Volume 24
Number 4
2003

[Table of Contents]

Cause-deleted health-adjusted life expectancy of Canadians with selected chronic conditions

Douglas G Manuel, Wei Luo, Anne-Marie Ugnat and Yang Mao

Abstract

Health-adjusted life expectancy (HALE) is life expectancy weighted or adjusted for the level of health-related quality of life (HRQOL). Cause-deleted probabilities of dying were derived using the cause-eliminated life table technique and death data from vital statistics for Canada in 1998/99. Life expectancy for men and women in Canada was 76.0 and 81.5 years respectively; HALE was 67.9 years for men and 71.1 years for women. Cancer represented the greatest burden of disease in the population, and eliminating it would increase men's life expectancy to 79.6 years and women's to 85.1 years. HALE would rise to 70.7 years for men and 73.6 for women. The gain in life expectancy would be very small if osteoarthritis were eliminated, but there would be an overall gain in HALE of approximately 1.0 years for men and 2.5 years for women. HALE estimated for chronic conditions using a utility-based measure of HRQOL from population health surveys should be regarded as a valuable component of population health surveillance.

Key words: burden of disease; health-adjusted life expectancy; health expectancy; health-related quality of life; health status; health utility index; life expectancy; morbidity

Introduction

In countries with high life expectancy, such as Canada, mortality is being delayed until older ages, and chronic diseases are causing illness and disability among those surviving.¹ To evaluate the likely effects of health interventions, it is important to capture two dimensions of health: quantity of life (mortality) and health-related quality of life (HRQOL) (morbidity). Summary measures of population health, which take into account both mortality and morbidity, are described as two major classes of measures: positive measures of health expectancy,^2,3 and measures of health gaps such as healthy life years⁴ or disability-adjusted life years (DALYs).⁵ Health expectancy, which is the focus of this study, estimates overall life expectancy or life years lived adjusted according to the amount of time spent in less than perfect health or with disability.⁶ 

The burden of specific conditions in a population can be estimated using either DALYs or cause-deleted health expectancy. Cause-deleted health expectancy estimates the increase in health expectancy if a specific cause did not exist in a population. It is calculated by removing both the deaths and the reduction in HRQOL attributable to a specific condition from the overall or all-cause mortality and HRQOL. The difference between cause-deleted health expectancy and current Canadian overall health expectancy is the "gap" in health resulting from the elimination of a condition - meaning that health expectancy is expressed as a health gap measure when reported as the difference between a reference (current Canadian) and potential health expectancy. 

In this study, we estimate cause-deleted health-adjusted life expectancy (HALE) for a number of chronic conditions in Canada for the period 1998/99. HALE refers to a health expectancy that is estimated using a utility-based measure of HRQOL. Utility-based measures assign a utility or value to the level of HRQOL, thereby making it easier to compare conditions with different HRQOL or mortality impact. We chose to estimate cause-deleted HALE over DALYs for three reasons. First, Canada has a relatively unique opportunity to estimate cause-deleted HALE with the availability of Statistics Canada population health surveys. Since 1994, the National Population Health Survey and the more recent Canadian Community Health Surveys collect information on both self-reported chronic conditions and HRQOL. These data allow for (ongoing) prevalence-based HRQOL assessment of chronic conditions. It is uncommon for population health surveys in other countries to include a utility-based measure of HRQOL, which is needed for HALE estimation.⁷ 

Second, health expectancy measures, like life expectancy, are expressed in intuitive terms - years of life or health - and, therefore, are helpful in describing the burden of disease to a wide audience. 

Finally, changes in health expectancy compared with life expectancy can be used to describe whether there is a "compression or expansion of morbidity". In the 1980s, Fries coined this term to describe this changing pattern of disease.⁸ He argued that an improvement in lifestyle would not only reduce death rates but would also slow the development of chronic diseases. This delayed onset, in turn, would lead to an increase in the proportion of life lived in a healthy state, or what he called "a compression of morbidity." Other authors have not been convinced by Fries's arguments, instead taking the view that increased medical care would lead to an "expansion of morbidity" because of an increase in survival without a change in the progression towards disability among the survivors.^1,9,10

Methods

Data sources

The Canadian Mortality Database was used to calculate the age-specific death rates, survival probabilities and life table estimates of life and health expectancy for the entire Canadian population in 1998/99.^11,12 Age-specific mortality estimates were calculated using adjusted Census population estimates from Statistics Canada.¹³ 

Data on health status and chronic conditions were derived from the 1998/99 National Population and Health Survey (NPHS 1998/99). The survey collected both cross-sectional and longitudinal data on household residents in all provinces (except people on Indian reserves, Canadian Forces bases and some remote areas in Quebec and Ontario) in 1998/99. There were two components to the interview, which was a computer-assisted telephone interview. The general component collected limited information on all members of the household who were 12 years and older; the health component, which is the component used in this study, was administered to one randomly selected member from each survey household for additional in-depth health information. 

For the first cycle (1994/95), a sample of approximately 20,000 households was drawn from the Labour Force Survey sampling frame. For Cycle 3 (1998/99), this frame was used to select an additional sample of recent immigrants and young children, thus ensuring that the data represent the 1998/99 Canadian population. The overall response rate was 88.2% at the household level. The response rate for the randomly selected respondents in these households was 98.5%.¹⁴ 

Variable definition and classification

Defining conditions: Disease groups for mortality statistics were defined using the ICD9 code for the most responsible underlying condition on the death certificate (see Table 1). Disease prevalence was estimated using the NPHS 1998/99 response for self-reported chronic conditions. This question asked respondents whether a health professional had ever diagnosed any of 24 chronic conditions. The presence of mental conditions was estimated using questions from the Composite International Diagnostic Interview (CIDI). Respondents were classified as having a mental condition if their CIDI score was 0.90 or higher, indicating that they had a high level of psychological distress.¹⁵ 

TABLE 1
Disease groups

Health-related quality of life measure (HRQOL): The HRQOL measure used to calculate HALE in this study was the Health Utilities Index (HUI3). The HUI3 is a utility-based, multi-attribute health classification system that estimates a summary value of individual health in which 0.0 = "dead" and 1.0 = "perfect health" (states worse than death are also possible), based on preference scores for different health states.¹⁶ Each respondent answered questions pertaining to eight attributes of functional health: vision, hearing, speech, mobility, dexterity, emotional state, cognition and level of pain and discomfort. Each attribute has from four to six possible levels, ranging from unrestricted to a highly disabled state (see Torrance¹⁷ for a description of health states). The eight attributes were then combined using preference scores from the mark III version according to the following multi-attribute utility function, where u is a HUI3 attribute:^18,19 

u = 1.371 * (u1 * u2 * u3 * u4 * u5 * u6 * u7 * u8) - 0.371 

Analysis methods

Life table analysis: Chiang's²⁰ method was used to calculate period life tables for 1998/99 for men and women in 20 standard age groups (< 1, 1-4, 5-9,…, 90+ years), except for an adaptation for the final age group.²¹ Cause-deleted life expectancy was calculated by subtracting the condition-deleted mortality rates from the overall mortality rates in the life table.²² 

HALE was calculated by means of a modified Sullivan method.²³ Sullivan used a period life table and the prevalence of disability to estimate the number of life years lived free of disability. After calculating life tables for each sex, we estimated HALE by weighting the years of life lived according to the age- and sex-specific mean HUI3 values. The cause-deleted mean HUI3 values were used to calculate cause-deleted HALE. Statistical error for life expectancy and health expectancy was calculated according to the method of Chiang and Mathers.^20,24 An example of the life tables used in this study are available in Microsoft Excel (http://www.ices.on.ca) . 

Cause-deleted HRQOL estimates: The cause-deleted methodology is based on the assumption that when a particular disease or condition is removed from the population, the pattern of morbidity and mortality in those without the disease/condition generalizes to the entire population.^25,26 

Cause-deleted mean HUI3 estimates were calculated in a manner similar to that of the cause-deleted mortality rates.²⁵ People with specific conditions were removed from the NPHS sample, and the mean HUI3 was estimated for each age-sex group. As the NPHS contains HUI3 scores for those over 12 years of age, the Canadian HUI3 estimates for age 12-15 were used for each of the age groups below 15 years old.

Results

Figure 1 illustrates the number of deaths and estimated prevalence (cases) for the various conditions in Canada in 1998/99 (total population 29.5 million). As expected, cancer and ischemic heart disease were responsible for the greatest number of deaths, although the number of prevalent cases was quite low. On the other hand, arthritis had the highest number of prevalent cases but resulted in few deaths. If one were to consider only the number of people affected by a condition, arthritis would have had the greatest population health impact.

FIGURE 1
Number of deaths and prevalence, chronic conditions, Canada 1998–1999

Data sources: Health Indicators 1999, Statistics Canada; 1996-97 Ontario Health Survey

The effect of a condition on HRQOL varied from one condition to another (Table 2). For example, Canadian women reporting the effect of stroke had a mean HUI3 of 0.57, whereas for those with chronic obstructive pulmonary disease it was 0.74. The mean HUI differences for conditions were smaller after age standardization, indicating that some conditions that have a large impact on HRQOL were more common in older people.

TABLE 2
Mean HUI3 by condition and sex, adjusted and unadjusted

*    Standardized to the 1996 Canadian population using the direct method.

^†    95% CI = 95% confidence interval

Table 3 shows that the life expectancy for men and women in Canada was 76.0 and 81.5 years respectively, and HALE was estimated to be 67.9 years for men and 71.1 years for women. All cancers represented the greatest burden of disease in the population, and eliminating them would have increased men's life expectancy to 79.6 from 76.0 years and women's to 85.1 from 81.5 years. HALE would rise to 70.7 years for men and 73.6 years for women. Eliminating ischemic heart disease had a similar effect on life expectancy but resulted in smaller gains in HALE compared with cancer.

TABLE 3
Cause-deleted life and health expectancy by disease group and sex

*    95% CI = 95% confidence interval

In Figure 2, the gain in life expectancy and HALE from the elimination of the various conditions is compared. This figure also provides a useful illustration of the compression or expansion of morbidity. Expansion of morbidity is evident when the years of HALE gained were less than the years of life expectancy gained, as in the case of cancer and heart disease. Compression of morbidity refers to a situation in which the proportion of life in less than perfect health decreases, or when the HALE gained is greater than the life expectancy gained. This was the case with osteoarthritis and mental conditions, for which the gain in life expectancy if these conditions were eliminated would be very small, but there would be an overall gain in HALE.

FIGURE 2
Gains in life expectancy (LE) and health-adjusted life expectancy (HALE) after eliminating conditions, Canada, 1998–1999

Discussion

This study used cause-deleted HALE to estimate the burden of disease from a number of chronic conditions through the use of a general population health survey, which contained questions on both a utility-based health status index and the presence of chronic conditions. 

As with previous studies, heart disease and cancer have the greatest impact on HALE because of the high death rates associated with them.^4,6,25,27,28 However, eliminating cancer would result in an expansion of morbidity. The cause-deleted approach assumes that the people surviving after a disease is eliminated will have the same health as the rest of the population. This may not be the case, depending on the approach taken to reduce the burden of disease. Reducing disease burden through prevention is thought to have a larger impact on HRQOL than on life expectancy because it will delay the onset of disabling disease.⁸ Since the outcome of current medical therapy is often improved HRQOL, secondary and tertiary care may also improve HRQOL more than life expectancy, resulting in a compression of morbidity for conditions such as ischemic heart disease and cancer. Evidence in Canada suggests that there has been a compression of morbidity in recent years.^6,29 

There are different methods of estimating the burden of health of chronic conditions using summary measures of population health (SMPH), most broadly defined as either DALYs or cause-deleted health expectancy. Deciding which method to use depends on the conceptual purpose of measuring disease burden and the sources of data that are available for their calculation.^30,31 Since Canada has the data sources necessary for estimating both types of measure, it is worth highlighting important method differences and relatively unique Canadian opportunities. Most importantly, DALYs are generally described as incidence-based measures of HRQOL impact, as compared with a prevalence-based method that is most commonly used in health expectancy measures. 

Incidence- and prevalence-based indicators measure different things, and which is more appropriate to use depends on the application. Incidence measures are generally regarded as useful for monitoring the trends of disease occurrence and, therefore, measure the progress towards disease prevention. Health expectancy measures the current impact of disease, which in turn is the combined influence of mortality and either past incidence and duration or current prevalence of disease conditions. Therefore, HALE, using Canadian population health surveys and mortality data, estimates the current overall impact of conditions on health, and cause-deleted HALE estimates the long-term consequence of eliminating or reducing specific conditions. 

DALYs typically use information for estimating condition incidence and HRQOL impact from different sources. If a broad definition of a condition is used to estimate incidence but a narrower (typically more severe) definition is used to estimate HRQOL impact, combining the two estimates would result in an overestimate of disease impact. In addition, HRQOL impact is usually approximated through a process of expert and lay panels that review different sources of epidemiologic evidence.

Population health surveys can be used to estimate the current Canadian HRQOL impact and the prevalence of different chronic conditions, reflecting the same definition of the condition for prevalence and HRQOL assessment without the need for other epidemiologic evidence or a panel ranking process. As an example of the benefit of using Canadian data, consider what would happen to HRQOL burden if a new medication that dramatically improved pain and mobility were to be widely introduced into Canada for the treatment of older patients with osteoarthritis. The ongoing Canadian Community Health Surveys and the National Population Health Surveys would capture the current improvement. The DALY method, as commonly derived, would require further epidemiologic methods and expert opinion to re-adjust disability weights and the incidence in different populations of disease severity to reassess disease burden. Without adjusting disability weights or severity levels for different ages, the DALY method may not appropriately adjust for the HRQOL effect of the medication introduction in older people. As there are many factors that affect disease burden (such as socio-economic conditions, physical and social environment, medical therapies, health risk behaviour) in different populations it would seem improbable that the current DALY approach could reflect the actual disease burden in any one population. 

Canada's population health surveys have several additional benefits. Measures of HRQOL can be combined with other survey components, such as sociodemographic and behavioural characteristics, to estimate HALE based on different factors. The NPHS has used this approach to estimate health expectancy based on socio-economic and smoking status together with other factors.³² The population health surveys allow for adjustment of comorbidity (defined as the effect of a person's HRQOL as influenced by other chronic conditions). 

The methodology of many studies, particularly those using dichotomous measures of disability and the WHO DALY method, assumes that eliminating a condition results in a non-disabled state (perfect health) regardless of age.^{25,27,28,33,34} We controlled for comorbidity by assuming that elimination of a condition would result in HRQOL equal to that of people of the same age without the condition. However, it is important to note that in our study we assumed that the HRQOL level of people reporting a condition was attributable only to that condition, even if a person had more than one chronic condition. This means that our estimates were not mutually exclusive between conditions, although it was possible to adjust for comorbidity arising from multiple conditions, as other studies have, since the NPHS captured information on the presence of multiple chronic conditions.^6,32 In the same manner, the recent addition of routinely recording multiple causes of death in vital statistics will allow different approaches to adjust for comorbidity or defining the cause of death in the mortality component of SMPH.³⁵

Schultz and Kopec have shown that comorbid conditions are common at older ages and, therefore, influence HUI3 estimates, although the rank order of HUI3 level does not change between conditions if comorbidity is considered.³⁶ This means that gains in cause-deleted HALE (and, potentially, cause-deleted life expectancy) would be smaller if comorbidity were considered, but the rank order of disease burden would likely not change. 

Increasingly, Canada's health surveys should not be considered as isolated sources of health data but, rather, as a family of cross-sectional and longitudinal surveys that can be linked by individual respondent to other sources of data. Repeated cross-sectional samples allow for the surveillance of health expectancy measures over time. Longitudinal surveys facilitate the assessment of disease incidence and/or the development of hybrid summary measures of population that consider both disease incidence and prevalence.³⁷ Similarly, the large selection of HRQOL measures in the health surveys can be used in the development of weights for Canadian DALY disability estimates. 

There are several important limitations to our study. Reliance on respondents' self report in health interviews that contain either an open-ended question or a checklist of chronic conditions may bias results. Compared with medical examinations and disease registries, self reports often underreport chronic conditions.^38-41 However, since the survey can be directly linked to disease registries and health care data, reporting bias can be overcame by ascertaining condition status (incidence or prevalence) using these alternative sources of information. This approach was used to estimate HALE and cause-deleted HALE for people with diabetes in Ontario.⁴² 

An additional important limitation of the study was the exclusion of people in institutions. Berthelot et al. have shown that this population would reduce population HRQOL utility estimates by up to 30% for women in the oldest age groups.⁴³ On the basis of their findings, the overall HALE estimates would be about 0.6 to 0.8 years lower if institutionalized people were included (calculations not shown). The bias resulting from excluding this population may be appreciably higher for conditions such as stroke, which are overrepresented in institutions. 

Conclusions

Population health surveys with a utility-based health status measure should be regarded as a valuable component of population health surveillance, as they can describe the incremental differences in HRQOL between conditions using fewer assumptions about the relation with age, sex or level of severity. As such, these surveys are well suited to describe the health status of a population for selected conditions - a product of all the health influences of that disease, from health promotion to palliation. The greatest limitation of health surveys for this purpose is the inherent difficulty in estimating disease prevalence based on self report. Opportunities exist to overcome this limitation by linking population health surveys with other health data better suited to estimate the prevalence/incidence of several conditions. These data sources introduce a number of other applications for improved and expanded surveillance of the burden of conditions in different populations.

Acknowledgements

Data were provided to Health Canada from the Canadian Vital Statistics databases at Statistics Canada. The cooperation of the provincial and territorial vital statistics registries, which supply the data to Statistics Canada, is gratefully acknowledged. Dr. Manuel receives Career Scientist support from the Ontario Ministry of Health and Long-term Care. 

References

1. Olshanksy S, Rudberg MA, Carnes BA. Trading off longer life for worsening health. J Aging Health 1991;3:216. 

2. Mathers CD, Robine J-M. Health expectancy indicators: a review of the work of REVES to date. In Mathers C, Robine J-M, eds. Calculation of health expectancies: harmonization, consensus achieved and future perspectives. London: John Libbey and Company Ltd., 1993:1-21. 

3. Robine JM, Romieu I, Cambois E. Health expectancy indicators. Bull WHO 1999;77: 181-5. 

4. Hyder AA, Rotllant G, Morrow RH. Measuring the burden of disease: healthy life-years. Am J Public Health 1998;88: 196-202. 

5. Murray CJ, Lopez AD. The global burden of disease: a comprehensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020. Washington: Harvard School of Public Health on behalf of the World Health Organization and the World Bank, 1996. 

6. Manuel D, Schultz S, Kopec J. Measuring the health burden of chronic disease and injury using health-adjusted life expectancy and the Health Utilities Index. J Epidemiol Community Health 2002;56: 843-50. 

7. Sadana R, Mathers CD. Lopez AD, Murray CJL, Iburg K. Comparative analyses of more than 50 household surveys on health status. evidence and information for policy discussion paper no. 15, 1-77. Geneva: World Health Organization. GPE Discussion Paper, 2000. 

8. Fries JF. The compression of morbidity. Milbank Mem Fund Q Health Soc 1983; 61:397-419. 

9. Vaupel JW, Gowan AE. Passage to Methuselah: some demographic consequences of continued progress against mortality. Am J Public Health 1986;76:430-3. 

10. Gruenberg E. The failures of success. Milbank Mem Fund Q 1977;55:3-24. 

11. Statistics Canada. Causes of death. Shelf tables. Ottawa : Statistics Canada, 1998, Catalogue 84F0208XPB. 

12. Statistics Canada. Causes of death. Ottawa: Statistics Canada, Health Statistics Division, 1999, Catalogue 84F208XPB. 

13. Statistics Canada. Annual demographic statistics 2001. Ottawa: Statistics Canada, Demography Division, 2002, Catalogue 91-213-XPB. 

14. Millar WJ. Patterns of use - alternative health care practitioners. Health Rep 2001;13:9-21. 

15. Rhodes AE, Lin E, Mustard CA. Self-reported use of mental health services versus administrative records: should we care? Int J Methods Psychiatr Res 2002;11:125-33. 

16. Feeny D, Furlong W, Boyle M, Torrance GW. Multi-attribute health status classification systems. Health Utilities Index. Pharmacoeconomics 1995;7:490-502. 

17. Torrance G, Furlong W, Feeny D, Boyle M. Multi-attribute preference functions - Health Utilities Index. Pharmacoeconomics 1995; 7:503-20. 

18. Furlong W, Feeny D, Torrance G, Barr R. The Health Utilities Index (HUI3) system for assessing health-related quality of life in clinical studies. Health Economics and Policy Analysis Research Working Paper, #01-02, 2001. 

19. Feeny D, Furlong W, Torrance GW, Goldsmith CH, Zhu Z, DePauw S et al. Multiattribute and single-attribute utility functions for the health utilities index mark 3 system. Med Care 2002;40:113-28. 

20. Chiang CL. The life table and its applications. Malabar, Florida: Robert E. Krieger Publ. Co., 1984. 

21. Hsieh J. A general theory of life table construction and a precise abridged life table method. Biomed J 1991;2:143-62. 

22. Newman SC. Formulae for cause-deleted life tables. Stat Med 1987;6:527-8. 

23. Sullivan DF. A single index of mortality and morbidity. HSMHA Health Reports 1971;86:347-54. 

24. Mathers C, Vos T, Stevenson C. The burden of disease and injury in Australia. Canberra: Australian Institute of Health and Welfare, 1999, PHE17. 

25. Mathers CD. Gains in health expectancy from the elimination of diseases among older people. Disabil Rehabil 1999;21: 211-21. 

26. Muennig PA, Gold MR. Using the years-of-healthy-life measure to calculate QALYs. Am J Prev Med 2001;20:35-9. 

27. Nusselder WJ, van der Velden K, van Sonsbeek JL, Lenior ME, van den Bos GA. The elimination of selected chronic diseases in a population: the compression and expansion of morbidity. Am J Public Health 1996;86:187-94. 

28. Picavet HS, van den Bos GA. The contribution of six chronic conditions to the total burden of mobility disability in the Dutch population. Am J Public Health 1997;87: 1680-2. 

29. Belanger A. Report on the demographic situation in Canada 1998-1999 (current demographic analysis). Ottawa: Statistics Canada, 1999, 91-209-XPE. 

30. WHO. Summary measures of population health: concepts, ethics, measurement and applications. Geneva: World Health Organization, 2002. 

31. Robine J-M, Jagger C, Mathers CD, Crimmins EM, Suzman RM. Determining health expectancies. Chichester, UK: John Wiley & Sons Ltd., 2003. 

32. Belanger A, Martel L, Berthelot JM, Wilkins R. Gender differences in disability-free life expectancy for selected risk factors and chronic conditions in Canada. J Women Aging 2002;14:61-83. 

33. Bonneux L, Barendregt JJ, Nusselder WJ, der Maas PJ. Preventing fatal diseases increases healthcare costs: cause elimination life table approach. Br Med J 1998;316: 26-9. 

34. Mathers C. Estimating the gains in health expectancy due to the elimination of specific diseases. Paper presented at the fifth meeting of the International Network on Health Expectancy (REVES-5). Ottawa, Canada, 1992. 

35. Wilkins K, Parsons GF, Gentleman JF, Forbes WF. Deaths due to dementia: an analysis of multiple-cause-of-death data. Chronic Dis Can 1999;20:26-35. 

36. Schultz SE, Kopec JA. Impact of chronic conditions. Health Rep 2003;14(4):41-53. 

37. Barendregt JJ. Incidence- and prevalence-based SMPH: making the twain meet. In: Murray CJL, Salomon J, Mathers CD, Lopez AD, eds. Summary measures of population health: concepts, ethics, measurement and applications I. Geneva: World Health Organization, 2002:221-2. 

38. Beckett M, Weinstein M, Goldman N, Yu-Hsuan L. Do health interview surveys yield reliable data on chronic illness among older respondents? Am J Epidemiol 2000; 151:315-23. 

39. Gross R, Bentur N, Elhayany A, Sherf M, Epstein L. The validity of self-reports on chronic disease: characteristics of underreporters and implications for the planning of services. Public Health Rev 1996;24:167-82. 

40. Heliovaara M, Aromaa A, Klaukka T, Knekt P, Joukamaa M, Impivaara O. Reliability and validity of interview data on chronic diseases. The Mini-Finland Health Survey. J Clin Epidemiol 1993;46:181-91. 

41. Jabine TB. Reporting chronic conditions in the National Health Interview Survey. A review of findings from evaluation studies and methodological test. Vital Health Stat 2 1987;1-45. 

42. Manuel DG, Schultz SE. Diabetes health status and risk factors. In: Hux JE, Booth G, Slaughter PM, Laupacis A, eds. Diabetes in Ontario: an ICES practice atlas. Toronto: Institute for Clinical Evaluative Sciences, 2003:4.77-4.94. 

43. Berthelot J-M, Roberge R, Wolfson M. The calculation of health-adjusted life expectancy for a Canadian province using a multi-attribute utility function: a first attempt. In: Robine JM, Mathers C, Bone I, Romieu I, eds. Calculation of health expectancies: harmonization, consensus achieved and future perspectives. London: John Libbey and Company Ltd., 1993.

Author References

Douglas G Manuel, Ontario Ministry of Health and Long-Term Care; Institute for Clinical Evaluative Sciences, Toronto, Ontario; Department of Public Health Science, University of Toronto, Toronto, Ontario, Canada 

Wei Luo, Surveillance and Risk Assessment Division, Public Health Agency of Canada, Health Canada, Ottawa, Ontario, Canada 

Anne-Marie Ugnat and Yang Mao, Surveillance and Risk Assessment Division, Public Health Agency of Canada, Health Canada; Department of Community Health and Epidemiology, University of Ottawa, Ottawa, Ontario, Canada 

Correspondence: Dr. Douglas G Manuel, Institute for Clinical Evaluative Sciences, G106-2075 Bayview Avenue, Toronto Ontario, Canada M4N 3M5;
Fax: (416) 480-6048; E-mail: doug.manuel@ices.on.ca

[Previous] [Table of Contents] [Next]