Why does the burden of disease persist? Relating the burden of anxiety and depression to effectiveness of treatment
Pourquoi la charge de morbidité persiste-t-elle? Le lien entre la charge de lanxiété et de la dépression et lefficacité des traitements
¿Por qué persiste la carga de ansiedad y depresión? Relación entre esa carga de morbilidad y la eficacia de los tratamientos
G. AndrewsI; K. SandersonII; T. SladeII; C. IssakidisII
IProfessor, WHO Collaborating Centre for Mental Health, and School of Psychiatry, University of New South Wales at St. Vincents Hospital, Sydney, Australia. 299 Forbes St, Darlinghurst, NSW 2010, Australia
IIResearch Psychologist, WHO Collaborating Centre for Mental Health, and School of Psychiatry, University of New South Wales at St. Vincents Hospital, Sydney, Australia
Why does the burden of mental disorders persist in established market economies? There are four possibilities: the burden estimates are wrong; there are no effective treatments; people do not receive treatment; or people do not receive effective treatments. Data from the Australian National Survey of Mental Health and Wellbeing about the two commonest mental disorders, generalized anxiety disorder and depression, have been used in examining these issues. The burden of mental disorders in Australia is third in importance after heart disease and cancer, and anxiety and depressive disorders account for more than half of that burden. The efficacy of treatments for both disorders has been established. However, of those surveyed, 40% with current disorders did not seek treatment in the previous year and only 45% were offered a treatment that could have been beneficial. Treatment was not predictive of disorders that remitted during the year. The burden therefore persists for two reasons: too many people do not seek treatment and, when they do, efficacious treatments are not always used effectively.
Keywords: anxiety disorders; depressive disorder; treatment outcome; prevalence; epidemiological studies; Australia.
Le projet relatif à la charge mondiale de morbidité amis au point une méthode qui permet de mesurer cette charge en ajoutant les années de vie perdues par suite dune maladie aux années de vie vécues avec une incapacité due à cette maladie. Les troubles mentaux cités dans le rapport du projet sont responsables de moins de 1% des années de vie perdues, de 26% des années vécues avec une incapacité et de 9% de la charge mondiale de morbidité. Dans les pays à économie de marché, les troubles mentaux sont responsables de 2% des années de vie perdues, de 43% des années vécues avec une incapacité et de 22% de la charge totale due à lensemble des maladies. Pourquoi la charge due aux troubles mentaux persiste-t-elle dans les pays à économie de marché? Il y a quatre possibilités : les chiffres sont erronés ; il nexiste pas de traitement efficace ; les malades ne sont pas soignés; les traitements administrés ne sont pas efficaces. On sest servi des données issues de lenquête nationale de santé et de bien-être mental réalisée en Australie sur les deux affections mentales les plus courantes, lanxiété généralisée et la dépression, pour étudier ces possibilités.
Lenquête nationale australienne de santé et de bien-être mental est une enquête auprès des ménages portant sur des adultes, qui a été menée en 1997. Sur les 13 625 adultes remplissant les conditions requises qui ont été recensés, 10 641 ont accepté de répondre aux enquêteurs. Lanxiété généralisée et les épisodes dépressifs aux termes de la CIM-10 sont les troubles mentaux qui ont été les plus couramment observés par la méthode appelée Composite International Diagnostic Interview. Afin de déterminer les raisons de la persistance de la charge associée à ces troubles, on a examiné, pour ces deux affections, la prévalence sur 12 mois et sur 1 mois, lincapacité mesurée sur léchelle SF-12 dévaluation de la santé mentale et les journées dincapacité au cours du mois précédent, lutilisation des services de santé, le traitement administré aux personnes soignées et les besoins perçus en matière de traitement par les personnes nayant pas été soignées. Toutes les données proviennent des entretiens menés dans le cadre de lenquête et elles ont donc été fournies par les sujets eux-mêmes.
La prévalence pondérée sur 12 mois de lanxiété généralisée était de 3,0 %, celle de la dépression de 6,7%. Les taux sur 1 mois étaient inférieurs dun tiers et de moitié respectivement, signe que les chances de rémission, dans le cas de lanxiété généralisée, sont moindres. Les taux sur 1 mois sont présentés à deux niveaux : personnes répondant aux critères le mois précédent, et personnes répondant aux critères le mois précédent cependant que ce diagnostic a été reconnu comme le trouble « unique ou principal », ou trouble de base. Ces deux troubles courants sont plus fréquents chez les femmes, ils sont chroniques, ils compromettent lavenir et ils sont actuellement incapacitants. Bien que la dépression soit plus incapacitante et plus fréquente, lun et lautre troubles devraient être au centre de lattention. Environ 40% des personnes atteintes de troubles navaient pas sollicité de soins au cours de lannée précédente, et 45 % seulement sétaient vu proposer un traitement qui aurait pu être bénéfique. Un traitement nétait pas indicatif dune rémission au cours de lannée.
Les résultats du projet relatif à la charge mondiale de morbidité ont été reproduits en Australie et ils continuent de faire apparaître les troubles mentaux comme une cause importante de charge de morbidité, et lefficacité des traitements de lanxiété généralisée et de la dépression a été établie. La charge persiste donc pour deux raisons: trop de personnes ne sollicitent pas de traitement et, pour celles qui le font, les traitements efficaces ne sont pas toujours utilisés convenablement.
En el marco del proyecto Carga Mundial de Morbilidad se formuló una medida de esa carga que permitía añadir los años de vida perdidos por una enfermedad a los años vividos con una discapacidad debida a esa dolencia. En el informe del proyecto los trastornos mentales enumerados representan menos del 1% de los años de vida perdidos, el 26% de los años vividos con una discapacidad y el 9% de la carga mundial de morbilidad. En las economías de mercado consolidadas los trastornos mentales representan el 2% de los años de vida perdidos, el 43% de los años vividos con una discapacidad y el 22% de la carga total de morbilidad. ¿Por qué persiste la carga de enfermedades mentales en las economías de mercado consolidadas? Cabe pensar en cuatro posibilidades: las estimaciones de la carga son erróneas; no hay tratamientos eficaces; la gente no recibe tratamiento, o la gente no recibe tratamientos eficaces. A fin de resolver esta cuestión se han analizado aquí los datos aportados por el Estudio Nacional de Australia sobre Salud Mental y Bienestar en lo que respecta a los dos trastornos mentales más frecuentes: la ansiedad generalizada y la depresión.
El Estudio Nacional de Australia sobre Salud Mental y Bienestar consistió en una encuesta domiciliaria de ámbito nacional realizada entre personas adultas en 1997. De los 13 625 adultos identificados que satisfacían los requisitos para participar, 10 641 aceptaron ser entrevistados. Los episodios de ansiedad generalizada y depresión fueron los trastornos mentales más comunes detectados mediante la Composite International Diagnostic Interview. Se analizaron esos dos trastornos, considerando la prevalencia sobre 12 meses y sobre un mes, el grado de discapacidad según la escala de salud mental SF-12 y los días de discapacidad durante elmes precedente, la utilización de servicios de salud, el tratamiento recibido por las personas tratadas y las necesidades de tratamiento percibidas por las no tratadas, con objeto de desentrañar las razones de la persistencia de la carga de estas enfermedades. Todos los datos procedían de las entrevistas realizadas, y reflejaban por tanto lo que los propios sujetos manifestaron.
La prevalencia ponderada sobre 12 meses del trastorno de ansiedad generalizada fue del 3,0%, y la de la depresión, del 6,7%. Las prevalencias sobre un mes fueron un tercio y un 50% más bajas, respectivamente, lo que sugiere que la ansiedad generalizada era una dolencia con menos probabilidades de remisión. Las prevalencias sobre un mes se presentaron a dos niveles: teniendo en cuenta los casos que habían cumplido los criterios durante el mes precedente, y sumando a ese requisito la identificación del diagnóstico en cuestión como la « única o principal » afección, considerada así como el trastorno básico. Estas dos dolencias comunes son más frecuentes en las mujeres, tienen carácter crónico, trastocan las perspectivas vitales y son a menudo discapacitantes. Aunque la depresión es más discapacitante y más frecuente, los dos trastornos son preocupantes. Un 40% de las personas afectadas en el momento de la entrevista no habían solicitado tratamiento durante el año precedente, y sólo a un'45% se les había ofrecido un tratamiento eventualmente eficaz. El tratamiento no permitía predecir los trastornos que remitieron durante el año.
Los resultados del proyecto Carga Mundial de Morbilidad se han reproducido en Australia, y siguen mostrando que los trastornos mentales contribuyen de forma importante a la carga de morbilidad, habiéndose establecido la eficacia de los tratamientos empleados contra la ansiedad generalizada y la depresión. Así pues, la carga de morbilidad persiste por dos razones fundamentales: son demasiadas las personas que no buscan tratamiento, y quienes lo hacen no siempre usan eficazmente unos tratamientos que han demostrado su utilidad.
The Global Burden of Disease project developed a measure of burden that allowed the years of life lost due to a disease to be added to the years lived with a disability due to that disease (1). The project, a worldwide collaboration sponsored by WHO and the World Bank and based at the Harvard School of Public Health, showed that the mental disorders enumerated accounted for less than 1% of the years of life lost, 26% of the years lived with disability, and 9% of the global burden of disease. In established market economies, mental disorders accounted for 2% of the years of life lost, 43% of the years lived with a disability, and a sizeable proportion (22%) of the total burden of all diseases. It is not that the developed countries are suffering from an epidemic of mental disorders but that the relative importance of mental disorders has become greater as the impact of infectious and other acute physical disease has been reduced. The estimates of the burden of mental disorders in disability-adjusted life years (DALYs) lost, were remarkably similar in the developed and developing world at 27.7 and 23.2 per thousand of the population, respectively (1). If these estimates are correct, what is causing the burden to persist in the established market economies given that they spend 513% of their considerable health budgets on mental health services? Although these two groups of countries cannot be compared exactly, the question remains as to why the burden of mental disorders persists in established market economies with appropriate services. There are four possibilities: the burden figures are wrong; there are no effective treatments; people do not receive treatment; or people do not receive effective treatments. Each will be discussed in turn and the last two illustrated with data from the Australian National Survey of Mental Health and Wellbeing.
Are the burden estimates correct?
The Global Burden of Disease project included only a limited number of mental disorders: two affective disorders (major depression and bipolar disorder), one psychosis (schizophrenia), two substance-use disorders (alcohol and drug use), three anxiety disorders (panic, obsessive-compulsive disorder and post-traumatic stress disorder) and dementia. Depression was estimated as likely to be the second most important determinant of the global burden of disease by the year 2020, a finding that has mobilized considerable support for mental health services. The project did not include a number of mental disorders, did not allow for comorbidity with physical or other mental disorders, and probably overvalued the burden of depression because of the disability weight used.
The Australian Institute of Health and Welfare has used the prevalence data from the Australian National Survey of Mental Health and Wellbeing (2, 3) and Netherlands disability weights (4), and has allowed for comorbidity by averaging the weights assigned to individuals with comorbid disorders, a rather simpler procedure than the one we had suggested (5). Their preliminary report (6) estimated that mental disorders (dementia excluded) account for 13% of the total burden of disease in Australia, more than half of this attributed to anxiety and depressive disorders. This is, in total, less than the World Bank estimates for established market economies such as Australia, the difference being explained as much by excluding dementia, not using age weighting, using more appropriate severity weights, and controlling for comorbidity, as by any true lessening of burden. Nevertheless, the questions remain as to why the burden persists and why mental disorders are the third most important determinant of burden in Australia, where there is subsidized access to medical services and potentially effective treatments for most disorders.
Are effective treatments available?
The range of treatments for mental disorders of proven efficacy is impressive. A recent publication (7) lists an average of 2.5 different treatments for each of the common mental disorders, each treatment having been shown, in replicated randomized controlled trials, to be superior to placebo. (Placebo-controlled trials establish the efficacy of a treatment, that is, whether it is effective in the particular setting of the study.) Whether the treatment is effective when used by ordinary practitioners with ordinary patients is, of course, the key question, but there is little evidence about this (8). To establish the effectiveness of treatment in routine medical practice it is necessary to have external end-points, like mortality or suicide rates which are gathered by independent mechanisms, or routine measurement of outcome so that practitioners become oblivious to the demand characteristics of measurement. Complex experimental designs may be suitable (9), but even pharmaco-economic data usually fail to relate to any end-point independent of the physicians opinion. Such studies as have been done suggest that the effectiveness of treatments declines closer to routine practice (8). The cost-effectiveness studies required by regulatory authorities are commonly cost-efficacy studies, that is, not studies of the efficiency of medical practice, but studies of the potential efficiency if the patients are uncomplicated and everything is done correctly. Neither of these conditions is usual.
Do most people get effective treatment?
There is considerable evidence, in many countries, that only a minority of people with mental disorders consult a physician for treatment (10). The majority do not get treatment and so the issue of effectiveness does not arise, for the system has simply failed these people. In this study, data from a population survey have been used to develop a method to explore treatment access and effectiveness to answer the question Why, for the two most common mental disorders, generalized anxiety disorder and depression, does the burden persist? Data are presented on the perceived health needs of people who did not consult, on the treatment experience of those who did remit during the previous year, and on the treatment experience of those who did consult yet whose disorders remained unresolved at the time of the survey.
Depression (depressive episode, single or recurrent, ICD-10 Diagnostic Criteria for Research F32.x) (11) is characterized by persistent depressed mood not fully explained by circumstances, loss of interest or pleasure, and decreased energy extending for at least 2 weeks coupled with loss of confidence, self-reproach or guilt, suicidal ideation, diminished ability to think or concentrate, change in psychomotor activity, and sleep or appetite disturbance. The diagnosis of mild depressive episode is made when four of the ten symptoms are positive, severe when eight of the ten are positive. Depression is a chronic disorder, even with treatment. There are a number of 12- and 15-year prospective studies of clinical samples from specialist practice (12, 13) that show that only one person in five recovers completely and that, on average, someone diagnosed clinically as having depression can expect symptoms in 60% of weeks over the next 12 years, and meet criteria for the full syndrome in 15% of those weeks. The natural history of the disorder in primary care is not known. Primary prevention programmes are in their infancy but there is good evidence about the efficacy of treatment: four classes of antidepressant drugs, cognitive behaviour therapy and electroconvulsant therapy have all been shown (7) to be superior (standard deviation 0.51.0) to placebo. The placebo effect in depression, which actually measures the effect of placebo and of spontaneous remission, is important given the chronic remitting nature of the condition; it is about twice the size of the treatment effect (14). Effectiveness appears to be much less than efficacy. Even so, many people who receive treatment do recover, unfortunately only to relapse. There are no randomized controlled trials on the influence of continued treatment on long-term outcome, and periods without treatment are surprisingly common and have been related to relapse (15).
Generalized anxiety disorder (ICD-10 Diagnostic Criteria for Research F41.1) is characterized by at least 6 months of prominent tension, worry and feelings of apprehension about everyday events and problems, coupled with at least four of 20 symptoms, most associated with autonomic arousal. Even more than with depression, effective prevention seems possible but is not widely applied (16). Follow-up studies are short and chronicity is marked, even in the face of treatment. In a 2-year follow-up of a specialist clinical sample, 88% stayed at the same symptom level over the first 6 months and only 25% experienced a period of remission exceeding 8 weeks during the follow-up period despite virtually all patients being actively treated (17). In a 5-year follow-up only 18% of another specialist clinical sample had no symptoms and 46% still met full criteria (18). Remission in these studies was not associated with treatment. There are no studies of the natural history of this disorder in primary care. The data on efficacy are different to those for depression. Response to placebo is of the order of one standard deviation, and cognitive behaviour therapy, buspirone, benzodiazepines and tricyclic anxiety depressants have all been shown to produce further improvements of 12 standard deviations over placebo (19). As with depression, why is this apparently treatable disorder so chronic, and why does the burden persist?
The Australian National Survey of Mental Health and Wellbeing was a nationwide household survey of adults conducted in 1997 (2, 3). Of the 13 625 eligible adults identified, 10 641 (78.1%) were interviewed. ICD-10 generalized anxiety disorder and depressive episode were the commonest mental disorders identified by the Composite International Diagnostic Interview (3). These two disorders were examined in terms of the 12- and 1-month prevalences, disability determined from the score on the SF-12 mental health scale (20) and disability days in the previous month, health service utilization, treatment received by those treated, and perceived treatment needs by those who were not, to explore reasons why the burden of these disorders persists. All data are from the Survey interview and hence are based on self-report.
Burden estimates can be reduced when prevalence is lowered by reducing the number of incident cases, by reducing the duration of the episode, or by reducing the severity or disability. The more important question is whether any reduction in prevalence or disability is associated with treatment and, especially, associated with a treatment deemed to be useful.
In this study, prevalence is presented as the proportions of the population meeting criteria for the disorder, according to ICD-10 criteria as determined by the Composite International Diagnostic Interview, in the previous 12 months, and the subgroup meeting the criteria in the previous month. Application of exclusion criteria reduced the frequency of generalized anxiety disorder by 43% (see ICD-10, F41.1, Criteria C and D), but did not affect the frequency of depression. Comorbidity among mental disorders is much more frequent than would be expected from prevalence estimates (21), so disability, service utilization, and treatment rates for all people with a disorder could reflect the influence of other disorders. To rely on people with only a single diagnosis would, however, focus on people who were more mildly affected than average (5). Respondents who met criteria for two or more disorders were therefore asked to identify their main set of complaints. Data for disability, service utilization and not seeking treatment were analysed for those with a current disorder and for the further subgroup who considered their current disorder to be their only or main complaint, referred to as a core disorder. It was considered that the latter subgroup would provide better information for the key question as to why the burden of the disorder persists. People who had met criteria for generalized anxiety disorder or depression during the previous year but who were no longer current cases were a further source of information as to whether their remission was likely to be associated with treatment.
The weighted 12-month prevalence of generalized anxiety disorder was 3.0%, that of depression 6.7%. The 1-month prevalences were one-third and one-half lower, respectively, suggesting, as described earlier, that generalized anxiety disorder was a condition in which remission was less likely. The 1-month prevalence is presented at two levels: meeting criteria in the previous month, and meeting criteria in the previous month coupled with identification of this diagnosis as the core disorder (Table 1). The demographic and disability characteristics of these current and core groups are also presented.
Half the people with current generalized anxiety disorder or depression identified this as their core disorder. People with generalized anxiety disorder were older and less likely to have never married than people with depression. Depression is more disabling than generalized anxiety disorder, whether this is measured by the overall SF-12 score or the disability days associated with each disorder. For both disorders, the people with core disorders reported fewer disability days than the total for current cases because the influence of other physical and mental disorders is reduced. Nevertheless, the SF-12 scores for core cases are in the moderate range of disability (22) and as such are significant. In summary, both these common disorders are more frequent in women, are chronic, impair life prospects and currently disable. Even though depression is more disabling and more frequent, both disorders should be a focus of concern.
The service utilization data showed that half to two-thirds of the current cases had had a consultation for a mental problem in the previous year; the rates in the previous month were lower (Table 1). Most reported receiving some treatment but a lower proportion reported receiving either medication or cognitive behaviour therapy, the two forms of treatment for which there is evidence of efficacy if delivered appropriately and the patient complies with the regime. In the core generalized anxiety disorder group only half had consulted for their principal disorder in the previous year (29% in the previous month). Most of those who had consulted had received treatment, but only six out of ten who had been treated received either medication or cognitive behaviour therapy. For depression the picture is better. Two-thirds of the core group had consulted in the previous year, 45% in the previous month. Virtually all of these had been treated, and eight out of ten had received medication or cognitive behaviour therapy. Many people in the core group did not consult. SF-12 scores were comparable to those for all current cases (Table 2) so, on average, they were moderately disabled. They were asked whether they needed any help for their symptoms. Four out of ten denied any need. The remainder, who said they had needs, were asked what type of help was needed. The most popular responses were information about mental illness, its treatments and available services and counselling or a talking therapy. The least favoured category was medicine or tablets.
One-third (124/375) of the people who had met the criteria for generalized anxiety disorder during the previous year and about half (399/822) of those who had met the criteria for depression during the same period had gone into remission and were not cases at the time of the survey. Again it was possible to identify a group for whom the diagnosis was their core disorder. At the time of the Survey, these people were less disabled and their SF-12 scores were near to normal levels. For both disorders, compared with those who were symptomatic, they were less likely to have consulted during the year, less likely to have received any treatment, and less likely to have received a potentially effective intervention. In the majority of cases it could not be concluded that the remission was related to treatment; much of it must be attributed to the natural history of these chronic conditions.
The key elements in effective health care for disorders for which there are efficacious treatments are coverage, clinician competence and compliance, and patient compliance. Despite universal health insurance in Australia, coverage was found to be poor. The Survey did not provide direct evidence of clinician competence or clinician or patient compliance, only data on what treatment the respondents reported receiving when they did consult or wanting if they did not consult. Chassin, discussing health care quality in the United States of America, concluded that, since 58% of people with depression were not detected or adequately treated, the treatment of depression was a procedure of very low quality, in his terminology a sigma 1.5 procedure (23). He likened the situation to a manufacturer, 58% of whose products were defective. The results of the present study are similar and indicate that the burden of anxiety and depression persists not because of a lack of effective treatments but because the public do not understand what treatments could be of benefit and because practitioners do not always offer them. Only 32% and 52% of people whose core concern was generalized anxiety disorder or depression, respectively, consulted and were offered a potentially effective treatment. Only some of this group (15, 24, 25) and none of the remainder could have been adequately treated. There was no evidence from the group who had remitted during the year that for the majority of cases their remission was treatment induced. Moreover, 50% of people with generalized anxiety disorder and 35% of those with depression as their core concern had not consulted for a mental health problem at any time in the previous 12 months, even though significantly disabled. Four out of ten of this group did not think they needed any help. Six out of ten agreed that they did need help and the majority identified information about their disorder or counselling as their principal needs. Medication, the only evidence-based therapy in the list of treatments enquired about, was not commonly identified as a need.
This study presents a method for identifying some of the reasons why the burden for mental disorders persists in the face of apparently efficacious treatments. The Composite International Diagnostic Interview is the most widely used diagnostic instrument for epidemiological studies (26). The self-identification of what was regarded as the main complaint when a person met criteria for more than one disorder is relatively new (27). While clinically defensible, the actual questions used should now be subject to cognitive probe studies to determine exactly what people understood by them. The measurement of disability followed standard procedures (28, 29). The identification of health service utilization parameters was closely modelled on the work of Kessler (30). The classification of interventions and the measurement of perceived health needs was developed for this Survey (31) and proven during pilot trials. In part, the expressed need for information about the disorder and the choice of non-effective remedies as the preferred treatments reflect a general problem with mental health literacy (32) that must be addressed.
Two-thirds to three-quarters of people identified in epidemiological surveys as meeting criteria for a mental disorder do not report receiving treatment (10). The present data are not exceptional. The problem remains as to why, given the expenditure on mental health services in developed countries, effective services are not more generally available and coverage, physician compliance with evidence-based medicine data and patient compliance are not better. There is a growing international literature, sponsored by the International Consortium of Psychiatric Epidemiology, which compares the data on mental health service use between countries. Treatment-seeking is related to morbidity and disability and to the usual socioeconomic variables. In addition, perceived need for help (and this includes reasonable expectations of benefit) on the part of the patient, and the structure and training of the clinical services are salient factors in determining health service utilization (3335).
The burden of disease in these two fluctuating but very chronic and disabling conditions will not be relieved while only a third or half of core patients are receiving potentially effective treatments. Furthermore, even in such patients there is evidence that the treatment given is usually inadequate (8), simply because few practitioners have the skill to implement the proven treatments correctly. The first implication is therefore that there is a need to raise the level of effectiveness of current practice. The present emphasis on evidence-based medicine should address this deficit once we have learned how to implement clinical practice guidelines, a considerable task (36). The second implication is the need to deal with the mental health literacy problem that leads to effective treatments being actively avoided. Again, a considerable problem, but surmountable given the success of health promotion strategies in other fields.
The most effective strategy would probably be to step back and view the trajectory of these chronic diseases, and design programmes for disease management, rather than responding to particular shortcomings in service delivery (10). For instance, specific prevention programmes in adolescence are very promising and very cost-effective (16). They should be routine for all schoolchildren at risk. Education for all adults about how to combat depression, anxiety and fear should be available through the common media. At present the media are fascinated by accounts of exceptional strategies used by individual survivors instead of promoting proven strategies of benefit to many. Professionally guided self-treatment by books, computer programs and self-help groups have been proven in randomized controlled trials to be efficacious (37, 38). Health services should recognize the need to fund professional input to such self-treatment strategies. At present, when such input occurs, it is too often aimed at comfort that serves to justify the suffering, rather than at strategies to cure it. Lastly, there is the panoply of direct patient care services, from primary care where the majority of patient contact occurs, to ambulatory and inpatient specialist care. We must strive to make these services efficient. The report of the WHO Ad Hoc Committee on Health Research provides a model for analysing the burden of a health problem to identify research needs (39). The data presented here indicate that only a small proportion of the burden attributed to two common mental disorders is being averted with the current mix of interventions and population coverage. A considerable proportion of the remainder would be averted with improved efficiency. The solution therefore lies in further research on health systems and policies, work which the authors are currently undertaking.
This study was supported by a contract from the Australian Department of Health and Aged Services to the WHO Collaborating Centre for Mental Health, Sydney, to support a survey data analysis consortium (G. Andrews, V. Carr, G. Carter, R. Crino, W. Hall, A. Henderson, I. Hickie, C. Hunt, L. Lampe, A. McFarlane, P. Mitchell, L. Peters, M. Teesson and K. Wilhelm).
1. Murray CJL, Lopez AD, eds. The global burden of disease: a comprehensive assessment of mortality and disability for diseases, injuries, and risk factors in 1990 and projected to 2020. Cambridge, MA, Harvard University School of Public Health on behalf of the World Health Organization and The World Bank, 1996.
2. McLennan W. Mental health and wellbeing: profile of adults, Australia, 1997. Canberra, Australian Bureau of Statistics, 1998 (Publication No. 4326.0).
3. Andrews G et al. The mental health of Australians. Canberra, Commonwealth Department of Health and Aged Care, 1999.
4. Stouthard MEA et al. Disability weights for diseases in the Netherlands. Rotterdam, Department of Public Health, Erasmus University of Rotterdam, 1997.
5. Andrews G, Sanderson K, Beard J. Burden of disease: methods of calculating the disability from mental disorder. British Journal of Psychiatry, 1998, 173: 123131.
6. Mathers C, Vos T, Stevenson C. The burden of disease and injury in Australia. Canberra, Australian Institute of Health and Welfare, 1999.
7. Nathan PE, Gorman J, eds. A guide to treatments that work. New York, Oxford University Press, 1998.
8. Andrews G. Efficacy, effectiveness and efficiency inmental health service delivery. Australian and New Zealand Journal of Psychiatry, 1999, 33: 316322.
9. Wells KB. Treatment research at the crossroads: the scientific interface of clinical trials and effectiveness research. American Journal of Psychiatry, 1999, 156: 510.
10. Andrews G. Meeting the unmet need through disease management. In: Andrews G, Henderson S, eds. The unmet need in psychiatry. Cambridge, Cambridge University Press, 2000.
11. The International Statistical Classification of Diseases and Related Health Problems, tenth revision. Geneva, World Health Organization, 1992.
12. Kiloh LG, Andrews G, Neilson MD. The long term outcome of depression. British Journal of Psychiatry, 1988, 153: 752757.
13. Judd LL et al. A prospective 12 year study of subsyndromal and syndromal depressive symptoms in unipolar major depressive disorders. Archives of General Psychiatry, 1998, 55: 694700.
14. Enserink M. Can the placebo be the cure? Science, 1999, 284: 238240.
15. Mueller TI et al. Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. American Journal of Psychiatry, 1999, 156: 10001006.
16. Dadds MR, Spence SH. Prevention and early intervention for anxiety disorders: a controlled trial. Journal of Consulting and Clinical Psychology, 1997, 65: 627635.
17. Yonkers KA et al. Phenomenology and course of generalized anxiety disorder. British Journal of Psychiatry, 1996, 168: 308313.
18. Woodman CL et al. A 5 year follow-up study of generalized anxiety disorder and panic disorder. Journal of Nervous and Mental Disease, 1999, 187: 39.
19. Andrews G et al. The treatment of anxiety disorders. New York, Cambridge University Press, 1994.
20. Ware JE, Kosinski M, Keller SD. A 12-item short-form health survey: construction of scales and preliminary tests of reliability and validity. Medical Care, 1996, 34: 220233.
21. Andrews G. Comorbidity and the general neurotic syndrome. British Journal of Psychiatry, 1996, 168 (Suppl. 30): 7684.
22. Sanderson K, Andrews G. Mental health-related disability in the community: prevalence, severity and relationship to diagnosis. Sydney, Clinical Research for Anxiety Disorder,Working Paper 2000/1.
23. Chassin MR. Is health care ready for six sigma quality? Milbank Quarterly, 1998, 76: 565591.
24. Wittchen H-U. Met and unmet need in anxiety disorders. In: Andrews G, Henderson S, eds. The unmet need in psychiatry. Cambridge, Cambridge University Press, 2000.
25. Hunt C. The unmet need for treatment in panic disorder and social phobia. In: Andrews G, Henderson S, eds. The unmet need in psychiatry. Cambridge, Cambridge University Press, 2000.
26. Andrews G, Peters L. Psychometric properties of the CIDI. Social Psychiatry and Psychiatric Epidemiology, 1998, 33: 8088.
27. Hunt C, Andrews G. Comorbidity of anxiety disorders: a life chart approach. Journal of Psychiatric Research, 1995, 29: 467480.
28. Ware JE, Kosinski M, Keller SD. A 12 item short form health survey: construction of scales and preliminary tests of reliability and validity. Medical Care, 1996, 34: 220233.
29. Kessler RC, Frank RG. The impact of psychiatric disorders on work loss days. Psychological Medicine, 1997, 27: 861873.
30. Kessler RC et al. Lifetime and 12 month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Archives of General Psychiatry, 1994, 51: 819.
31. Meadows G et al. The assessment of perceived need. In: Andrews G, Henderson S, eds. The unmet need in psychiatry. Cambridge, Cambridge University Press, 2000.
32. Jorm AF, Angermeyer M, Katschnig H. Public knowledge of and attitudes tomental disorders: a limiting factor in the optimal use of treatment services. In: Andrews G, Henderson S, eds. The unmet need in psychiatry. Cambridge, Cambridge University Press, 2000.
33. Katz SJ et al. Mental health care use, morbidity, and socioeconomic status in the United States and Ontario. Inquiry, 1997, 34: 3849.
34. Katz SJ et al. The use of outpatients mental health services in the United States and Ontario: the impact of mental morbidity and perceived need for care. American Journal of Public Health, 1997, 87: 11361143.
35. Allegria M et al. Comparing data on mental health service use between countries. In: Andrews G, Henderson S, eds. The unmet need in psychiatry. Cambridge, Cambridge University Press, 2000.
36. Andrews G. Randomised controlled trials in psychiatry: important but poorly accepted. British Medical Journal, 1999, 319: 562564.
37. Ghosh A, Marks I, Carr A. Therapist contact and outcome of self exposure treatment for phobias. British Journal of Psychiatry, 1988, 152: 234238.
38. Griest J. COPE self help program for depression. Middleton, WI, Healthcare Technology Systems, 1998.
39. Ad Hoc Committee on Health Research Relating to Future Intervention Options. Investing in health research and development. Report of the Ad Hoc Committee on Health Research Relating to Future Intervention Options. Geneva, World Health Organization, 1996 (document TDR/GEN/96.2).
WHO Collaborating Centre for Mental Health, and School of Psychiatry, University of New South Wales at St. Vincents Hospital, Sydney, Australia
299 Forbes St, Darlinghurst, NSW 2010, Australia