged 8-12 yr. This small group of community-based studies provided some benchmarks for the rates of psychiatric disturbance among Indian children. Reasons for the wide disparity among the prevalence rates included the varying modes of case ascertainment, sampling methods, instruments and the informants chosen across the studies. The methodological lacunae included the use of purposive sampling techniques, single-stage sampling designs, small sample sizes, and unspecified clinical criteria for case ascertainment. Most studies did not have multiple informants (e.g., parent, child, teacher) and did not assess the disability caused by the psychiatric disorder. In addition, there is no information about the prevalence of psychiatric disturbance among adolescents. It is difficult to comment about differences between rural and urban areas, given the lack of studies using a uniform methodology in both settings. Some epidemiological studies in other countries7 found that a substantial proportion of individuals who met symptomatic criteria for a diagnosis appeared to be functioning normally. This suggested that caseness might be best judged by the presence of both symptoms and impairment. There is only one Indian study6 that included impairment criterion, resulting in a drop in the total prevalence rate from 9.4 to 5.2 per cent. An understanding of the associated degree of disability can help in assessing the need for secondary and tertiary services. The lack of information about the psychosocial correlates of psychiatric disorders also limits the clinical and policy implications of Indian studies.

An earlier epidemiological study by us suggested a lower prevalence of psychiatric morbidity when compared to studies in Western countries8. Epidemiological data from Indian studies would be critical in outlining needs for mental health services, training requirements for mental health professionals, planning for optimum resource deployment, and formulating government policy initiatives as well as for cross-cultural comparisons. Many critical methodological issues need to be integrated into these studies. The need for dependable epidemiological data led the Indian Council of Medical Research (ICMR) to initiate a multicentre study (Lucknow and Bangalore) to determine the prevalence rates of psychiatric problems in children and adolescents from urban and rural areas. In addition, it also aimed to study the psychosocial correlates of the psychiatric disorders and to assess the perceived needs of the family for help or treatment. The present paper focuses on the findings related to the prevalence of psychiatric morbidity at the Bangalore Centre, National Institute of Mental Health and Neuro Sciences (NIMHANS).

Material & Methods

Sample

A sample size of at least 2000 was estimated to ensure that cases of psychiatric disorders with an expected prevalence rate of 1 per cent or higher were not missed out. The sample was selected by stratified multistage sampling from middle class urban, urban slum and rural areas. All children below 16 yr from each selected family were included in the study after obtaining informed consent.

Middle class urban areas: A primary census abstract of 1991 census of Bangalore urban area was obtained. There were 87 divisions, each divided into blocks, ranging from 26 to 157 blocks. Five divisions were selected randomly and then one block randomly chosen from each division. Informed consent was sought from families in every house.

Urban slum areas: Lists of the slums in Bangalore city were obtained from the Slum Clearance Board and 28 slums with a population of 500 and above were considered. Eight slums were then chosen based on the language (Kannada) that the majority of inhabitants spoke. Three slums were randomly chosen from the list of eight slums. Every alternate house, every third house and every fifth house was selected from each of the three slums respectively, with the first house for each slum being chosen randomly.

Rural areas: Three taluks from among eight taluks in Bangalore rural district were randomly chosen for the study. Three villages from each taluk, with a population between 1000 and 2000, were randomly chosen and visited. One village was finally chosen from each taluk and all the families were approached for consent.

Tools

(i) Socio demographic proforma (SDP) - This 17-item proforma was developed based on existing proformas and expert discussions. It elicited details like type of area, informant, name of the head of the household, address, caste/religion, family income, family size and number of children below the age of 16 yr.

(ii) Screening checklist (SCL) - This 33 item checklist covers a range of behaviour problems usually present among children aged 0-3 yr and was developed by combining the Child Behaviour Checklist (CBCL) for 2-3 yr9 and the unpublished Behaviour Checklist developed at the All India Institute of Medical Sciences, New Delhi (Mehta M, personal communication). Each item was rated 0 (absent), 1 (sometimes/somewhat true), and 2 (very true/often true). A score of 1 or 2 on any item was used as a cut off.

(iii) Child behavior checklist (CBCL) - The 113-item behaviour problems section of the CBCL10 was used as a screening tool for 4-16 yr olds. Each item was rated by the parent as 0 = not true, 1 = sometimes or somewhat true, and 2 = very true or often true. The original cut off scores were found to be too high for the Indian cultural setting and were modified based on unpublished data collected at Ranchi and discussions with experts at the Indian Council of Medical Research (Male: 6 to 11 yr ≥ 21, 12 to 16 yr ≥ 18; Female: 6 to 11 yr ≥ 16, 12 to 16 yr ≥ 13).

(iv) Additional module (AM) - An additional module with 12 items in a yes/no format was added to screen for developmental disorders, scholastic problems, epilepsy, mental retardation, enuresis and phobias. The CBCL is not sensitive to these diagnoses, as the total scores would be lower than the cut off. Hence a score of >1 on the AM was considered as the cut off.

(v) Children's behaviour questionnaire (CBQ) - The Children's Behaviour Questionnaire11 was rated by teachers. Part A sought qualitative information about academic performance, attendance, leadership, nicknames, handicaps and teacher's opinion about need for psychological help. Part B had 26 items covering common emotional, conduct and behaviour problems in the school setting. The scale was rated as 0 (does not apply), 1 (applies somewhat) and 2 (certainly applies). A score of 9 and above was used as the cut off11.

(vi) Felt treatment needs (FTN) - This 8-item questionnaire in a yes/no format was prepared to assess parental awareness of their child's problems, their perceived need for treatment, type of management and the expected outcome.

(vii) Diagnostic interview schedule for children (DISC) - The DISC12 is a structured diagnostic interview schedule, with two parallel versions. One with the parent as informant (DISC-P) for children aged 6 yr and above and the other with the child as informant (DISC-C), for children aged 12 yr and above. Six independent modules were used to assess children for disorders as per ICD - 10/DCR13 criteria in Rutter's penta-axial scheme14.

(viii) Structured interview schedule (SIS) - The SIS was compiled from the DISC - P to evaluate children aged 4-6 yr. It assesses disorders like generalized anxiety disorder, separation anxiety disorder, attention deficit hyperactivity disorder, oppositional defiant disorder, conduct disorder and elimination disorder, that are most likely to occur in this age group.

(ix) Parent interview schedule (PIS) - A shortened version of this semi-structured interview schedule15 was used to assess deviance or disturbance in the child's family and environment. The areas covered were intrafamilial communication, qualities of upbringing, nature of immediate environment, life events, societal stresses, and interpersonal stress at home, school or work. Some culturally relevant items like excessive parental control, family level stresses/life events, change of academic environment, and excessive academic pressure at school, were also added.

(x) Vineland social maturity scale (VSMS) - An Indian adaptation16 of the Vineland Social Maturity Scale17 was used to assess children aged 0-16 yr in the areas of self-help general, self-help dressing, self-help eating, self-direction, locomotion, communication, occupation and socialization. The scale yields a social age and a social quotient, which can be considered a proximate intelligence quotient.

(xi) Binet kamat test (BKT) - The original Stanford-Binet test was modified and standardized to measure general mental ability for the age group of 3-22 yr. This Indian adaptation18 has items at each age level and yields a mental age and intelligence quotient.

(xi) Specific learning disability (SLD) battery - A battery of tests to assess attention, reading, writing, spelling, comprehension, arithmetic, visuo-motor skills and auditory and visual memory was compiled at NIMHANS19. This battery has face and content validity. If the child's performance was 2 classes below what was expected for him/ her, the diagnosis of SLD was made. The ICD-10/DCR13 category of specific developmental disorder of scholastic skills is referred to as specific learning disability (SLD) for convenience.

(xiii) Children's global assessment scale (C-GAS) - The C-GAS is an adaptation of the Global Assessment Scale20, and was developed by Schaffer et al21. It reflects the lowest level of functioning of the child during a specified period of time and measures the degree of functional impairment. A cut off of 70 or less (Bird H, unpublished data) was used in this study. The lower ranges 41-50 denote moderate degree of interference in functioning in most social areas, 51-60 is variable functioning with sporadic difficulties or symptoms in several, but not all social areas, 61-70 indicates some difficulty in a single area but generally functioning pretty well.

(xiv) Physical examination proforma (PE) - A pro forma was developed to enter the details of physical examination findings. This included a general physical examination and a systems review.

Procedure

The instrument adaptation phase was followed by the second phase where the main study was conducted. The period of the study was from September 1995 to April 2000.

1st Phase - Instrument adaptation phase - The tools numbered 2 to 9 were translated into Kannada by a bilingual project staff member and subsequently circulated among clinicians knowing both the languages for comments. The final draft was given to bilingual lay people for back translation. Necessary modifications were done after the back-translated version was compared with the original instrument.

A pilot study was conducted in the clinic and community settings to assess the feasibility, applicability and reliability of the tools. The inter-rater reliability of the CBCL10 was satisfactory (r=0.998). The additional module (AM) was added to increase the sensitivity and specificity of the CBCL10 for certain disorders. A sample of 66 children aged 4 to 16 yr who received an ICD-10-DCR13 diagnosis in the Child Psychiatry Clinic at NIMHANS Bangalore, were also administered the CBCL and the AM. Fifty-nine children (89.4%) scored above the modified cut off score of the CBCL and/or the AM.

2nd Phase - Main study - The main study commenced with the screening followed by the detailed evaluation stage. All assessments were done in the community. A flow chart indicating the instruments used at the two stages, with different age groups, is outlined in the Fig. Field workers, who had a Masters In Psychology and clinical experience, or had an M.Phil. (Clinical Psychology) administered the VSMS16, BKT18 and SLD19 battery. Field workers with a postgraduate diploma or degree in Psychiatry did the physical and clinical examination. All other tools were administered by any of the field workers. Training for the DISC12 was given over a three to four day period using the DISC User's Manual.

Screening stage - All children aged 0-16 yr were screened using age appropriate screening instruments. Children aged 0-3 yr were screened using the SCL and VSMS16. Children aged 4-16 yr were screened using the CBCL10 with a family member as an informant. The CBQ11, filled by teachers, was also used to screen school going children.

Detailed evaluation stage - Children selected as positive in the first stage were taken for a detailed evaluation.

Parents of children aged 0-3 yr were interviewed with the FTN questionnaire after which the doctor conducted physical examination of the children. ICD-10-DCR13 diagnoses were assigned after clinical interviews.

The SIS, PIS15 and the FTN questionnaire were used for children aged 4-5 yr. A detailed physical examination was done. The Vineland Social Maturity Scale16 or Binet- Kamat Test18 were used when the score was 1 or 2 on questions reflecting intellectual and/or academic backwardness on the AM.

The DISC-P12, PIS15 and the FTN questionnaire were used for children aged 6-16 yr. Children between 12-16 yr were also assessed using the DISC-C12. Intellectual assessment was done using the Vineland Social Maturity Scale16 and Binet-Kamat Test18, wherever necessary. Children were also assessed with the SLD battery19 based on teacher or parental report of academic problems, despite having normal intelligence. A doctor physically examined all children. Each child was also assessed on the C-GAS21 by the interviewer.

The final diagnosis was assigned based on the results of the detailed assessments. For children aged 12-16 yr, information from both the DISC-P12 and DISC-C12 was combined, with the parent's version being given more weightage. In the case of psychotic, obsessive and substance use disorders, the child's information was acceptable even if contradicted by the DISC-P12. Clinician's judgment was used to combine the information and make the final ICD-10-DCR13 diagnosis. Rutter's penta-axial system14 was used to record diagnostic information. The five axes were (i) Psychiatric diagnoses; (ii) Specific developmental delays; (iii) Intellectual level; (iv) Significant non-psychiatric medical diagnoses; and (v) Associated abnormal psychosocial conditions. The 1st axis had three possible diagnoses, the 2nd axis had two possible diagnoses, the 3rd axis had one diagnostic possibility, the 4th axis had two diagnostic possibilities and the 5th axis had five possibilities.

Of the children who were negative on the 1st stage evaluation, 10 per cent were chosen randomly for a 2nd stage evaluation. This was done to check for 'false negatives' during the screening and thereby evaluate the sensitivity and specificity of the tools.

The average time taken for the screening and evaluation ranged from 30 to 90 min for the 0-3 yr group. For the 4-16 yr group, it was 120 to 300 min. An additional 70 min was taken when the BKT18 and SLD battery19 was administered.

Statistical analysis was done using frequencies and percentages, and by applying Chi-square tests.

Results

The total number of houses approached in the different areas numbered 1484. Forty three (2.9%) families refused to take part in the study, 66 (4.5%) houses were locked and 31 (2.1%) houses were empty. Of the 1344 families who gave informed consent for the study, 346 (25.7%) had no children below the age of 16 yr. Initially, 2082 children from 998 families were included in the study. Of these, 18 children from the urban areas could not be screened (13 refused, 3 migrated from the area and 2 were left pending). Therefore, the total sample screened was 2064, with 833 children from rural areas, 629 from urban slum areas and 602 from middle class urban areas. The sample included 486 (23.5%) children in the 0-3 yr age group and 1578 (76.5%) in the 4-16 yr age group (Table I).

Chi-square analysis indicated no significant differences between the number of male and female children in each of the three areas. Urban middle-class families tended to be small, with almost half comprising of 1-4 members per household. In comparison, medium to large sized families (5-10 members) made up almost two-third of the slum and rural samples. About 50 per cent of the rural sample and 33-36 per cent of the slum and urban middle class samples were either unaware of or reluctant to divulge the family income. As expected, the family income was highest in the urban middle-class areas, followed by urban slum and rural areas. Urban families far outnumbered the other two areas in having just one child, while having three or more children was far more common in urban slum and rural areas.

Chi square analyses also indicated that rates of consanguinity were significantly (P

The response rates for the SCL, CBCL and AM, PIS, FTN Questionnaire and CBQ were very good (100, 98.9, 92.0, 91.6 and 90.2% respectively). The response rates for the DISC were lower due to refusal or incomplete assessments, with the rates for DISC-P and DISC-C being 88.3 and 67.1 per cent respectively.

The screening stage involved 2064 children and adolescents aged 0-16 yr. The 505 (24.5%) selected as positive on the SCL, CBCL plus AM or the CBQ, were taken for detailed evaluation in the 2nd stage. Mothers were the most common informants for the SCL (80%) and the CBCL (73%). The other respondents for the SCL were older siblings who were caregivers (10.5%), fathers (5.8%), grandparents (3.1%) and uncles/aunts (0.8%). For the older children, the other respondents were fathers (10.6%), grandparents (7.2%), siblings (4.9%), uncles/aunts (2.5%) and others (1.8%). Of the screening tools, teachers' ratings on the CBQ identified only 53 children as manifesting behaviour that caused concern and 145 children with academic problems. Of the 505 children, 256 were found to have a psychiatric disorder (51% of 1st stage positives) after the detailed evaluation, 196 children did not receive any diagnosis (39% of the 1st stage positives), while the evaluation was incomplete for 53 (10% of the 1st stage positives).

More than half of the incomplete assessments were due to lack of co-operation for the lengthy testing for specific learning disabilities. The SLD battery identified 149 (9.4%) children as having scholastic problems (rural: 11.7%; slum: 9.0%; urban: 6.6%). Of these children, 114 (7.2%) did not have any other Axis 1 disorder but did poorly only on the SLD Battery. These children could not be diagnosed as having SLD, as per ICD-10-DCR criteria, as most of them lacked adequate schooling. They were therefore excluded from calculation of the total prevalence rate.

The total prevalence rate for the sample aged 0-16 yr was 12.4 per cent (256 out of 2064 children). The total prevalence rates included the Axis IV diagnoses of epilepsy and breath holding spells, since these disorders often present in psychiatric settings. The total and disorder-specific prevalence rates for the 0-3 yr group are described in Table II.

The total prevalence rate for 0-3 yr old children was 13.8 per cent. The confidence interval (CI: 10.6-17%) indicated a 95 per cent certainty that it included the true population prevalence. There were no significant differences among the prevalence rates in the middle-class urban, slum and the rural areas (χ^sup 2^ = 1.001; df=2 ; P=0.606). The most common diagnoses were breath holding spells, pica, behaviour disorder NOS, expressive speech disorder and mental retardation (Table II).

The total prevalence rate for the 4-16 yr age group was found to be 12 per cent. The confidence interval (CI: 10.3-13.6%) indicated a 95 per cent certainty that it included the true population prevalence (Table III). There were no significant differences among the prevalence rates in the middle-class urban, slum and the rural areas (χ^sup 2^ = 1.788; df=2; P=0.409). The most common diagnoses were nonorganic enuresis, specific phobia, attention deficit hyperactivity disorder, stuttering and oppositional defiant disorder.

When the C-GAS scores were included, with disability defined as scores below 70, 5.3 per cent (CI: 4.2 - 6.5%) of the sample had significant functional impairment associated with psychiatric disturbances. Results indicated that only 45 per cent of the identified cases (84 of 189 children) had global impairment associated with their diagnoses.

A comparison of prevalence rates in the four age groups (0-3, 4-6, 7-11 and 12-16 yr) was carried out. The Bonferroni correction for 6 pair-wise comparisons at P

The results of the detailed 2nd stage evaluation of 10 per cent of the cases that were screened as negative in the first stage (N=159) indicated that 8.2 per cent were found to have an Axis 1 psychiatric disorder (18.4% in slum areas; 6.4% in rural areas; 0% in urban middle-class areas). Specific phobias, enuresis and behaviour disorder NOS were the only Axis 1 disorders identified.

The results indicated that the SCL had a sensitivity of 1.0 and specificity of 0.8. The CBCL and AM together had lower positive predictive power with a sensitivity of 0.74 and specificity of 0.62.

The results indicated that 162 (37.5%) of the 432 respondents to the FTN questionnaire perceived that their child had some kind of problem. Interestingly, 9.3 per cent of the total positive responses were given by families whose children only had some symptoms but did not receive any psychiatric diagnosis in the second stage of detailed evaluation.

The frequency of abnormal psychosocial situations was assessed using Parent Interview Schedule and compared between the two groups. The first group comprised of 189 children, aged 4-16 yr, who had a psychiatric diagnosis after the detailed evaluation stage (excluding the children who had only specific learning disability). The second group consisted of 78 children aged 4-16 yr from the 10 per cent negative group, who had no psychiatric morbidity. The results revealed that the following psychosocial situations were most common in children with a psychiatric diagnosis; living conditions that create a potentially hazardous psychosocial situation, physical abuse, parental mental disorder / deviance and intrafamilial discord. Analysis showed that physical abuse and the presence of mental disorder or deviance in the family were both significantly higher in the children with psychiatric morbidity (χ^sup 2^ = 11.67; df=1; P=0.0006 and χ^sup 2^ = 6.11; df=1; P=0.0130 respectively).

Discussion

The rate of psychiatric morbidity was 13.8 percent among children aged 0 to 3 yr, a largely neglected group among community-based studies. One study22 reported a prevalence rate of 7.8 per cent in Dutch children aged 2-3 yr. Two community-based Indian studies23,24 reported surprisingly low prevalence rates in the 0-4 yr age group: 0.6 and 0.0 per cent respectively. This strengthens the case for using developmentally appropriate assessment tools to identify childhood psychiatric disorders. The most common diagnoses were breath holding spells, pica, behaviour disorder NOS (primarily temper tantrums), expressive speech disorders and mental retardation. A clinic-based study25 also confirmed that epilepsy, mental retardation and breath holding spells were the most frequent reasons for referral among Indian children below 5 yr of age. These children are more likely to reach the paediatric and general health services instead of mental health services in our country. It is important to ensure early recognition, intervention or referral by these professionals. Developmental disorders like pica and breath holding spells are likely to be transient and long-term follow up would provide information about their developmental trajectory.

The prevalence rates among the 4 - 16 yr group, 12 per cent overall and 5.3 per cent with impairment criteria, were lower compared with findings from other community-based studies in Western countries. The MECA study26 reported a much higher rate of 20.9 per cent among 9-17 yr olds, using the same impairment criteria (C-GAS

Unexpectedly, the urban slum areas had the lowest total prevalence rates. One might speculate that low awareness of the importance of psychiatric disorders, increased tolerance for deviance, poor living conditions and the presence of multiple stressors could have combined to decrease the focus on children's problems. The results suggested that the screening tools were most sensitive in the urban middle class areas and the least sensitive in the urban slum areas. The finding of Axis 1 psychiatric disorders in 18.4 per cent of the '10 per cent negative' sample in urban slum areas, lends credence to the suggestion that the psychiatric morbidity may well be higher.

A sizeable number of children (9.4%) had scholastic problems. This is in line with an earlier school-based epidemiological study from north India30. The findings suggested that the needs of children with scholastic underachievement must be addressed, despite the lack of a psychiatric diagnosis.

The reasons for the absence of certain psychiatric conditions need to be explored. A larger sample would be needed to identify disorders like psychosis and autism, which have a low base rate of occurrence in the population. The present study might have underestimated psychiatric morbidity among adolescents, and particularly among adolescent girls whose vulnerability to emotional or internalizing disorders is well documented31. The low response rate of the DISC-C12 could be related to the lack of incentives or privacy, time taken for the interview, and adolescents' perceptions of the mental health profession. The complete absence of substance abuse in the present study could be linked to the low response rate for DISC-C12, youths' unwillingness to reveal such information and parents' ignorance about the abuse. The absence of conversion or dissociative disorders is puzzling, particularly since they are among the more common disorders seen at child psychiatric clinics in India32,33. It has been argued that since the expression of emotional distress is generally discouraged in the Indian cultural setting, physical illness may be the most acceptable avenue for help seeking or expression of psychological distress34. Clearly, culture-specific assessment instruments must be developed to aid identification of certain psychiatric disorders.

About 55 per cent of the children and adolescents with a psychiatric diagnosis had no global impairment (C-GAS

In contrast to earlier literature35, there were no significant gender differences in total prevalence rates. Certain disorder-specific prevalence rates might have been expected to reveal differences; for example, higher rates of conduct disorder among boys and depression among adolescent girls. However, the relatively small numbers of these disorders identified in the study made such comparisons difficult.

Physical abuse and the presence of parental psychopathology were significantly associated with psychiatric morbidity in children aged 4-16 yr. Though information about the correlates of specific psychiatric disorders was lacking, the results threw some light on potential psychosocial risk factors for psychiatric morbidity.

The limitations of the study included the need for a larger sample size to accurately assess prevalence in the 0-3 and 4-16 yr age groups separately. The utility of the CBCL and the modified cut-off scores, as a first-stage screening tool needs to be reconsidered, considering its low sensitivity in slum areas for certain psychiatric disorders. Teachers proved to be better informants for identifying children with academic difficulties but were almost unable to identify potential mental health problems on the CBQ15. Limitations also included the low response rate to the DISC-C, and the absence of adequate reliability and validity data for tools like the SCL and SLD battery.

The results of the study have implications for clinical training, practice and policy initiatives. The low level of awareness among families suggested that dissemination of accurate information about psychological problems and available help seeking avenues was imperative. There is a need to plan models of service delivery in both rural and urban areas and focus on the integration of mental health into general health care. Clearly, manpower and training issues for pharmacological and psychosocial interventions will be paramount in the planning of effective mental health services. Effective networking between mental health professionals, paediatricians, neurologists, speech therapists, community-based health services and professionals from the education sector, would be essential.

Further analysis can enhance the understanding of the patterns of co-morbidity, perceived treatment needs and psychosocial correlates. The area of scholastic difficulties also warrants further exploration. A longitudinal cohort study should be the next step in understanding the natural history of child and adolescent disorders.

Acknowledgment

The authors acknowledge the initiative and the financial assistance provided by the Indian Council of Medical Research, New Delhi. In particular we wish to acknowledge the contributions of Late Dr C.R. Ramachandran, Director Grade Scientist and Chief, Division of NCD, Dr Bela Shah, Senior Deputy-Director General and Dr Tripti Khanna, Senior Research Officer from the ICMR, New Delhi.

The authors thank the two consultants Dr Savita Malhotra, Professor, Department of Psychiatry, Postgraduate Institute of Medical Education & Research, Chandigarh and Dr Manju Mehta, Professor, Department of Psychiatry, All India Institute of Medical Sciences, New Delhi for their expert opinion and assistance during the study, and the research staff Ms. Veena Muthanna, Dr Manoranjan Hegde, Ms. S.S. Kusuma, Shri. K.S. Divesh, Ms. Thriveni, Shri. T.G. Srinivas, Shri. D. Shadaksharaiah, Dr S.R. Sujay Kumar, Dr Rajendra Kumar Katte and Dr Sudhir Hebbar for their contribution.

References

1. Sethi BB. Gupta SC, Kumar R. 300 urban families (A psychiatric survey). Indian J Psychiatry 1967; 9: 280-302.

2. Sethi BB, Gupta SC, Kumar R, Kumar P. A psychiatric survey of 500 rural families. Indian J Psychiatry 1972; 14: 183-96.

3. Verghese A, Beig A. Psychiatric disturbances in children: An epidemiological study. Indian J Med Res 1974; 62: 1538-42.

4. Nandi DN, Ajmany S, Ganguly H, Banerjee G, Boral GC, Ghosh A, et al. Psychiatric disorder in a rural community in West Bengal - an epidemiology study. Indian J Psychiatry 1975; 17: 87-9.

5. Lal N, Sethi BB. Estimate of mental ill health in children in an urban community. Indian J Paediat 1977; 4: 55-64.

6. Hackett R, Hackett. L, Bakta P, Cowers S. The prevalence and association of psychiatric disorder in children in Kerala, South India. J Child Psychol Psychiatry 1999; 40: 801-7.

7. Costello E, Tweed D. A review of recent empirical studies linking the prevalence of functional impairment with that of emotional and behavioural illness or disorder in children and adolescents. Washington, DC: Report to the Center for Mental Health Services; 1994.

8. Srinath S, Girimaji SC. Epidemiology of child and adolescent mental health problems and mental retardation. NIMHANS J 1999; 17: 355-66.

9. Achenbach TM. Manual for the Child Behaviour Checklist/2-3 and 1992 profile. 1992; Burlington, VT: University of Vermont, Department of Psychiatry.

10. Achenbach TM, Edelbrock CS. Manual for the Child Behaviour Checklist and Revised Child Behaviour Profile. 1983; Burlington, VT: University of Vermont, Department of Psychiatry.

11. Rutter M. A children's behaviour questionnaire for completion by teachers. Preliminary findings. J Child Psychol Psychiatry 1967; 8: 1-12.

12. National Institute of Mental Health. Diagnostic Interview Schedule for Children. 1991; Rockville, MD: National Institute of Mental Health.

13. World Health Organisation. The ICD-10 classification of mental and behavioural disorders: Diagnostic criteria for research. 1993; Geneva: World Health Organisation.

14. Rutter, M. Classification. In: Rutter M, Hersov L, editors. Child psychiatry. Oxford: Blackwell Scientific Publications; 1977 p. 359-84.

15. World Health Organisation. Psychosocial axis of the multiaxial classification of child and adolescent psychiatric disorders. Parent Interview Schedule. Draft for comments and field testing. Geneva: World Health Organization; 1990: NIMH/MND/90.13.

16. Malin AJ. Vineland Social Maturity Scale: Nagpur adaptation. 1971. Lucknow, Indian Psychological Corporation.

17. Doll EA. The manual of directions of the Vineland Social Maturity Scale. 1947. Minneapolis, Educational Test Bureau.

18. Kamat VV. A revision of the Binet scale for Indian children (Kanarese and Marathi speaking). Br J Edu Psychol 1934; 4: 296-309.

19. Kapur M, John A, Rozario J, Oommen A. NIMHANS Index for Specific Learning Disabilities 1992. Bangalore; Department of Clinical Psychology: National Institute of Mental Health and Neuro Sciences.

20. Endicott J, Spitzer RL, Fleiss JL, Cohen J. The Global Assessment Scale: A procedure for measuring the overall severity of psychiatric disturbance. Arch Gen Psychiatry 1977; 33: 766-71.

21. Shaffer D, Could MS, Brasic J, Ambrosini P, Fisher P, Bird H, et al. A Children's Global Assessment Scale (C-GAS). Arch Gen Psychiatry 1983; 40 : 1228-31.

22. Koot HM, Verhlust FC. Prevalence of problem behaviour in Dutch children aged 2-3. Acta Psychiatr Scand 1991; 367 Suppl : 1-37.

23. Sen B, Nandi DN, Mukherjee SP, Mishra DC, Banerjees G, Sarkar S. Psychiatric morbidity in an urban slum dwelling community. Indian J Psychiatry 1984; 26: 185-93.

24. Banerjee T, Mukherjee SP, Nandi DN, Banerjee G, Mukherjee A, Sen B, et al. Psychiatric morbidity in an urbanized tribal (Santhal) community: A field survey. Indian J Psychiatry 1986; 28: 243-8.

25. Chadda RK. Psychiatric morbidity in preschool children: A clinic based study. Indian J Pediatr 1995; 62: 77-81.

26. Shaffer D, Fisher P, Dulcan MK, Davies M, Piacentini J, Schwab-Stone ME, et al. The NIMH Diagnostic Interview Schedule for children Version 2.3 (DISC 2.3): Description, acceptability, prevalence rates and performance in the MECA study. J Am Acad Child Adolesc Psychiatry 1996; 35: 865-77.

27. Offord DR, Boyle MH, Szatmari P, Rae-Grant NI, Links PS. Cadman DT, et al. Ontario Child Health Study II: Six-month prevalence of disorder and rates of service utilization. Arch Gen Psychiatry 1987; 44: 826-31.

28. Bird HR, Canino G, Rubio-Stipec M, Gould MS, Ribera J. Sesman M. et al. Estimates of the prevalence of childhood maladjustment in a community survey in Puerto Rico: The use of combined measures. Arch Gen Psychiatry 1988; 45: 1120-6.

29. Roberts RE, Attkisson CC, Rosenblatt A. Prevalence of psychopathology among children and adolescents. Am J Psychiatry 1998; 155: 715-25.

30. Malhotra S, Kohli A, Arun P. Prevalence of psychiatric disorders in school children in Chandigarh, India. Indian J Med Res 2002; 116: 21-8.

31. Ollendick TH, King NJ. Diagnosis, assessment, and treatment of internalizing problems in children: the role of longitudinal data. J Consult Clin Psychol 1994; 62: 918-27.

32. Chadda RK, Saurabh. Patterns of psychiatric morbidity in children attending a general psychiatry unit. Indian J Pediatr 1994; 61: 281-5.

33. Sidana A, Bhatia MS, Choudhary S. Prevalence and pattern of psychiatric morbidity in children. Indian J Med Sci 1998; 52: 556-8.

34. Srinath S, Bharat S, Girimaji S, Seshadri S. Characteristics of a child inpatient population with hysteria in India. J Am Acad Child Adolsc Psychiatry 1993; 32: 822-5.

35. Zahn-Waxler C. Warriors and worriers: Gender and psychopathology Dev Psychopathol 1993; 5: 79-89.

Shoba Srinath, Satish Chandra Girimaji, G. Gururaj*, Shekhar Seshadri, D.K. Subbakrishna", Poornima Bhola+ & Narender Kumar++

Departments of Psychiatry, * Epidemiology & ** Biostatistics, National Institute of Mental Health & Neuro Sciences, Bangalore, + Department of Psychiatry, St. John's Medical College & Hospital, Bangalore ++ Indian Council of Medical Research, New Delhi, India

Received February 23, 2004

Reprint requests: Dr Shoba Srinath, Professor and Head, Department of Psychiatry Head, Child and Adolescent Psychiatry Services, National Institute of Mental Health & Neuro Sciences, Bangalore 560009, India

e-mail: shobas@nimhans.kar.nic.in

Copyright Indian Council of Medical Research Jul 2005
Provided by ProQuest Information and Learning Company. All rights Reserved