Mortality from Chronic Obstructive Pulmonary Diseases and Asthma in France, 1969-1983
1990; Elsevier BV; Volume: 97; Issue: 1 Linguagem: Inglês
10.1378/chest.97.1.213
ISSN1931-3543
AutoresJ Cooreman, Thomas J. Thorn, Millicent Higgins,
Tópico(s)Health Promotion and Cardiovascular Prevention
ResumoThis article focuses on international similarities and differences in levels and trends of mortality for chronic obstructive pulmonary disease (COPD) and asthma in the US, Canada and France from 1969 to 1983. Comparisons have been made of national vital statistics data for age groups 55-64 years, 65-74 years and 75-84 years. From 1969 to 1978, under the 8th revision of the International Classification of Diseases (ICD), the COPD and asthma category included the codes 490-493 and, for the US and Canada, a special code 519.3. From 1979, under the 9th revision, COPD and asthma codes 490-493, 496 were in use in all three countries. The analyses of US, Canadian and French data show lower death rates for COPD and asthma in France. The ratio of male to female deaths from COPD increased with age in the US and Canada, but not in France. The proportion of COPD and asthma deaths attributed to bronchitis was higher in France. An increased use of code 496 (under 9th revision) was observed in the US and above all in Canada. In the three countries, death rates increased faster between 1979 and 1983 in women than in men and increases in women were steeper in the US and Canada than in France. Intercountry comparability is better for COPD mortality in the 9th revision than the 8th revision. Some differences observed between the three countries can be partly explained by coding practices and ICD revisions, but, allowing for differences in coding and classifying repiratory causes of death as well as ICD revisions, death rates are obviously higher in the US and Canada than in France. This suggests that the difference is real.(Chest 1990; 97:213-19) This article focuses on international similarities and differences in levels and trends of mortality for chronic obstructive pulmonary disease (COPD) and asthma in the US, Canada and France from 1969 to 1983. Comparisons have been made of national vital statistics data for age groups 55-64 years, 65-74 years and 75-84 years. From 1969 to 1978, under the 8th revision of the International Classification of Diseases (ICD), the COPD and asthma category included the codes 490-493 and, for the US and Canada, a special code 519.3. From 1979, under the 9th revision, COPD and asthma codes 490-493, 496 were in use in all three countries. The analyses of US, Canadian and French data show lower death rates for COPD and asthma in France. The ratio of male to female deaths from COPD increased with age in the US and Canada, but not in France. The proportion of COPD and asthma deaths attributed to bronchitis was higher in France. An increased use of code 496 (under 9th revision) was observed in the US and above all in Canada. In the three countries, death rates increased faster between 1979 and 1983 in women than in men and increases in women were steeper in the US and Canada than in France. Intercountry comparability is better for COPD mortality in the 9th revision than the 8th revision. Some differences observed between the three countries can be partly explained by coding practices and ICD revisions, but, allowing for differences in coding and classifying repiratory causes of death as well as ICD revisions, death rates are obviously higher in the US and Canada than in France. This suggests that the difference is real. (Chest 1990; 97:213-19) Respiratory diseases are among the leading causes of morbidity and mortality in many countries. Chronic obstructive pulmonary disease (COPD) is the most common respiratory cause of death and is mostly attributable to smoking.1US Department of Health and Human Services. The health consequences of smoking: Chronic obstructive lung disease. Rockville, Maryland: Report of the Surgeon General, 1984, (DHHS publication [PHS] 84-50205Google Scholar,2Fielding JE Smoking. Health effects and control (first of two parts).N Engl J Med. 1985; 313: 491-498Crossref PubMed Scopus (277) Google Scholar The mortality trend for COPD is of particular interest because, in the United States, and probably in other countries, trends are upward in contrast to sharply downward trends for heart disease, stroke, and other chronic diseases.3Abrams HK Aggravation of lung disease.Scand J Work Environ Health. 1984; 10: 487-493Crossref PubMed Scopus (4) Google Scholar,4Goldman L Cook EF The decline in ischemic heart disease mortality rates. An analysis of the comparative effects of medical interventions and changes in lifestyle.Ann Intern Med. 1984; 101: 825-836Crossref PubMed Scopus (413) Google Scholar Assessment of international differentials in levels and trends for COPD has been complicated, however, not only by intercountry differences in cause-of-death classification and coding practices, but also by revisions of the International Classification of Diseases which the World Health Organization uses to collect and report mortality statistics. This article compares recent mortality trends for COPD and the related condition, asthma, in France, the United States (US) and Canada addressing these complicating factors and the possible role of cigarette smoking in the mortality differentials. Information is provided for ages 55 to 84 where mortality from COPD is common. Mortality data for COPD and asthma described in this study are from Vital Statistics of Canada, the US National Center for Health Statistics, L'Institut National de la Santé et de la Recherche Médicale for France, and issues of World Health Statistics Annual of the World Health Organization (WHO). Comparisons of death rates from 1969 to 1983 are made between the US, France and Canada for age groups 55-64 years, 65-74 years and 75-84 years. For years prior to 1974, data from Canada were not available. The focus is on mortality from COPD. Brief mention is made of trends for asthma because in adults the distinction between these diseases as causes of death is not always clear. Indeed, the WHO combines them in its reports on mortality by country. The diagnostic categories for COPD and asthma used between 1969 and 1978, in the 8th Revision of the International Classification of Diseases (ICD), are given in Table 1, which also shows the numbers of deaths in 1978 in the three countries. Inclusion of “other diseases of the respiratory system” is discussed below. During the 8th revision, there was no category in France for COPD not specified as bronchitis or emphysema, but the US and Canada introduced a special code (519.3), “chronic obstructive lung disease without mention of asthma, bronchitis or emphysema” (COLD). As more and more US and Canadian death certificates contained nonspecific terminology for COPD instead of bronchitis or emphysema, there was a large increase in the proportion of COPD and asthma deaths classified as COLD, as well as an increase in the total. By 1978, there were more deaths attributed to COLD than to emphysema and bronchitis combined in the US. Code 519.3 comprised 57 percent and 38 percent, respectively, of all COPD and asthma deaths in the US and Canada. The effects of this on mortality comparisons with France will be described.Table 1Distribution of deaths from COPD and Asthma in men and women all ages. United States, Canada and France in 1978 and 1979United StatesCanadaFranceCause of death197819791978197919781979COPD and Asthma100.0100.0100.0100.0100.0100.0(Numbers of deaths)50,49049,9334,7294,6207,24111,076Bronchitis unqualified (490)*From 1979: Bronchitis, not specified as acute or chronic.1.50.83.41.319.812.3Chronic bronchitis (491)7.16.524.920.042.025.7Emphysema (492)31.027.227.123.817.19.2Asthma (493)3.75.26.48.621.113.6Chronic obstructive lung disease without mention of asthma, bronchitis or emphysema (519.3)56.7—38.2—†This diagnostic term does not exist in the mortality statistics of France.—Other diseases of the respiratory system (519.9)n = 819—n = 51—n = 5,354—Chronic airways obstruction not elsewhere classified (496)‡In US tabulations, this category also includes deaths from bronchiectasis (494) and extrinsic allergic alveolitis (495) which accounted for only 703 and 10 deaths respectively in 1979.—60.3—46.3—39.2NOTE: Numbers in parentheses are codes of the International Classification of Diseases.* From 1979: Bronchitis, not specified as acute or chronic.† This diagnostic term does not exist in the mortality statistics of France.‡ In US tabulations, this category also includes deaths from bronchiectasis (494) and extrinsic allergic alveolitis (495) which accounted for only 703 and 10 deaths respectively in 1979. Open table in a new tab NOTE: Numbers in parentheses are codes of the International Classification of Diseases. Beginning in 1979, the ICD/9 codes for COPD and asthma were 490-492 and 493 respectively. Death certificates containing generalized terminology such as COLD were classified to code 496, “chronic airways obstruction, not elsewhere classified” (NEC). By basing COPD and asthma mortality for years 1979-83 on ICD/9 codes 490-493,496 (490-496 for the US), comparison is enhanced over that possible from WHO publications which omit deaths classified to chronic airways obstruction (496). The total number of deaths from COPD and asthma in France increased from 7,247 in 1978 to 11,076 in 1979 because of the exclusion of a generalized COPD category in ICD/8 and inclusion of code 496 in ICD/9. The number of deaths from chronic bronchitis, emphysema, or asthma did not change appreciably in 1979 as compared with 1978. In an attempt to improve comparability of time trends in France across these two ICD revisions, mortality for the category 519.9 “Other diseases of the respiratory system,” for 1969 to 1978 is also analyzed. This is a large category of mortality in France and it presumably contains deaths certified to generalized COPD. COPD and asthma death rates from 1969 to 1983 for the three countries for the 65-74 years age group, the middle age-group of our study, are shown in Table 2. Intercountry comparisons for this age group typify comparisons for the other age groups. The proportion of COPD and asthma deaths due to bronchitis is much higher in France (29 percent in 1983) than in the US (5 percent) and Canada (9 percent). There is greater mortality coded to the term chronic airways obstruction NEC in Canada and the US than in France between 1979 and 1983. Between 1978 and 1979, there is no sudden increase in mortality from COPD and asthma in the US and Canada.Table 2Death Rates (per 100,000 population) from COPD and Asthma Age Group 65-74, in the US, Canada, and Bronce, 1969-1983YearsBronchitis 490-491Emphysema 492CAO-NEC*Chronic airways obstruction not elsewhere classified: for US=519.3 until 1978,494-496 from 1979; for Canada=519.3 until 1978, 496 from 1979; for France = 496 from 1979.Asthma 493TotalUS196914.969.18.03.895.8197014.468.713.33.599.9197113.965.517.83.7100.9197215.165.223.03.8107.1197313.163.030.83.1110.0197412.055.638.73.2109.5197510.851.245.73.2110.9197610.247.054.33.4114.919779.641.660.23.0114.419789.338.770.83.0121.819797.633.671.44.5117.019807.633.583.14.9129.119817.232.687.75.1132.719826.530.089.84.9131.219836.630.397.85.1139.8Canada197437.938.726.67.3110.6197536.138.030.88.3113.2197636.734.440.65.3117.0197730.932.142.55.2110.7197832.934.947.76.1121.7197921.127.454.48.3111.1198019.728.454.17.4109.6198119.230.552.57.0109.2198214.726.567.58.5117.2198310.725.570.78.1115.0France196918.512.4—12.243.0197016.59.9—10.036.4197118.111.7—11.241.0197223.010.4—10.443.8197323.89.5—10.844.0197424.18.2—9.742.0197524.59.4—9.443.3197623.78.3—9.841.8197721.47.8—9.138.3197821.38.0—8.938.2197918.86.324.98.458.4198019.06.128.78.562.3198118.36.329.99.363.8198218.56.330.39.064.1198318.05.432.07.462.7NOTE: ICD/8 and ICD/9 codes are in parentheses except as noted.* Chronic airways obstruction not elsewhere classified: for US=519.3 until 1978,494-496 from 1979; for Canada=519.3 until 1978, 496 from 1979; for France = 496 from 1979. Open table in a new tab NOTE: ICD/8 and ICD/9 codes are in parentheses except as noted. Because of the lack of a special category (519.3) for COLD during ICD/8 in France, however, there is a sudden increase, as deaths classified as chronic airways obstruction (496) are included beginning with ICD/9 in 1979. Apart from this sudden increase, trends for total COPD and asthma are markedly upward in the US, modestly upward in Canada, but not upwards in France. For the total population (all ages), between 1979 and 1983, deaths coded chronic airways obstruction NEC increased in frequency and dominated the COPD and asthma category (Fig 1), in the US and Canada but not in France. Age-adjusted death rates for COPD and asthma by sex in France from 1969 to 1983 for the broad age range 55 to 84 years are shown in Figure 2. For each cause, death rates have been adjusted to the age distribution of the French population of 1968, a census year. COPD and asthma mortality rates were relatively constant during 1969-1978. The marked increases between 1978 and 1979 coincide with beginning of the 9th revision, which introduced the separate tabulation of deaths from chronic airways obstruction NEC. The main respiratory cause of death (almost 1/3) is bronchitis, especially in men. Death rates for bronchitis increased in men and women at the beginning of the study period. On the contrary, mortality from emphysema (for both sexes) and asthma (for men) decreased during the entire study period. Trends in France by sex for the three age groups are shown in Figure 3. Mortality is noticeably higher in men than women. In each age group, the curves for men are at the same level as for women ten years older. The same sharp increase (as in Fig 2) due to revision in cause-of-death coding is observed between 1978 and 1979. Trends were flat prior to 1979, but slightly upward since then. Age-sex specific death rates for COPD and asthma for the three countries, from 1979 to 1983, are presented in Figure 4. Male/female differentials in mortality are of similar magnitude among the three countries. However, with respect to trends, there are much steeper increases for women than men in the US and Canada so that the differentials are narrowing in those two countries, but not in France. The US and French death rates for COPD and asthma including mortality for the special ICD/8 code 519.3 for the US and the closest comparable ICD/8 code for France, (519.9), other diseases of the respiratory system, are shown in Figure 5. Even with inclusion of this code the US curves remain above those for France, especially for women.Figure 5Death rates for COPD and asthma by age and sex, US and France, 1969-1983.View Large Image Figure ViewerDownload (PPT) The male/female ratios of COPD and asthma death rates separately by age group in the US, Canada and France are presented in Table 3. For asthma, the ratios are nearly constant, but higher in France than in the US, especially before 1979. For COPD, the sex ratios rise with age in the US and Canada, but not in France. In the US and Canada, the sex ratios were higher in the earlier years. For the three countries, the sex differences at each age are much smaller for asthma than COPD, especially in recent years. These data suggest that some deaths which are attributed to asthma in French men would be attributed to COPD in the US.Table 3Male/Female Ratios for COPD and Asthma for U.S. Canada and France 1969-1983USCanadaFranceCOPD*ICD/8 code 490-492, 519.3; ICD/9 code 490-492, 494-496Asthma‡code 493COPD*ICD/8 code 490-492, 519.3; ICD/9 code 490-492, 494-496Asthma‡code 493COPD†ICD/8 code 490-492; ICD/9 code 490-492, 496Asthma‡code 49355-64 years19695.131.00Data not available7.761.7619704.540.89Data not available8.212.3519714.400.81Data not available7.321.6919724.060.86Data not available6.261.9219733.760.82Data not available5.902.0819743.320.913.341.405.341.4719753.090.914.101.156.211.3619763.010.794.600.746.331.2719772.800.933.660.896.911.1019782.450.593.381.835.050.9419792.440.923.020.974.551.1719802.300.883.420.845.071.0019812.200.902.991.015.381.0919822.100.792.950.694.431.1719832.100.792.730.834.820.9865-74 years19697.871.33Data not available4.931.6819707.061.12Data not available5.401.9619714.761.15Data not available5.571.9019725.931.17Data not available5.271.9519735.631.21Data not available5.631.8519745.211.137.111.945.311.7719754.961.245.711.105.811.7919764.621.195.012.485.791.7619774.211.034.791.485.471.3019783.851.004.321.506.201.5519793.701.174.571.194.801.5019803.261.254.560.975.021.3319813.050.944.291.405.101.2519822.790.964.090.945.051.4619832.580.983.580.964.740.7975-84 years19697.421.66Data not available3.631.7819707.761.30Data not available3.501.7419717.091.17Data not available4.101.6919727.301.25Data not available4.381.6619736.891.18Data not available4.231.4719746.461.347.331.774.181.7819756.551.076.902.394.521.4519765.891.146.300.884.531.8519775.811.026.091.394.501.6419785.311.096.171.134.381.3619795.101.216.561.574.451.2619804.641.476.051.874.031.2919814.251.165.901.544.291.4819824.111.105.421.274.221.3519833.831.055.551.584.731.21* ICD/8 code 490-492, 519.3; ICD/9 code 490-492, 494-496† ICD/8 code 490-492; ICD/9 code 490-492, 496‡ code 493 Open table in a new tab Analysis of US, Canadian and French mortality data shows lower death rates for COPD and asthma in France and dissimilar patterns by sex among these countries. The proportion of deaths from bronchitis was higher in France, while an increased use of generalized terms for COPD in death certificates was observed in Canada and the US. The likely reason is differing diagnostic fashion and coding practice. Death rates for COPD increased faster in women than in men in the US and Canada; the difference was less in France. Unlike the US and Canada, where COPD and asthma mortality generally increased over the 1969 to 1983 period, especially in women, trends in France were relatively flat and were lower than rates in the US and Canada. For the 1969-78 period, the trend would also be downwards in the US and Canada if it were not for inclusion of the special category COLD. By including the category ICD/8 code 519.9 in COPD and asthma mortality for France between 1969 and 1978, comparability with Canadian and US mortality is still not very good. Using a broad diagnostic category to define COPD and asthma in France (Fig 5), death rates in middle-aged men and women (age 55-84) are clearly lower in France than in the US (and Canada), and trends are generally level in France, upwards in the US and Canada. The reliability of any comparison of mortality data depends on several factors. Four are very important: accuracy of diagnosis, death certificate entries by physicians, coding practices, and changes in the ICD. In each country it can be difficult for a physician to choose between similar diagnoses and to specify underlying or contributory cause of death, and the coding practices are not simple. This is why, in some countries, the doctor is contacted if there is doubt about a code. This is being tested in France now.5Rumeau-Rouquette C Bréart G Padieu R Statistiques sanitaires.in: Méthodes en Epidémiologie. 3rd ed. Flammarion Médecine-Science ed, Paris1985: 22-36Google Scholar In addition, death certification varies among countries and certificates of death are not actually standardized. Some comparisons were made in the European Economic Community of coding by the usual national center and by a reference center in London. Large differences in the coded causes of death were found within and between countries. For instance, there was a net overestimate of 13 percent in deaths from asthma in the United Kingdom.6Kelson MC Heller RF The effect of death certification and coding practices on observed differences in respiratory disease mortality in 8 EEC countries.Rev Epidemiol Santé Publique. 1983; 31: 423-432PubMed Google Scholar, 7Neukirch F Maguin P Perdrizet S Pariente R Validité des données de mortalité par maladies respiratoires en France et dans sept autres pays de la CEE.Rev Mal Respir. 1984; 1: 361-367PubMed Google Scholar, 8Subcommittee of the BTA Research Committee. Accuracy of death certificates in bronchial asthma. Accuracy of certification procedures during the confidential inquiry by the British Thoracic Association. Thorax 1984; 39:505-09Google Scholar The most common error was failure to follow the procedure advised for completion of death certificates. Revisions in the ICD also complicate these comparisons, especially when related diseases are reported on the same certificate. For instance, under the 8th revision, the deaths certified as being due to asthma with mention of bronchitis, bronchiolitis, or emphysema were to be attributed to the other condition. Under the 9th revision, deaths certified as being due to asthma, whether or not other conditions coexisted, are to be attributed to asthma. Nevertheless, some authors9Alderson M, Ashwood F. Projection of mortality rates for the elderly. Medical Statistics Division, OPCS: 22-29Google Scholar believe there is a very limited discontinuity in trends in mortality statistics for chronic bronchitis. They show that a dual coding of a sample of deaths in 1978 suggests that deaths ascribed to chronic bronchitis under the 9th revision would have formed 97 percent of those coded to the equivalent codes under the 8th revision. Intercountry comparability is better for mortality in the 9th revision of the ICD than in the 8th revision. Unfortunately, only a few years of time trend data are available to date and it will be necessary to continue this observation during a longer period. Results of the present analysis suggest that it is necessary to extend mortality tabulations beyond ICD/9 codes 490-493 by including mortality from chronic airways obstruction (code 496) when comparing COPD and asthma mortality among countries. That gives the most comparable COPD and asthma category possible for 1979 and beyond. International comparisons of COPD and asthma mortality should not rely only on mortality reported in WHO publications. To do so would repeat the error that occurred in the 1969 to 1978 trend in the US under the 8th revision: without the special code 519.3 for COLD, a decrease in COPD mortality was observed principally in men, whereas with inclusion of this code, there was a marked increase in mortality from COPD. In each country, COPD death rates are higher in men than in women, but male/female differentials are relatively small in France. According to a number of studies, men have an enhanced risk of developing bronchitis and airways obstruction and their risk of bronchitis is reported in some investigations to be more than three times greater than for women.10Tockman MS Khoury MJ Cohen BH The epidemiology of COPD.in: Lenfant C Chronic obstructive pulmonary disease. Marcel Dekker, New-York1985: 43-92Google Scholar After cigarette smoking and age, “male sex is the risk factor most closely associated with the presence of airways obstruction.”10Tockman MS Khoury MJ Cohen BH The epidemiology of COPD.in: Lenfant C Chronic obstructive pulmonary disease. Marcel Dekker, New-York1985: 43-92Google Scholar Between 1969 and 1983, COPD and asthma death rates increased faster in women than in men for the three age groups studied in the US and Canada. The changes in smoking habits of women could be responsible, at least in part. In France, the proportion of adult women who smoke increased for the generations studied.11Les consommateurs de boissons alcooliques et de tabac. Enquête sur la santé et les soins médicaux 1980-1981. Ministère des Affaires Sociales et de la Solidarité Nationale. Solidarité, Santé, Cahiers statistiques 1984; 1:109Google Scholar The prevalence of smoking in the US is described in the 1985 Surgeon General's Report where the estimate for American women was 18 percent in 1935 and about 30 percent between 1965 and 1979.12US Department of Health and Human Services. The health consequences of smoking for women. Rockville, Maryland: Report of the Surgeon General, 1985 (DHHS publication n° [PHS] 85-470-822Google Scholar The Surgeon General added “the health consequences of this enhanced exposure to cigarette smoke among women are likely to be more prominent in the coming decades.” French women began to smoke later than American women and the French Department of Health estimated that French women (15 years old or more) usually smoking in 1980/81 was 16.3 percent,11Les consommateurs de boissons alcooliques et de tabac. Enquête sur la santé et les soins médicaux 1980-1981. Ministère des Affaires Sociales et de la Solidarité Nationale. Solidarité, Santé, Cahiers statistiques 1984; 1:109Google Scholar but this proportion was 35 percent in the youngest age groups. Among men, the proportion of smokers in the US12US Department of Health and Human Services. The health consequences of smoking for women. Rockville, Maryland: Report of the Surgeon General, 1985 (DHHS publication n° [PHS] 85-470-822Google Scholar decreased regularly from 52.5 percent in 1936 to 36.9 percent in 1979. It does not seem reasonable to think that a similar decrease existed in France, since the proportion of men who smoked was 43.5 percent in 1980/81.11Les consommateurs de boissons alcooliques et de tabac. Enquête sur la santé et les soins médicaux 1980-1981. Ministère des Affaires Sociales et de la Solidarité Nationale. Solidarité, Santé, Cahiers statistiques 1984; 1:109Google Scholar In Canada, between 1966 and 1986, there was much less decline for women than men in the percentage who smoke regularly.13Manfreda J Mao Y Litven W Morbidity and mortality from COPD in Canada; Proceedings of the workshop on the rise in chronic obstructive pulmonary disease mortality. Bethesda, Maryland.December 3-4, 1987Google Scholar According to Durieux,14Durieux P Tabagisme: des pneumologues français trop discrets.Rev Mal Respir. 1986; 3: 121-122PubMed Google Scholar smoking is responsible for 300,000 deaths yearly in the US and for 70,000 in France. Petty15Petty TL Definitions. Clinical assessment and risk factors.in: Lenfant C Chronic obstructive pulmonary disease. Marcel Dekker, New-York1985: 1-30Google Scholar also noted that COPD is precipitated and aggravated by external factors such as smoking, but development of disease probably takes 30 years or more in the majority of cases. Consequently, the changes in smoking habits mentioned above are likely to involve modifications in mortality rates from COPD in the coming decades. Some differences between the three countries can be partly explained by coding practices and ICD revisions. These problems are known to complicate mortality analyses, especially between countries.16Bouvier MH Guidevaux M Mortality from disorders of the respiratory system throughout the world between 1950 and 1972.World Health Stat. 1979; 32: 174-197PubMed Google Scholar,17Thorn TJ Epstein FH Feldman JJ Leaverton PE Trends in total mortality from heart disease in 26 countries from 1950 to 1978.Int J Epidemiol. 1985; 14: 510-520Crossref PubMed Scopus (92) Google Scholar Results of international comparisons, therefore, have to be interpreted cautiously. Even though different diagnostic categories are used to define COPD and asthma, death rates were obviously higher in the US and Canada than in France from 1969 to 1983, and differences persisted when allowance was made for use of different categories. A more definitive analysis will be possible as additional mortality data under the 9th revision of the ICD become available.
Referência(s)