Review Article Indian J Med Res 126, September 2007, pp 183-192

Psychiatric epidemiology in India Suresh Bada Math, C. R. Chandrashekar & Dinesh Bhugra*

Department of Psychiatry, National Institute of Mental Health & Neuro Sciences (Deemed University) Bangalore, Karnataka, India & *Health Services Research Department, Institute of Psychiatry, London, UK

Received October 18, 2006

Epidemiological studies report prevalence rates for psychiatric disorders from 9.5 to 370/1000 populations in India. This review critically evaluates the prevalence rate of mental disorders as reported in Indian epidemiological studies. Extensive search of PubMed, NeuroMed and MEDLARS using search terms “psychiatry” and “epidemiology” was done. Manual search of literature was also done. Retrieved articles were systematically selected using inclusion and exclusion criteria. Only sixteen prevalence studies fulfilled the study criteria. Most of the epidemiological studies done in India neglected anxiety disorders, substance dependence disorders, co-morbidity and dual diagnosis. The use of poor sensitive screening instruments, single informant and systematic underreporting has added to the discrepancy in the prevalence rate. The prevalence of mental disorders reported in epidemiological surveys can be considered lower estimates rather than accurate reflections of the true prevalence in the population. Researchers have focused on broad non-specific, non-modifiable risk factors, such as age, gender and social class. Future research focused on the general population, longitudinal (prospective), multi-centre, co-morbid studies, assessment of disability, functioning, family burden and quality of life studies involving a clinical service providing approach, is required.

Key words Epidemiology - mental health - prevalence - psychiatric disorders

prevalence rates, ranging from 9.54 to 370 per 1000 population5. These discrepancies are not specific to Indian studies but are also seen in international studies like the Epidemiological Catchment Area Program and the National Comorbidity Survey6,7. This discrepancy will impact planning, funding and health care delivery. Providing accurate data about the prevalence of mental disorders in the community would help to justify the allocation of scarce resources and planning of health services.

Psychiatric epidemiology has gone through various stages of growth over the past five decades in India, starting from the first psychiatric epidemiological study by K.C. Dube1, in 1961 at Agra, to the development of tools like the Present Status Examination (PSE) 2 and the Indian Psychiatric Survey Schedule (IPSS)3. A major advance in psychiatric epidemiology is the development of reliable and valid diagnostic interviews. Many epidemiological studies conducted in India on mental and behavioural disorders report varying 183

184

INDIAN J MED RES, SEPTEMBER 2007

Psychiatric epidemiology lags behind other branches of epidemiology due to difficulties encountered in conceptualizing, diagnosing, defining a case, sampling, selecting an instrument, lack of resources and stigma8. The descriptive epidemiological studies have undergone unprecedented growth in India, but at the same time advances with respect to costeffective, analytical and prospective experimental epidemiological studies have been minimal8. A major challenge for psychiatric epidemiologists is to increase the relevance of their research with regard to their counterparts in preventive psychiatry and to the policy makers8. Researchers have expressed reservations about the comparison of various epidemiological studies because of methodological differences. However, at present one has to rely on available studies to generalise the findings. Hence this article attempts to critically evaluate the (overall) prevalence rate of psychiatric disorders as reported in epidemiological studies from India. This review also attempts to answer the following questions (i) What are the reasons for wide variation in the prevalence rate in India; (ii) Is the prevalence rate of psychiatric disorders stable or changing? (iii) What is the cost of treating psychiatric patients?; and (iv) What should be the focus of future epidemiological studies?. Methodology Extensive search was done of NeuroMed (19821997) and MEDLARS for published Indian psychiatric epidemiological studies. The search terms included “psychiatry” “prevalence”, “community”, and “epidemiology”. Attempts were also made to retrieve Indian epidemiological studies published in international journals through PubMed using search terms “psychiatry” and “epidemiology”. Extensive manual search of issues of the Indian Journal of Psychiatry, NIMHANS Journal (before 1982 and after 1997) and Indian Journal of Medical Research was also done. Cross-references of the psychiatric prevalence studies were also reviewed. Inclusion criteria were: general population studies either urban, rural or mixed from India; community study design involving door-to-door/house-to-house enquiry of families and random sampling of families; inclusion of all psychiatric disorders or at least priority/ major mental disorders. (Various researchers have used this term to assess schizophrenia, manic depressive psychosis, organic psychosis, epilepsy and mental retardation in the community); and covering all age groups.

Studies were excluded if those were: specific syndrome/illness/disorder studies; hospital or clinic based studies; and particular age group studies. Only data published in scientific journals were considered because of logistic reasons. This is one of the main limitations of this review. Only 16 prevalence studies (Table I) fulfilled the inclusion criteria. However, an attempt was made to summarize other studies like high-risk/special population, cost-effective analysis, incidence, follow up and meta analysis separately to have a comprehensive picture of epidemiological studies till date. Evaluation of prevalence in each study was done on the following parameters; definition of a “case”, screening and diagnostic instrument used, diagnostic break-up assessed, methods of arriving at the diagnosis, sampling method, location of study, general population/ high risk population study and number of informants used to collect data. What are the reasons for wide variations noted in the prevalence rates of psychiatric epidemiological studies? Defining a case: Inherent limitations in defining clinical cases in epidemiological studies are: (i) the definition of mental disorder fails to provide a clear boundary between psychopathology and normality; (ii) the concepts “clinical significance” and “medical necessity” are difficult to operationalize and to assess reliably; and (iii) lay interviewers do not have the experience necessary to judge clinical significance6,23. Defining a case was one of the factors which contributed to the huge variation in prevalence rate. If the threshold for defining a case is very low then prevalence rate will be very high5,24. Earlier studies had difficulty in defining a case, when clear-cut diagnostic criteria were not available, but Indian researchers used a broader diagnostic classification such as psycho-neurotic, schizophrenia, epilepsy, depression, mental deficiency, miscellaneous group9-11,25. Researchers have also used case definitions as per WHO technical report series no. 185, 196015,26,27, WHO International Classification of Diseases (ICD) (1965 R), Diagnostic and Statistical Manual of Mental Disorders (DSM) of American Psychiatric Association (APA)196814, and 1965 ICD12 for diagnosis. The ‘presence of a disorder’ implies ‘need for treatment’ in a clinical population; however, in epidemiological studies these differ. For example, in

MATH et al: PSYCHIATRIC EPIDEMIOLOGY IN INDIA

Indian culture, people suffering from social phobia are generally considered shy and it is not considered that they require treatment. To determine the presence of a disorder, the need for treatment, distress, dysfunction, disability and availability of resources need to be established6,28. Indian researchers made attempts to overcome this difficulty by assessing the positive cases on screening by qualified psychiatrist(s) for final diagnosis by using their clinical judgment. However, minor mental disorders which could not be tapped on screening were missed. Considering the difficulty of defining a case and diagnosis, Indian researchers made all attempts to overcome them by assessing all positive cases on screening. Screening instruments: The majority of the epidemiological studies considered two-phase sampling for assessing prevalence1,4,13,14,17. The first step was to screen the population for mental illness using a screening instrument and the next step was to confirm diagnosis. Screening instruments can be self-reported, observer-rated, or based on information from an

185

informant. Self-reported ratings are highly dependent on the co-operation of patients and their ability to understand either written or verbal instructions. Observer-based ratings can be time-consuming, and lay interviewers can misinterpret the severity and impact of the illness7,23. Hence the use of self-reported and observer-rated tools was not possible in our population. The most common method used by Indian researchers was based on information from one or more informants. Screening scales or instruments included exploratory questions, and symptoms or diagnostic checklists for screening the community for possible mental illness, with positive cases referred for more extensive evaluation. Psychosis becomes less and less common at greater distances from the hospital patient, and in the community the epidemiologist is faced with large numbers of respondents who present with fewer, minor and non-specific symptoms 29 . Indian epidemiological studies were largely inadequate to tap the most non-psychotic disorders like panic disorder, social phobia, obsessive compulsive disorder, sexual

Table I. Prevalence of psychiatric morbidity Investigator

Year

4

1964 1967 1970 1971 1972 1973 1974 1975 1975 1979 1980 1985 1986 1993 1995 2001

Surya Sethi et al9 Dube1 Elangar et al10 Sethi et al11 Verghese et al12 Sethi et al13 Thacore et al14 Nandi et al15 Nandi et al16 Shah et al17 Mehta et al18 Sachdeva et al19 Premarajan et al20 Shaji et al21 Sharma et al22

Centre Pondicherry Lucknow Agra Hoogly Lucknow Vellore Lucknow Lucknow West Bengal West Bengal Ahmedabad Vellore Faridkot Pondichery Erankulam Goa

Location Sampling U U M R R U R U R R U R R U R M

H-H H-H H-H H-H H-H SRS 3SPS H-H H-H H-H H-H S-S H-H RS H-H SRS

Tool MHSQ(P) QAPF DCP CHM & DCP(2) CHQ & CHM MHIS & DCP as per ICD (1965) PSQ & DCP as per DSM-II (1968) PHQ & DCP HS, QS & CRS as per ICD (1965 R) HS, SESS, CDS & CRS MHSQ & DCP IPSS & DCP HS, SESS & CDS IPSS & DCP as per ICD-9R IPSS, SESS, CRS & DCP, ICD-10 RPES & DCP as per ICD-9

Population

Prevalence/1000

2731 1733 29.468 1393 2691 1887 4481 1977 1060 3718 2712 5941 1989 1115 5284 4022

9.5 72.7 18 27 39.4 66.5 67.0 81.6 102.8 102 47.2 14.5 22.12 99.4 14.57 60.2

U- urban; R- rural; H-H- house to house survey; S-S – systematic sampling; SRS – stratified random sampling 3SPS- 3 stage probability sampling; RS- random sampling, ICD- international classification of diseases DSM-II- diagnostic and statistical manual of mental disorders. Tools: MHSQ= Mental health Screening Questionnaire, QAPF=Questionnaire for the assessment of psychiatric state of the family, DCP = Diagnosis confirmed by a psychiatrist(s) CHM = Case history method CHQ = Case history questionnaire IPSS=Indian Psychiatric Survey Schedule SFQ=Social functioning questionnaire MHIS=Mental health item sheet, PSQ=Psychiatric screening questionnaire PHQ=Psychiatric health questionnaire, HS=Household schedule QS=Questionnaire Schedule, CRS=Case record schedule, SESS=Socio-economic status schedule

CDS = Case detection schedule, RPES= Rapid psychiatric examination schedule

186

INDIAN J MED RES, SEPTEMBER 2007

dysfunctions, substance use etc., in the community. Earlier studies prepared their own screening questionnaire, which was applied to the entire population to be studied without testing their validity for high-risk populations such as the children, the elderly and substance users, thereby missing mental disorders during the initial screening4,9,10,14. This was a major drawback of these studies, which might have led to under-reporting of mental disorders. Issac and Kapur30, in their study on cost-effective analysis of three different methods of psychiatric case finding in the general population, answered the difference in prevalence reported by various researchers. Comparison of cases in various diagnostic categories identified through screening by using questionnaires and confirming the diagnosis by a psychiatrist (Method-1) missed out on depression, possession disorder, neurotic disorders, somatisation disorder and substance dependence during the initial screening. This was later picked up by using the Indian Psychiatric Survey Schedule (IPSS)3, schedule in the general population (Method-2). Major mental disorders that were identified were almost similar with both methods. Cost-effective analysis of the above two methods revealed that using the IPSS was 10 times more expensive. This study clearly demonstrated the significant difference in prevalence rates: 76/1000 in method-1 and 344/1000 on method-2. Determining which screening instrument should be selected must always follow an analysis of the underlying purpose of whether this instrument can be applied to the chosen sample and to all age groups31. Sensitivity rather than the specificity of the screening instrument is very important in prevalence studies to avoid false negatives. However, further essential criteria in determining choice of the scale are, cultural relevance, validity, time available and the person who

is to carry out the rating. Instruments such as the Self Reporting Questionnaire (SRQ) and General Health Questionnaire (GHQ)31 have been used successfully for screening purposes in epidemiological studies32. Informants: A family informant is one who reports about his/her illness as well as on other biological relatives. It takes 30 to 90 min to administer the screening instrument, depending on the size of the family and the presence of family illness. There is agreement that compared with direct interviews, the family history by one informant method underreports illness and using two or more informants can partially correct this problem33. Utilizing data from multiple informants provides a more valid assessment of the disorder in studies related to mental health problems34,35. However, in certain epidemiological studies, the investigators interviewed only the head of the family or the housewife or any other responsible family member for data collection9,11,18,19,21,22. This will lead to responder bias since the chances of under-reporting symptoms of psychiatric disorders are very high especially for minor mental disorders. But whenever researchers assessed all the family members or used multiple informants to collect data, the prevalence rate was found to be twice that of the single informant method14-16,20. This is one of the main reasons that can account for discrepancies in the psychiatric prevalence rate reported in India. Sampling methods: Another important methodological issue is the sampling approach. In order to properly estimate the prevalence of mental disorders, it is essential to choose the sample in a manner that will accurately reflect the general population. Eighty per cent of researchers had used the house-to-house survey method9-11,14. In the design of studies of people with mental disorders, it is necessary to consider the likelihood of not being able to access people who are hospitalized due to illness, homeless people, wandering

Table II. Special/high-risk population studies Investigator group

Year

Carstairs & Kapur5 Nandi et al26 Nandi et al39 Nandi et al37 Nandi et al43 Sen et al38 Banerjee et al41 Nandi et al40

1973 1977 1978 1980 1980 1984 1986 1992

Centre Kota West West West West West West West

Bengal Bengal Bengal Bengal Bengal Bengal Bengal

Abbrevation- As given in the foot note of Table I

Location R R R R M U U U

Sampling H-H H-H H-H RS H-H H-H H-H H-H

Tool IPSS & SFQ HS, QS & CDS HS, QS & CDS HS, SESS, CDS HS, SESS, CDS HS, SESS, CDS HS, SESS, CDS HS, SESS, CDS

& & & & &

CRS CRS CRS CRS CRS

Population

Prevalence/ 1000

1233 2918 1259 4053 1862 2168 771 1424

370 58.2 47.6 50.3 129.9 48.7 51.9 47.75

MATH et al: PSYCHIATRIC EPIDEMIOLOGY IN INDIA

mentally ill patients and people who are not available for other reasons like occupation 36 . Only a few studies12,13,20,37 have used the random sampling method. Researchers have sampled special population like urban slum dwellers38, uprooted communities39 and tribal communities 40, to study the effect of stress5, urbanization 41 and to compare prevalence across cultures26,37. Studies done on high risk populations yielded a high prevalence rate5. On plotting the studies on the Indian map, studies were found to be concentrated only in certain areas like West Bengal (40%) and UP (10%), which led to difficulty in generalizing the findings. The extent to which findings from one cultural unit can be generalised to other populations is still open to debate. Can we generalise findings from one State to another or from one ethnic group to another within the same country? Multi-site studies are the answer42. Hence we decided to highlight the studies on various special populations (Table II). Variations seen in prevalence (47.6 to 370) were huge on comparing with Table I (general population) (9.5 to 102.8). Disability: Mental disorders often result in profound burden of illness and disability. The World Health Organization has determined that mental illness is one of the largest contributors to disability worldwide45,46. The disability-adjusted life years (DALY) measure combines burden from premature mortality with that from living with disability, and provides a comprehensive assessment of the burden of illness. When the loss of life and ability are estimated using DALY, more than 10 per cent of the total burden of human disease and loss are attributed to mental disorders. The overall DALYs burden for mental disorders is projected to increase to 15 per cent by the year 2020 and this increase is proportionately larger than that for cardiovascular diseases46. Assessment of disability in social and occupational functioning is necessary in epidemiological studies24. Caregiver’s burden: The family, as a unit of the society forms an integral part of the care giving system for persons with mental illness, especially in our country. The demands of being involved in the care of a seriously mentally ill patient have serious impact on the caregiver’s health. The costs that families incur in terms of economic hardships, social isolation and psychological strain, are referred to as the family burden47,48. It is our duty to treat the patient and also to evaluate the mental state of the caregivers, since in our community family members are the caregivers. Research has clearly demonstrated the burden of mental

187

disorders on individuals, families, communities and health services globally. If the impact of mental illness on the economy is assessed by combining all the costs like health care cost, loss of wages and patient/family costs, then the magnitude of the impact of mental illness becomes evident. Psychiatric illness has disabling consequences49. The contribution of Indian researchers towards the burden assessment schedule (BAS) is noteworthy 50 , and can be applied in future epidemiological studies. Systematic under-reporting: A survey of an urban community in southern India, found that one-third of people with schizophrenia had never accessed any treatment resources 51 . Even after the diseased individuals and their families were offered treatment, a third of them remained untreated52. The important question remains as mental health professionals, how can we reduce this stigma against mental illness in the mind of the public? The question is important because how the public perceives mental illness will decide how our services will be utilized by the society53. Mental disorders are highly stigmatized conditions that many people want to keep private because of their embarrassment or fear of discrimination54. In the case of patients seeking professional treatment, there is reason to believe that self-reports will be complete and honest. However, this is not the case in epidemiological surveys, where patients do not seek help. It is not surprising then that concerns have been raised that under-reporting is a very serious problem in surveys of this sort55. Even when clinicians carry out interviews, methodological research has shown that some respondents are less liable to disclose embarrassing information when they are aware that their interviewer is a mental health professional 56. The problem of systematic under-reporting continues to be a major challenge for the future of psychiatric epidemiology in India. Is the prevalence rate of psychiatric disorders stable or changing? Does globalization (liberalization) play any role on the mental health of the individuals? Does changing family structure (joint family to nuclear) influence the psychiatric morbidity? Unfortunately we do not have any long-term epidemiological studies to answer these questions. A 10 yr follow up study reported that there was not much change in the prevalence rates over a decade

188

INDIAN J MED RES, SEPTEMBER 2007 Table III. Follow-up studies on prevalence of psychiatric disorders in India

Investigator

Year

Centre

Location Sampling

Nandi et al57 (1972-1982) Nandi et al58 (1972-1992)

1986

West Bengal

R

2000

West Bengal

R

Tool

Year Population

Prevalence/ 1000

H-H

HS, SESS, CDS & CRS

1972 1982

1060 1539

84.9 81.9

H-H

HS, SESS, CDS, CRS & DCP

1972 1992

2183 3488

116.8 105.2

Abbrevation- As given in the foot note of Table I

(prevalence rate in 1972 was 84.9 and in 1982 it was 81.9)57. This was further reinforced by the 20 yr follow up study, where prevalence rate in 1972 was 116.8 and in 1992 it was 105.258 (Table III). These two studies are milestones in psychiatric epidemiology looking at the same population cross-sectionally at two points. Though the prevalence rate did not change, the morbidity pattern (affective illness had increased) changed significantly. However, researchers and policy makers should exercise caution before generalizing the findings to whole country. Reason being follow up studies had small sample size and were on rural population. Adding to the existing unmet needs of mental health, changing health scenario in India has led to imminent epidemic of non communicable diseases along with the unmet agenda of controlling infectious diseases59. Life style disease (like obesity and substance use) and stress can increase the risk of several leading causes of death like cardiovascular diseases, cancer, diabetes, metabolic and endocrine disorders among others. In the management of non communicable and life style diseases, mental health component plays a crucial role. Psychiatric co-morbidity is common in non communicable diseases and chronic medical conditions. This co-morbidity further adds to the prevalence of mental health disorders. What does the meta-analysis of epidemiological studies of psychiatric disorders report? A meta-analysis of 13 epidemiological studies consisting of 33,572 persons, who met the following criteria; door-to-door survey, all age groups included and prevalence rate for urban and rural being available60, reported a total morbidity of 58.2 per 1000. Though meta-analysis has its own limitations, this was the first attempt to analyze the epidemiological studies. Another meta-analysis of 15 epidemiological studies reported a total morbidity of 73 per 100061. This study included 15 studies on the basis of having similar design which were conducted in three phase/steps. The difference noted in the two available meta-analyses was due to

the methodology of selecting the papers for the analysis. Earlier study had a criterion of door to door survey, whereas the latter had a criteria of studies being conducted in three phases/steps. What do the international epidemiological studies report? NIMH-Epidemiological Catchment Area study62 of US reported psychiatric morbidity as follows: one month prevalence of 151/1000 population; life time prevalence of 322/1000 population; one year incidence of 60/1000 population. National Co-morbidity Study of US63 reported 12 months prevalence of 277/1000 population and life time prevalence of 487/1000 population. Compared to these studies it is clear that prevalence rates and incidence rates reported in India are very low. Possible reasons for this difference may be (i) Indian epidemiological studies did not adequately measure psychiatric morbidity; (ii) The psychiatric prevalence rates are truly low in India; and (iii) Combination of both the above factors. Available evidence supports the first possibility of the under-reporting by Indian epidemiological studies because of poor sensitivity of the screening instrument, high-risk populations (children and adolescents, elderly) were not assessed, neurotic disorders and substance use disorders were not assessed adequately, stigma and single informant method would have lead to under-reporting. However, the possibility of genuine low prevalence of psychiatric disorders in Indian population cannot be disregarded because of low rates of substance use in general population compared to western countries and good outcome of psychiatric disorders due to various factors like religious, cultural, social and family support. Answer to this question can be obtained through multi-centred studies involving various countries with prospective design using single/ common research protocol. What is the economic cost of treating psychiatric patients? Macro-economic commission report of 2005 64 considered prevalence rate of 65/1000 population

MATH et al: PSYCHIATRIC EPIDEMIOLOGY IN INDIA

(average of two meta-analysis-60, 61) and projected the prevalence rate for next two decades. However, if we consider prevalence of individual mental disorder and add, the overall prevalence rate will be approximately 115-130/1000 population. Even if we consider 65/1000 population as prevalence rate, then 6.5 crores populations require professional help. If each patient requires INR 300 per month for the treatment, the total cost required per month will be 1,950 crores INR and per year will be 23,400 crores INR. If we do not address this issue by investing in mental health, the indirect costs in terms of loss of wages by the patient, disability and then families facing social isolation, burden, stigma and psychological strain will be enormous. Discussion In majority of the studies the researchers were able to assess major mental disorders accurately, but minor mental disorders were not assessed adequately. Indian researchers made attempts to overcome the difficulty of diagnosis by assessing the positive cases on screening by qualified psychiatrist(s) for final diagnosis by using their clinical judgment and available diagnostic guidelines at that time, thus avoiding clinician bias in diagnosis. Prevalence rates of psychiatric disorders not only vary across the populations, but also in the same population from time to time. In addition, weak agreement at the level of diagnosis and systematic under-reporting continue to threaten the credibility of estimates of prevalence65. Prevalence rates can vary markedly with screening instruments used, changes in the assessment questions used in community surveys, minor changes in diagnostic criteria (thresholds for defining mental disorders), number of informants, and sampling methodology24. With the exception of one study1, all the past epidemiological studies have surveyed a population less than 6000. This raises a query whether the findings can

189

be generalized to even one State in a country like India, which is well known for its geographical, linguistic and ethnic diversity. Mental health care priorities need to be shifted from psychotic disorders to common mental disorders like depression, anxiety disorders, somatoform disorder, etc., which are also associated with high disability in all measures49. The available evidence suggests that though there has been no increase in prevalence rate of psychiatric disorders in the past few decades in India, the changing health scenario has led to imminent epidemic of non communicable diseases59 and psychiatric co-morbidity being common in non communicable diseases and chronic medical conditions provides indirect evidence of rise in psychiatric prevalence. Scope of the future epidemiological studies: Most of the past epidemiological studies were descriptive in nature. Thus the future epidemiological studies should be more analytical and experimental. High-risk individuals (survivors of disaster, people suffering from chronic general medical conditions, the destitute and homeless) with modifiable risk factors need to be identified and included in the studies. The effects of modifying risk factors on prevalence rates have to be explored. Studies to document the impact of organizing mental health services and preventive strategies are required. There is a need to study the effect of global socio-economic policies like liberalization and privatization, the fast growth of information technology, life style changes and its impact on the mental health of individuals. The effectiveness of various techniques and programmes of stress management and life skill implementation on individuals also need to be included as a part of epidemiological studies. These is a need to determine the quality of life, co-morbidity, disability and burden of various mental disorders66. Longitudinal/prospective (experimental) epidemiological studies need to be carried out in which the natural course of all the disorders in the community can be studied and modifiable risk factors are identified

Table IV. Incidence studies done in India Investigator

Year

Centre

Location Sample

Nandi et al26 (1972-1973) Nandi et al27 (1972-1973)

1976

West Bengal

R

1978

West Bengal

R

Abbrevation- As given in the foot note of Table I

Tool

Year

H-H

HS, QS & CDS

H-H

HS, CDS & CRS

1972 1973 1972 1973

Population Prevalence/ Incidence/ 1000 1000 1060 1078 2230 2250

102.1 107.6 110.3 108.4

17.6 16

190

INDIAN J MED RES, SEPTEMBER 2007

and targeted for interventions. It is also vital to study the factors (barriers) affecting better service use, the role of culture and religion in help-seeking behaviours and modifying the identified factors, which may help in better delivery of mental health care at the community level. Mental health care priority also requires to be shifted from mental hospitals to primary health centres59. Systems need to be set up to carry out small area estimations for needs assessment and programme planning. Remaining challenges include the refinement of diagnostic categories and criteria and using accurate assessment tools for studying disorders of children, adolescents and the elderly in community surveys8. By applying valid instruments and proper sampling techniques, generalization of the results can be achieved. Systematic under-reporting in epidemiological surveys of mental disorders must be evaluated so that the accurate prevalence can be assessed. The translation of these estimates from the community to policy, however, is exactly what most critics challenge67. As only two incidence studies 25,27 are available (Table IV), there is a need to carry out more incidence studies in representative populations. Prevalence studies need to be repeated in a given area to evaluate the effectiveness of intervention programmes like the District Mental Health Programme, school or college mental health programmes, life skills programmes and other preventive programmes68. Conclusions Considering the facts of systematic underreporting, collecting data from single informant, use of low sensitivity screening instruments, poor sampling methods, assessing only priority mental disorders and changing health scenario indicates that the prevalence of mental disorders as reported in Indian epidemiological surveys can be considered lower estimates rather than accurate reflections of the true prevalence in the population. This should be kept in mind especially for policy making and for mental health care planning. Mental health care priorities need to be shifted from psychotic disorders to common mental disorders and from mental hospitals to primary health centres. Future research needs to focus on the general population, longitudinal (prospective), multi-centre, co-morbid studies, assessment of disability, functioning, family burden and quality of life studies involving a clinical service providing approach.

Acknowledgment Authors thank Dr N. Girish, Department of Epidemiology, National Institute of Mental Health and Neuro Sciences (Deemed University), Bangalore, for his valuable suggestions and inputs and Dr Maria Christine Nirmala for her valuable comments on the manuscript.

References 1.

Dube KC. A Study of prevalence and biosocial variables in mental illness in rural and urban community in Uttar Pradesh, India. Acta Psychiatr Scand 1970; 46 : 327-59.

2.

Wing JK, Cooper JE, Sartorius N. The Measurement and Classification of Psychiatric Symptoms. Cambridge: Cambridge University Press; 1974.

3.

Kapur RL, Kapur M, Carstairs GM. Indian Psychiatric survey schedule. Soc Psychiatry 1974; 9 : 71-6.

4.

Surya NC. Mental morbidity in Pondicherry. Transaction-4, Bangalore: All India Institute of Mental Health; 1964.

5.

Carstairs GM, Kapur RL. The Great University of Kota, London: The Hogarth Press; 1976.

6.

Regier DA, Kaelber CT, Rae DS, Farmer ME, Knauper B, Kessler RC, et al. Limitations of diagnostic criteria and assessment instruments for mental disorders: implications for research and policy. Arch Gen Psychiatry 1998; 55 : 109-15.

7.

Murphy JM, Monson RR, Laird NM, Sobol AM, Leighton AH. A comparison of diagnostic interviews for depression in the Stirling County study: challenges for psychiatric epidemiology. Arch Gen Psychiatry 2000; 57 : 230-6.

8.

Kessler RC. Psychiatric epidemiology: selected recent advances and future directions. Bull World Health Organ 2000; 78 : 464-74.

9.

Sethi BB, Gupta SC, Rajkumar S. Three hundred urban families - a psychiatric study. Indian J Psychiatry 1967; 9 : 280-302.

10. Elnager MN, Maitra P, Rao MN. Mental health in an Indian rural community. Br J Psychiatry 1971; 118 : 499-503. 11. Sethi BB, Gupta SC, Mahendru RK, Kumari P. A psychiatric survey of 500 rural families. Indian J Psychiatry 1972; 14 : 183-96. 12. Verghese A, Beig A, Senseman LA, Sundar Rao PSS, Benjamin V. A social and psychiatric study of a representative group of families in Vellore town. Indian J Med Res 1973; 61 : 608-20. 13. Sethi BB, Gupta SC, Mahendru RK, Kumari, P. Mental Health and urban life: A study of 850 families. Br J Psychiatry 1974; 124 : 243-6. 14. Thacore VR, Gupta SC, Suriya M. Psychiatric Morbidity in North Indian Community. Br J Psychiatry 1975; 126 : 364-9. 15. Nandi DN Ajmany S, Ganguly H, Banerjee G, Boral GC, Ghosh A, et al. Psychiatric disorders in a rural community in West Bengal - an epidemiological study. Indian J Psychiatry 1975; 17 : 87-99. 16. Nandi DN, Ajmany S, Sachdev S, Ganguli H, Banerjee G, Boral GC, Ghosh A, et al. Socio-economic status and prevalence of mental disorders in certain rural communities in India. Acta Psychiatr Scand 1979; 59 : 276-93.

MATH et al: PSYCHIATRIC EPIDEMIOLOGY IN INDIA 17. Shah A V, Goswami U A, Maniar R C, Hariwala DC, Sinha B K. Prevalence of psychiatric disorders in Ahmedabad: an epidemiological study. Indian J Psychiatry 1980; 22 : 384-8. 18. Mehta P, Joseph A, Verghese A. An epidemiological study of psychiatric disorders in a rural area in Tamil Nadu. Indian J Psychiatry 1985; 27 : 153-8. 19. Sachdeva JS, Singh S, Sidhu BS, Goyal RKD, Singh J. An epidemiological study of psychiatric disorders in rural Faridkot (Punjab). Indian J Psychiatry 1986; 28 : 317-23. 20. Premrajan KC, Danabalan M, Chandrasekhar R, Srinivasa DK. Prevalence of psychiatric morbidity in an urban community of Pondicherry. Indian J Psychiatry 1993; 35 : 99-102. 21. Shaji S, Verghese A, Promodu K, George B, Shibu VP. Prevalence of priority psychiatric disorders in a rural area of Kerala. Indian J Psychiatry 1995; 37 : 91-6. 22. Sharma S, Singh MM. Prevalence of mental disorders: An epidemiological study In Goa. Indian J Psychiatry 2001; 43 : 118-26. 23. Frances A. Problems in defining clinical significance in epidemiological studies. Arch Gen Psychaitry 1998; 55 : 119. 24. Spitzer RL. Diagnosis and need for treatment are not the same. Arch Gen Psychaitry 1998; 55 : 120. 25. Nandi DN, Ajmany S, Ganguli H, Banerjee G, Boral GC, Sarkar S. The incidence of mental disorders in one year community in West Bengal. Indian J Psychiatry 1976; 18 : 79-87. 26. Nandi DN, Mukherjee SP, Boral GC, Banerjee G, Ghosh A, Ajmany S, et al. Prevalence of psychiatric morbidity in two tribal communities in certain villages of West Bengal - a cross cultural study. Indian J Psychiatry 1977; 19 : 2-12. 27. Nandi DN, Banerjee G, Ganguli H, Ajmany S, Sachdev S, Boral GC et al. The Natural history of mental disorders in a rural community - a longitudinal field - survey. Indian J Psychiatry 1978; 21 : 390-6.

191

36. Jenkins R, Bebbington P, Brugha T, Farrell M, Gill B, Lewis G, et al. The National Psychiatric Morbidity Surveys of Great Britain: Strategy and methods. Psychol Med 1997; 27 : 76574. 37. Nandi DN, Mukherjee SP, Boral GC, Banerjee G, Ghosh A, Sarkar S, et al. Socioeconomic status and mental morbidity in certain tribes and castes in India- a cross cultural study. Br J Psychiatry 1980; 136 : 73-85. 38. Sen B, Nandi DN, Mukherjee SP, Mishra DC, Banerjee G, Sarkar S. Psychiatric morbidity in an urban slum-dwelling community. Indian J Psychiatry 1984; 26 : 185-93. 39. Nandi DN, Mukherjee SP, Banerjee G, Boral GC, Ghosh A, Sarkar S, et al. Psychiatric morbidity in a uprooted community in rural West Bengal. Indian J Psychiatry 1978; 20 : 137-42. 40. Nandi DN, Banerjee G, Chowdhury AN, Banerjee G, Boral GC, Sen B. Urbanization and mental morbidity in certain tribal communities in West Bengal. Indian J Psychiatry 1992; 34 : 334-9. 41. Banerjee T, Mukherjee SP, Nandi DN, Banergee G, Mukerjee A, Sen B, et al. Psychiatric morbidity in an urbanised tribal (Santal) community - A field-survey. Indian J Psychiatry 1986; 28 : 243-8. 42. Ommeren MV. Validity issues in transcultural epidemiology, Br J Psychiatry 2003; 182 : 376-8. 43. Nandi DN, Das NN, Chaudhuri A, Banerjee G, Datta P, Ghosh A, et al. Mental morbidity and urban life- an epidemiological study. Indian J Psychiatry 1980; 22 : 324-30. 44. Greenberg P, Stiglin L, Finkelstain S. Berndt E. Depression: A neglected major illness. J Clin Psychiatr 1993; 54 : 419. 45. World Bank. World Development Report. Investing in health. New York: Oxford University Press; 1993 p. 213.

28. Pincus HA. Zarin DA, First M. Clinical significance and DSMIV. Arch Gen Psychiatry 1998; 55 : 1145.

46. Murray C, Lopez A. The global burden of disease vol. 1. A comprehensive assessment of mortality and disability from diseases, injuries and risk factors in 1990, and projected to 2020. Cambridge, MA: Harvard University Press; 1996.

29. Williams P, Tarnopolsky A, Hand D. Case definition and case identification in psychiatric epidemiology: review and assessment. Psychol Med 1980; 10 : 101-14.

47. Pai S, Kapur RL. The burden on the family of a psychiatric patient: Development of an interview schedule. Br J Psychiatry 1981; 138 : 332-5.

30. Isaac MK, Kapur RL. A cost effectiveness analysis of three different methods of psychiatric case finding in the general population. Br J Psychiatry 1980; 137 : 540-7.

48. Schene A, Van Wijngaarden B, Koeter M. Family caregiving in schizophrenia: Domains and distress. Schizophr Bull 1998; 24 : 609-18.

31. Goldberg DP, Williams P. A user’s guide to the general health questionnaire. Windsor: NFER-Nelson; 1998. 32. Chandrashekar CR, Isaac MK. Development of psychiatric epidemiology in India. NIMHANS J 1999; 17 : 297-306.

49. Patel V, Pereira J, Coutinho L, Fernandes R, Fernandes, J, Mann A. Poverty, psychological disorder and disability in primary care attenders in Goa, India. Br J Psychiatry 1998; 172 : 533-6.

33. Romitti PA, Burns TL. Feasibility of collecting disease reports from relatives for genetic epidemiologic investigations. Hum Hered 1997; 47 : 351-7.

50. Thara R, Padmavati R, Kumar S, Latha S. Burden assessment Schedule, Instrument to assess burden on caregivers of chronic mentally ill. Indian J Psychiatry 1998; 40 : 21-29.

34. Spitzer RL. Psychiatric diagnosis: Are clinician still necessary? In: Williams JBW, Spitzer RL, editors. Psychotherapy research: where are we and where should we go? New York, NY: Guilford Press; 1983 p. 273-92.

51. Padmavati R, Rajkumar S, Srinivasan TN. Schizophrenic patients who were never treated - a study in an Indian urban community. Psychol Med 1998; 28 : 1113-7.

35. Kraemer HC, Measelle JR, Ablow JC, Essex MJ, Boyce WT, Kupfer DJ. A new approach to integrating data from multiple informants in psychiatric assessment and research: Mixing and matching contexts and perspectives. Am J Psychiatry 2003; 160 : 1566-77.

52. Srinivasan TN, Rajkumar S, Padmavati R. Initiating care for untreated schizophrenia patients and results of one-year follow-up. Int J Soc Psychiatry 2001; 47 : 73-80. 53. Prabhu GG, Raghuram A, Verma N, Maridass A. Public attitudes towards mental illness: a review. NIMHANS J 1984; 2 : 1-14.

192

INDIAN J MED RES, SEPTEMBER 2007

54. Jorm F. Mental health literacy, Public knowledge and beliefs about mental disorders. Br J Psychiatry 2000; 177 : 396-401. 55. Lee RM, Renzetti CM. The problems of researching sensitive topics. Am Behav Sci 1990; 33 : 510-28. 56. Reissman CK. Interviewer effects in psychiatric epidemiology: a study of medical and lay interviewers and their impact on reported symptoms. Am J Public Health 1977; 69 : 485-91. 57. Nandi DN, Banerjee G, Mukherjee SP, Sarkar S, Boral GC, Mukrjee A, et al. A study of psychiatric morbidity of a rural community at an interval of ten years. Indian J Psychiatry 1986; 28 : 179-84. 58. Nandi DN, Banerjee G, Mukherjee SP, Ghosh A, Nandi PS, Nandi S. Psychiatric morbidity of a rural Indian community. Changes over a 20-year interval. Br J Psychiatry 2000; 176 : 351-6. 59. Chandrashekar CR, Math SB. Psychosomatic disorders in developing countries: current issues and future challenges. Curr Opin Psychiatry 2006; 19 : 201-6. 60. Reddy MV, Chandrasekar CR. Prevalence of mental and behavioural disorders in India : A metaanalysis. Indian J Psychiatry 1998; 40 : 149-57.

63. Kessler RC, McGonagle Ka, Zhao S, Nelson CB, Hughes M, Eshleman S, et al. Lifetime and 12-month prevalence of DSMIII-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry 1994; 51 : 8-19. 64. Gururaj G, Girish N, Isaac MK. Mental, neurological and substance abuse disorders: Strategies towards a systems approach, NCMH Background Papers-Burden of Disease in India; Available from: http://whoindia.org/en/section102/ section201_888.htm. New Delhi: Ministry of Health & Family Welfare; 2005. 65. Eaton WW, Neufeld K, Chen L, Cai G, Comparison of selfreport and clinical diagnostic interviews for depression, Diagnostic Interview Schedule and Schedules for Clinical Assessment in Neuropsychiatry in the Baltimore Epidemiologic Catchment Area Follow-up. Arch Gen Psychiatry 2000; 57 : 217-22. 66. Belfer ML, Child and adolescent mental health around the world: Challenges for progress. J Indian Assoc Child Adolescent Mental Health 2005; 1 : 31-6.

61. Ganguli HC. Epidemiological finding on prevalence of mental disorders in India. Indian J Psychiatry 2000; 42 : 14-20.

67. Blazer DG, Kaplan BH. Controversies in community-based psychiatric epidemiology, let the data speak for themselves. Arch Gen Psychiatry 2000; 57 : 227-228.

62. Regier DA, Myers JK, Kramer M, Robins LN, Blazer DG, Hough RL, et al. The NIMH Epidemiologic Catchment Area program. Historical context, major objectives, and study population characteristics. Arch Gen Psychiatry 1984; 41 : 934-41.

68. Math SB, Girimaji SC, Benegal V, Uday Kumar GS, Hamza A, Nagaraja D. Tsunami: Psychosocial aspects of Andaman and Nicobar islands. Assessments and intervention in the early phase. Int Rev Psychiatry 2006; 18 : 233-9.

Reprint requests: Dr Suresh Bada Math, Assistant Professor of Psychiatry, Department of Psychiatry, National Institute of Mental Health & Neuro Sciences (Deemed University), Bangalore 560 029, India e-mail: [email protected]