Evidence-based health policy - Lessons from the global burden of disease study

Science

Volumn 274, Issue 5288, 1996, Pages 740-743

  Murray, Christopher J L ^a   Lopez, Alan D ^b  

^a HARVARD SCHOOL OF PUBLIC HEALTH   (United States)

^b WORLD HEALTH ORGANIZATION   (Switzerland)

Author keywords

[No Author keywords available]

Indexed keywords

AGE DISTRIBUTION; CAUSE OF DEATH; COMMUNICABLE DISEASE; DISEASE CLASSIFICATION; HEALTH CARE POLICY; HUMAN; INJURY; MORBIDITY; MORTALITY; PRIORITY JOURNAL; PUBLIC HEALTH; REVIEW;

COST OF ILLNESS; HEALTH POLICY; HUMANS; MORBIDITY; MORTALITY; RISK FACTORS; UNITED NATIONS; WORLD HEALTH; WORLD HEALTH ORGANIZATION;

EID: 0029907597     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.274.5288.740     Document Type: Review

References (23)

1. Interest in tuberculosis had waned to an historic low in the mid-1980s, but a series of analyses stimulated by the World Bank as part of their Health Sector Priorities Review [for example, see C. J. L. Murray, K. Styblo, A. Rouillon, Bull. IUATLD 65, 2 (1990);
2. C. J. L. Murray et al., Lancet 338, 1305 (1991)], the rising case numbers of tuberculosis in developed nations, and the interaction between HIV and tuberculosis led to tuberculosis being declared a global emergency by WHO.
3. Preliminary results from the GBD were published in brief in the World Development Report 1993: Investing in Health (Oxford Univ, Press, New York, 1993) and more extensively in 1994 [C. J. L. Murray, Bull. WHO 72, 429 (1994);_and A. D. Lopez, ibid., p. 447; ibid., p. 481; C. J. L. Murray et al., ibid., p. 495]. The complete details on the methods and final results are being published in a 10-volume series. The first two volumes of this series, which provide an overview of the study, have been published: C. J. L. Murray and A. D. Lopez, Eds., The Global Burden of Disease, vol. I: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020 (Harvard Univ. Press, Cambridge, MA, 1996) and The Global Burden of Disease, vol. II: Global Health Statistics: A Compendium of Incidence, Prevalence, and Mortality Estimates for Over 200 Conditions (Harvard Univ. Press, Cambridge, MA, 1996).
4. Preliminary results from the GBD were published in brief in the World Development Report 1993: Investing in Health (Oxford Univ, Press, New York, 1993) and more extensively in 1994 [C. J. L. Murray, Bull. WHO 72, 429 (1994);_and A. D. Lopez, ibid., p. 447; ibid., p. 481; C. J. L. Murray et al., ibid., p. 495]. The complete details on the methods and final results are being published in a 10-volume series. The first two volumes of this series, which provide an overview of the study, have been published: C. J. L. Murray and A. D. Lopez, Eds., The Global Burden of Disease, vol. I: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020 (Harvard Univ. Press, Cambridge, MA, 1996) and The Global Burden of Disease, vol. II: Global Health Statistics: A Compendium of Incidence, Prevalence, and Mortality Estimates for Over 200 Conditions (Harvard Univ. Press, Cambridge, MA, 1996).
5. Preliminary results from the GBD were published in brief in the World Development Report 1993: Investing in Health (Oxford Univ, Press, New York, 1993) and more extensively in 1994 [C. J. L. Murray, Bull. WHO 72, 429 (1994);_and A. D. Lopez, ibid., p. 447; ibid., p. 481; C. J. L. Murray et al., ibid., p. 495]. The complete details on the methods and final results are being published in a 10-volume series. The first two volumes of this series, which provide an overview of the study, have been published: C. J. L. Murray and A. D. Lopez, Eds., The Global Burden of Disease, vol. I: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020 (Harvard Univ. Press, Cambridge, MA, 1996) and The Global Burden of Disease, vol. II: Global Health Statistics: A Compendium of Incidence, Prevalence, and Mortality Estimates for Over 200 Conditions (Harvard Univ. Press, Cambridge, MA, 1996).
6. Preliminary results from the GBD were published in brief in the World Development Report 1993: Investing in Health (Oxford Univ, Press, New York, 1993) and more extensively in 1994 [C. J. L. Murray, Bull. WHO 72, 429 (1994);_and A. D. Lopez, ibid., p. 447; ibid., p. 481; C. J. L. Murray et al., ibid., p. 495]. The complete details on the methods and final results are being published in a 10-volume series. The first two volumes of this series, which provide an overview of the study, have been published: C. J. L. Murray and A. D. Lopez, Eds., The Global Burden of Disease, vol. I: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020 (Harvard Univ. Press, Cambridge, MA, 1996) and The Global Burden of Disease, vol. II: Global Health Statistics: A Compendium of Incidence, Prevalence, and Mortality Estimates for Over 200 Conditions (Harvard Univ. Press, Cambridge, MA, 1996).
7. Preliminary results from the GBD were published in brief in the World Development Report 1993: Investing in Health (Oxford Univ, Press, New York, 1993) and more extensively in 1994 [C. J. L. Murray, Bull. WHO 72, 429 (1994);_and A. D. Lopez, ibid., p. 447; ibid., p. 481; C. J. L. Murray et al., ibid., p. 495]. The complete details on the methods and final results are being published in a 10-volume series. The first two volumes of this series, which provide an overview of the study, have been published: C. J. L. Murray and A. D. Lopez, Eds., The Global Burden of Disease, vol. I: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020 (Harvard Univ. Press, Cambridge, MA, 1996) and The Global Burden of Disease, vol. II: Global Health Statistics: A Compendium of Incidence, Prevalence, and Mortality Estimates for Over 200 Conditions (Harvard Univ. Press, Cambridge, MA, 1996).
8. Preliminary results from the GBD were published in brief in the World Development Report 1993: Investing in Health (Oxford Univ, Press, New York, 1993) and more extensively in 1994 [C. J. L. Murray, Bull. WHO 72, 429 (1994);_and A. D. Lopez, ibid., p. 447; ibid., p. 481; C. J. L. Murray et al., ibid., p. 495]. The complete details on the methods and final results are being published in a 10-volume series. The first two volumes of this series, which provide an overview of the study, have been published: C. J. L. Murray and A. D. Lopez, Eds., The Global Burden of Disease, vol. I: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020 (Harvard Univ. Press, Cambridge, MA, 1996) and The Global Burden of Disease, vol. II: Global Health Statistics: A Compendium of Incidence, Prevalence, and Mortality Estimates for Over 200 Conditions (Harvard Univ. Press, Cambridge, MA, 1996).
9. Preliminary results from the GBD were published in brief in the World Development Report 1993: Investing in Health (Oxford Univ, Press, New York, 1993) and more extensively in 1994 [C. J. L. Murray, Bull. WHO 72, 429 (1994);_and A. D. Lopez, ibid., p. 447; ibid., p. 481; C. J. L. Murray et al., ibid., p. 495]. The complete details on the methods and final results are being published in a 10-volume series. The first two volumes of this series, which provide an overview of the study, have been published: C. J. L. Murray and A. D. Lopez, Eds., The Global Burden of Disease, vol. I: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020 (Harvard Univ. Press, Cambridge, MA, 1996) and The Global Burden of Disease, vol. II: Global Health Statistics: A Compendium of Incidence, Prevalence, and Mortality Estimates for Over 200 Conditions (Harvard Univ. Press, Cambridge, MA, 1996).
10. For each disease or injury, experts (or groups of experts in some cases) collaborated on the study. These experts were drawn from WHO; the World Bank; the U.S. Centers for Disease Control and Prevention; and academic institutions in many countries, including the United States, the United Kingdom, France, Mexico, New Zealand, Japan, India, Sri Lanka, China, and South Africa, Volumes III through X of the Global Burden of Disease and Injury Series, [C. J. L. Murray and A. D. Lopez, Eds. (Harvard Univ. Press, Cambridge, MA, 1996)] provide a series of disease-, injury-, and risk factor-specific chapters authored by these collaborators. In addition, a number of other individuals have provided critical input to the study (see the acknowledgments in the series.)
11. C. J. L. Murray and A. D. Lopez, in The Global Burden of Disease, C. J. L. Murray and A. D. Lopez, Eds. (Harvard Univ. Press, Cambridge, MA, 1996), vol. 1, pp. 325-396.
12. A computer program, DisMod, was developed specifically for the GBD to facilitate the analysis of internal consistency. In this program, the user inputs a set of age-specific incidence, remission, and case-fatality rates, and the program calculates the expected duration and prevalence and death rates in a stable population. Further information on the currently available versions of this software can be obtained at http://www.hsph.harvard.edu/organizations/bdu or by contacting the authors.
13. DALYs can be considered a standardized form of quality-adjusted life years (QALYs). For useful discussions of the strengths, limitations, and uses of QALYs, see M. C. Weinstein, Int. J. Tech. Assess. Health Care 6, 93 (1990); E. Nord, Soc. Sci. Med. 41, 1429 (1992); J. Richardson, ibid. 39, 7 (1994).
14. DALYs can be considered a standardized form of quality-adjusted life years (QALYs). For useful discussions of the strengths, limitations, and uses of QALYs, see M. C. Weinstein, Int. J. Tech. Assess. Health Care 6, 93 (1990); E. Nord, Soc. Sci. Med. 41, 1429 (1992); J. Richardson, ibid. 39, 7 (1994).
15. DALYs can be considered a standardized form of quality-adjusted life years (QALYs). For useful discussions of the strengths, limitations, and uses of QALYs, see M. C. Weinstein, Int. J. Tech. Assess. Health Care 6, 93 (1990); E. Nord, Soc. Sci. Med. 41, 1429 (1992); J. Richardson, ibid. 39, 7 (1994).
16. βx (where C is a constant, x is age, and β is 0.04) to reflect the greater social role played by adults in caring and providing for the young and the old. Time lost due to premature mortality has been discounted at 3%, so that a year of life lost in the future is less valuable than a year of life lost today. Years Lived with Disability (YLDs) are time lived in health states worse than perfect health, weighted by a severity weight for each health state. The severity weights range from 0 (which is perfect health) to 1 (which is the weight for a health state equivalent to death). The preferences of groups of individuals for different health states have been measured by the person trade-off (PTO) method. In the most simple form of a PTO, a person is offered two alternatives. Alternative 1 is to extend life for individuals in normal health, and alternative 2 is to extend life for y individuals in health state i. The term y is varied until the respondent is indifferent to the choice between the two alternatives, at which point the preference for state i is (1-x/y) Other forms of PTOs can be structured in which people are asked to trade off restoring health to x individuals in health state i versus extending life for y healthy individuals for a fixed period of time. Severity weights for 22 indicator conditions have been developed with the PTO method applied to a group that was representative of each region of the world, convened at WHO. Very similar results have been found in eight other preference measurement exercises in other countries with the same PTO protocol. On the basis of the disability severity weights for the 22 indicator conditions, seven classes of disability have been defined, so that each class of disability contains two to three indicator conditions. For each of the remaining several hundred disabling sequelae included in the study, the group convened at WHO selected the distribution of severity across the seven classes for both treated and untreated forms of the condition. Time lived with disability is also age-weighted and discounted in the same manner as YLLs. For a complete discussion of various methods to measure preferences for living with various health conditions and the methods used to apply the PTO method in the GBD, see C. J. L. Murray, in The Global Burden of Disease, C. J. L. Murray and A. D. Lopez, Eds. (Harvard Univ. Press, Cambridge, MA, 1996), vol. 1, pp. 22-43; ibid., pp. 90-98.
17. βx (where C is a constant, x is age, and β is 0.04) to reflect the greater social role played by adults in caring and providing for the young and the old. Time lost due to premature mortality has been discounted at 3%, so that a year of life lost in the future is less valuable than a year of life lost today. Years Lived with Disability (YLDs) are time lived in health states worse than perfect health, weighted by a severity weight for each health state. The severity weights range from 0 (which is perfect health) to 1 (which is the weight for a health state equivalent to death). The preferences of groups of individuals for different health states have been measured by the person trade-off (PTO) method. In the most simple form of a PTO, a person is offered two alternatives. Alternative 1 is to extend life for individuals in normal health, and alternative 2 is to extend life for y individuals in health state i. The term y is varied until the respondent is indifferent to the choice between the two alternatives, at which point the preference for state i is (1-x/y) Other forms of PTOs can be structured in which people are asked to trade off restoring health to x individuals in health state i versus extending life for y healthy individuals for a fixed period of time. Severity weights for 22 indicator conditions have been developed with the PTO method applied to a group that was representative of each region of the world, convened at WHO. Very similar results have been found in eight other preference measurement exercises in other countries with the same PTO protocol. On the basis of the disability severity weights for the 22 indicator conditions, seven classes of disability have been defined, so that each class of disability contains two to three indicator conditions. For each of the remaining several hundred disabling sequelae included in the study, the group convened at WHO selected the distribution of severity across the seven classes for both treated and untreated forms of the condition. Time lived with disability is also age-weighted and discounted in the same manner as YLLs. For a complete discussion of various methods to measure preferences for living with various health conditions and the methods used to apply the PTO method in the GBD, see C. J. L. Murray, in The Global Burden of Disease, C. J. L. Murray and A. D. Lopez, Eds. (Harvard Univ. Press, Cambridge, MA, 1996), vol. 1, pp. 22-43; ibid., pp. 90-98.
18. The term "demographic transition" is often used to describe the decline in fertility and mortality rates that most developed and a number of developing countries have undergone during this century. Omran [Millbank Mem. Fund Q. 49, 509 (1971)] coined the term "epidemiological transition" to refer to the characteristic shift in the composition of causes of death from infectious and parasitic to noncommunicable diseases. More recently, J. Frenk et al. [Health Policy Planning 4, 29 (1989)] have used the "health transition" to refer to the changes in fertility, mortality, cause of death composition, disability, and the health care system's response to these trends. We use "demographic transition" to refer to the characteristic changes in fertility and mortality rates that accompany development, and "epidemiological transition" to refer to changes in the age-specific mortality and disability rates that may accompany development.
19. The term "demographic transition" is often used to describe the decline in fertility and mortality rates that most developed and a number of developing countries have undergone during this century. Omran [Millbank Mem. Fund Q. 49, 509 (1971)] coined the term "epidemiological transition" to refer to the characteristic shift in the composition of causes of death from infectious and parasitic to noncommunicable diseases. More recently, J. Frenk et al. [Health Policy Planning 4, 29 (1989)] have used the "health transition" to refer to the changes in fertility, mortality, cause of death composition, disability, and the health care system's response to these trends. We use "demographic transition" to refer to the characteristic changes in fertility and mortality rates that accompany development, and "epidemiological transition" to refer to changes in the age-specific mortality and disability rates that may accompany development.
20. The impressive rise in the burden associated with neuropsychiatric conditions is exclusively due to the projected increase in the number of young and middle-aged adults who are at the highest risk of these conditions. As a result of past changes in fertility, the number of children under age 5 in the developing world will increase from 1990 to 2020 by 25% and the number of adults between ages 45 and 59 will increase by 140%. We were unable to find convincing evidence concerning the trends or determinants of trends in the age-specific rates of most neuropsychiatric conditions. We nave therefore assumed for many of these conditions that age-specific incidence rates will remain constant. Some may argue persuasively that these estimates are conservative, as a number of socioeconomic developments may contribute to higher rates of mental illness; for example, the higher rates of war experienced in recent years may contribute to higher rates of post-traumatic stress disorder in the coming decades.
21. R. G. A. Feachem et al., The Health of Adults in the Developing World (Oxford Univ. Press, New York, 1992) is one of the few examples of such research.
22. D. T. Jamison et al., Disease Control Priorities in Developing Countries (Oxford Univ. Press, New York, 1993).
23. For a review of these 10 studies, see C. J. L. Murray and A. D. Lopez, in The Global Burden of Disease, C. J. L. Murray and A. D. Lopez, Eds. (Harvard Univ. Press, Cambridge, MA, 1996), pp. 295-324. More complete discussions of methods and results are provided in chapters on each specific risk factor.