LIBRARY MIchigan State Unlversity PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MTE DUE DATE DUE DATE DUE L . 1/” COMM.“ AGE-PERIOD—COHORT-ANALYSIS OF CARDIOVASCULAR MORTALITY, MICHIGAN 1945-1995 By Fawzia K. Ahmed A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Epidemiology 1997 ABSTRACT AGE-PERIOD-COHORT-ANALYSIS OF CARDIOVASCULAR MORTALITY, MICHIGAN 1945-1995 By Fawzia K. Ahmed Period and cohort effects on cardiovascular disease mortality in Michigan between 1945 and 1995 were studied. Age, race, and gender-specific deaths in persons >19 years of age, and attributed to Major Cardiovascular Diseases (MCD), Diseases of the Heart (DOH), Ischemic Heart Disease (IHD), Cerebrovascular Disease (CD), and Hypertensive Heart Disease (HHD), were obtained. Population denominators for each year were estimated by linear interpolation of decennial census data. Age, race and sex specific mortality rates were calculated for successive 5 year periods. Overall, Michigan trends paralleled national data. MCD mortality was highest in white males, followed by black males, black females and white females respectively. It declined consistently in all age, race and gender groups over 50-year period. Deaths from CD and HHD were notably higher in blacks than whites. There was an overall decline in mortality due to DOH. In 1995, IHD was responsible for 70% of DOH deaths in whites and 46% in blacks. A cohort effect was not evident in any of the race/gender groups. These results emphasize the burden of premature MCD mortality, particularly among black males. To my family iii ACKNOWLEDGMENTS In researching and writing this thesis, I have been the beneficiary of three extraordinary groups of people: 1) my faculty at the Department of Epidemiology at Michigan State University; 2) the staffs of Michigan Public Health Institute(MPHI) and the Michigan Department of Community Health(MDCH); and 3) the members of my family. Dr. Aryeh Stein of MSU and Dr. Mathew Reeves of MDCH were invaluable as my principal advisors in this project. Together, they not only offered their gentle coaching but also their keen insights. Two other members of the MSU faculty, Drs. Nigel Paneth and Dr. Dorothy Pathak, remained steadfast in their interest and encouragement. Dr. Charlotte Pratt of MPH] sponsored this research, and offered useful feedback on my work. Much of the raw data analyzed in this thesis was drawn from the library of the Michigan Department of Community Health. and The Library of Michigan. I thank the staff of these institutions. Finally, I would like to thank my husband and my three children for their patience and good humor during these hours in the library or on the computer. My parents and brothers who were visiting during this writing were also full of encouragement. TABLE OF CONTENTS List of Tables-----.-- viii List of Data Sets- ix List of Figures xvi Chapter I Background I General trends of Cardiovascular Diseases ....................... 11. Trends of specific cardiovascular diseases. ...................... A) Ischemic heart disease ....................................................... B) Cerebrovascular diseases ................................................... C) Hypertensive heart disease ................................................ HI. Age-Period -Cohort Analysis ....................................................... Chapter 11 Issues A. Accuracy of mortality statistics ................................................... 1. Registration of deaths ...................................................... 2. Enumeration of population .............................................. B. Comparability of data .................................................................. l 4 8 14 16 23 3O 30 31 31 D. E. Chapter III Methods A. Chapter IV Results A. B. 1. Diagnostic accuracy ........................................................ 2. Change in coding ............................................................ Issues with comparability data ..................................................... Age-adjusted and age-specific rates ............................................ Race .......................................................................................... Data Sets .................................................................................. 1. Data sources ............................................................. 2. Count data ................................................................... 3. Population data ............................................................. Analysis ................................................................................... 1. Period analysis ............................................................. 2. Cohort analysis ............................................................ Major cardiovascular diseases ................................................. Diseases of heart .................................................................. vi 32 33 35 43 44 49 49 50 50 54 54 55 57 61 C. Ischemic heart disease ............................................................ 64 D. Cerebrovascular diseases ..................................................... 67 E. Hypertensive heart disease .................................................. 69 Chapter V Discussion .................................................................................................. 73 Conclusion .................................................................................................. 78 Appendix A Age-period-cohort graphs generated by W.H. Frost on Massachusetts 187 death rates from tuberculosis - for the time period, 1880,1910,1930 Appendix B Age-period-cohort graphs on peptic ulcer mortality in England 188 and Wales (1900 to 1977) generated by Mervyn Susser. List of References 189 vii LIST OF TABLES TABLE 1 Summary of studies on trends of cardiovascular diseases in United States TABLE 2 List year of adoption and use of Revisions of International Classification of Diseases (1CD) in United States. TABLE 3 Comparability ratios of International Classification of Diseases TABLE 4 Estimation of death rate in Michigan from count data and population data (cerebrovascular disease, 1960) TABLE 5 Intercensal estimate of white male population in Michigan from 1940 to 1970 TABLE 6 Estimation of Ischemic Heart Disease events in Michigan from US. data TABLE 7 Absolute and annual mortality rate change in cardiovascular diseases in Michigan viii LIST OF DATA SETS PERIOD DATA TABLES DATA 0001 Age specific mortality rates (by 5 yr intervals) for major cardiovascular diseases (ICD-9 codes 390-448) among white males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e. g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0002 Age specific mortality rates (by 5 yr intervals) for major cardiovascular diseases (ICD-9 codes 390-448) among white females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0003 Age specific mortality rates (by 5 yr intervals) for major cardiovascular diseases (ICD-9 codes 390—448) among black males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e. g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0004 Age specific mortality rates (by 5 yr intervals) for major cardiovascular diseases (ICD-9 codes 390-448) among black females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e. g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0005 Age specific mortality rates (by 5 yr intervals) for diseases of heart (ICD-9 codes 390- 429) among white males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0006 Age specific mortality rates (by 5 yr intervals) for diseases of heart (ICD-9 codes 390- 429) among white females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0007 Age specific mortality rates (by 5 yr intervals) for diseases of heart (ICD-9 codes 390- 429) among black males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0008 Age specific mortality rates (by 5 yr intervals) for diseases of heart (ICD-9 codes 390- 429) among black females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0009 Age specific mortality rates (by 5 yr intervals) for ischemic heart disease (ICD-9 codes 410-414) among white males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5- year period, the rate is an average of 3 years (e. g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0010 Age specific mortality rates (by 5 yr intervals) for ischemic heart disease (ICD-9 codes 410—414) among white females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0011 Age specific mortality rates (by 5 yr intervals) for ischemic heart disease (ICD-9 codes 410-414) among black males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5- year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0012 Age specific mortality rates (by 5 yr intervals) for ischemic heart disease (ICD-9 codes 410—414) among black females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5- year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0013 Age specific mortality rates (by 5 yr intervals) for cerebrovascular disease (ICD-9 codes 430—438) among white males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5- year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0014 Age specific mortality rates (by 5 yr intervals) for cerebrovascular disease (ICD-9 codes 430-438) among white females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0015 Age specific mortality rates (by 5 yr intervals) for cerebrovascular disease (ICD-9 codes 430-438) among black males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5- year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0016 Age specific mortality rates (by 5 yr intervals) for cerebrovascular disease (ICD-9 codes 430-438) among black females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5- year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0017 Age specific mortality rates (by 5 yr intervals) for hypertensive heart disease (ICD-9 code 402) among white males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. xi DATA 0018 Age specific mortality rates (by 5 yr intervals) for hypertensive heart disease (ICD-9 code 402) among white females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0019 Age specific mortality rates (by 5 yr intervals) for hypertensive heart disease (ICD-9 code 402) among black males, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (c. g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0020 Age specific mortality rates (by 5 yr intervals) for hypertensive heart disease (ICD-9 code 402) among black females, ages 20-84 years, in Michigan, 1945 to 1995. For each 5-year period, the rate is an average of 3 years (e.g. the rate for 1950 is an average of 1949, 1950, and 1951 rates). Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. COHORT DATA TABLES Data 0021 Age specific mortality rates by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among white males, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909 Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. Data 0022 Age specific mortality rates by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among white females, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909 Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. Data 0023 Age specific mortality rates by 5 year intervals for major cardiovascular diseases (ICD-9 ' codes 390-448) among black males, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909 Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. xii Data 0024 Age specific mortality rates by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among black females, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909 Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0025 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among white males, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909 Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0026 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among white females, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909 Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0027 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among black males, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909. Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0028 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among black females, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909. Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0029 Age specific mortality rates by 5 year intervals for ischemic heart disease (ICD-9 codes 410—414) among white males, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909. Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. xiii DATA 0030 Age specific mortality rates by 5 year intervals for ischemic heart disease (ICD-9 codes 410-414) among white females, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909. Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0031 Age specific mortality rates by 5 year intervals for ischemic heart disease (ICD-9 codes 410-414) among black males, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909. Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0032 Age specific mortality rates by 5 year intervals for ischemic heart disease (ICD-9 codes 410-414) among black females, ages 20-84 years, in Michigan, by year of birth. Birth years are shown for the central year of birth for a given 5 year period i.e. 1908 is for the birth year 1906-1909. Mortality data were obtained from vital records, population estimates were derived from decennial census and linear inter-censal interpolations. DATA 0033 Mortality rates from cerebrovascular disease (ICD-9 codes 430-438) among white males, ages 20-84 years, in Michigan. For each birth year (the central birth year), the rate is an average of 3 years, the rate is an average of 3 years(e.g. the rate is age specific rates for 1949, 1950, and 1951). Mortality data were obtained from vital records, population denominator derived from decennial census and linear inter-censal interpolation. DATA 0034 Mortality rates from cerebrovascular disease (ICD-9 codes 430-438) among white females, ages 20-84 years, in Michigan. For each birth year (the central birth year), the rate is an average of 3 years, the rate is an average of 3 years(e.g. the rate is age specific rates for 1949, 1950, and 1951). Mortality data were obtained from vital records, population denominator derived from decennial census and linear inter-censal interpolation. DATA 0035 Mortality rates from cerebrovascular disease (ICD-9 codes 430-438) among black males, ages 20-84 years, in Michigan. For each birth year (the central birth year), the rate is an average of 3 years, the rate is an average of 3 years(e.g. the rate is age specific rates for 1949, 1950, and 1951). Mortality data were obtained from vital records, population denominator derived from decennial census and linear inter-censal interpolation. DATA 0036 xiv Mortality rates from cerebrovascular disease (ICD-9 codes 430-438) among black females, ages 20-84 years, in Michigan. For each birth year (the central birth year), the rate is an average of 3 years, the rate is an average of 3 years(e.g. the rate is age specific rates for 1949, 1950, and 1951). Mortality data were obtained from vital records, population denominator derived from decennial census and linear inter-censal interpolation. DATA 0037 Mortality rates from hypertensive heart disease (ICD-9 code 402) among black males, ages 20-84 years, in Michigan. For each birth year (the central birth year) , the rate is an average of 3 years, the rate is an average of 3 years(e.g. the rate is age specific rates for 1949, 1950, and 1951). Mortality data were obtained from vital records, population denominator derived from decennial census and linear inter-censal interpolation. DATA 0038 Mortality rates from hypertensive heart disease (ICD-9 code 402) among black males, ages 20-84 years, in Michigan. For each birth year (the central birth year) , the rate is an average of 3 years, the rate is an average of 3 years(e.g. the rate is age specific rates for 1949, 1950, and 1951). Mortality data were obtained from vital records, population denominator derived from decennial census and linear inter-censal interpolation. DATA 0039 Mortality rates from hypertensive heart disease (ICD-9 code 402) among black males, ages 20-84 years, in Michigan. For each birth year (the central birth year) , the rate is an average of 3 years, the rate is an average of 3 years(e.g. the rate is age specific rates for 1949, 1950, and 1951). Mortality data were obtained from vital records, population denominator derived from decennial census and linear inter-censal interpolation. DATA 0040 Mortality rates from hypertensive heart disease (ICD-9 code 402) among black females, ages 20-84 years, in Michigan. For each birth year (the central birth year) , the rate is an average of 3 years, the rate is an average of 3 years(e.g. the rate is age specific rates for 1949, 1950, and 1951). Mortality data were obtained from vital records, population denominator derived from decennial census and linear inter-censal interpolation. XV LIST OF FIGURES PERIOD EFFECT GRAPHS FIGURE 0001 Age specific mortality by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among white males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0002 Age specific mortality by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-4-48) among white females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0003 Age specific mortality by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among black males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0004 Age specific mortality by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among black females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0005 Age specific mortality by 5 year intervals for diseases of heart (ICD-9 codes 390-429) among white males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0006 Age specific mortality by 5 year intervals for diseases of heart (ICD-9 codes 390-429) among white females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0007 Age specific mortality by 5 year intervals for diseases of heart (ICD-9 codes 390-429) among black males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. xvi FIGURE 0008 Age specific mortality by 5 year intervals for diseases of heart (ICD-9 codes 390-429) among black females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0009 Age specific mortality by 5 year intervals for ischemic heart disease (ICD-9 codes 410- 414) among white males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0010 Age specific mortality by 5 year intervals for ischemic heart disease (ICD-9 codes 410- 414) among white females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa FIGURE 0011 Age specific mortality by 5 year intervals for ischemic heart disease (ICD-9 codes 410- 414) among black males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0012 Age specific mortality by 5 year intervals for ischemic heart disease (ICD-9 codes 410- 414) among black females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0013 Age specific mortality by 5 year intervals for cerebrovascular disease (ICD-9 codes 430- 438) among white males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0014 Age specific mortality by 5 year intervals for cerebrovascular disease (ICD-9 codes 430- 438) among white females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0015 Age specific mortality by 5 year intervals for cerebrovascular disease (ICD-9 codes 430- 438) among black males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0016 Age specific mortality by 5 year intervals for cerebrovascular disease (ICD-9 codes 430- 438) among black females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. xvii FIGURE 0017 Age specific mortality by 5 year intervals for hypertensive heart disease (ICD-9 code 402) among white males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0018 Age specific mortality by 5 year intervals for hypertensive heart disease (ICD-9 code 402) among white females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0019 Age specific mortality by 5 year intervals for hypertensive heart disease (ICD-9 code 402) among black males, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. FIGURE 0020 Age specific mortality by 5 year intervals for hypertensive heart disease (ICD-9 codes 430-438) among black females, ages 20-84 years, in Michigan. Log of mortality rate per 100,000 is plotted as ordinate and year of death as abscissa. COHORT EFFECT GRAPHS FIGURE 0021 Age specific mortality rates by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among white males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0022 Age specific mortality rates by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among white females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0023 Age specific mortality rates by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among black males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa FIGURE 0024 Age specific mortality rates by 5 year intervals for major cardiovascular diseases (ICD-9 codes 390-448) among black males, ages 20-84 years, in Michigan, by year of birth. Log xviii of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa FIGURE 0025 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among white males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0026 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among white females, ages 20—84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0027 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among black males, ages 20—84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0028 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among black females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0025 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among white males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa FIGURE 0026 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among white females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa FIGURE 0027 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among black males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa xix FIGURE 0028 Age specific mortality rates by 5 year intervals for diseases of heart (ICD-9 codes 390- 429) among black females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa FIGURE 0029 Age specific mortality rates by 5 year intervals for ischemic heart disease (ICD-9 codes 410-414) among white males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa FIGURE 0030 Age specific mortality rates by 5 year intervals for ischemic heart disease (ICD-9 codes 410-414) among white females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa FIGURE 0031 Age specific mortality rates by 5 year intervals for ischemic heart disease (ICD-9 codes 410-414) among black males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa FIGURE 0032 Age specific mortality rates by 5 year intervals for ischemic heart disease (ICD-9 codes 410—414) among black females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0033 Age specific mortality rates by 5 year intervals for cerebrovascular disease (ICD-9 codes 430-438) among white males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0034 Age specific mortality rates by 5 year intervals for cerebrovascular disease (ICD-9 codes 430-438) among white females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0035 Age specific mortality rates by 5 year intervals for cerebrovascular disease (ICD-9 codes 430-438) among black males, ages 20-84 years, in Michigan, by year of birth. Log of XX mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0036 Age specific mortality rates by 5 year intervals for cerebrovascular disease (ICD-9 codes 430-438) among black females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0037 Age specific mortality rates by 5 year intervals for hypertensive heart disease (ICD-9 code 402) among white males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0038 Age specific mortality rates by 5 year intervals for hypertensive heart disease (ICD-9 code 402) among white females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0039 Age specific mortality rates by 5 year intervals for hypertensive heart disease (ICD-9 code 402) among black males, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. FIGURE 0040 Age specific mortality rates by 5 year intervals for hypertensive heart disease (ICD-9 code 402) among black females, ages 20-84 years, in Michigan, by year of birth. Log of mortality rate per 100,000 is plotted as ordinate and central year of birth is shown for 5 year periods on the abscissa. xxi Age-period-cohort analysis of cardiovascular mortality, Michigan 1945-1995 Chapter 1 - Background A. General trends of Cardiovascular Diseases Over the past three decades, there has been a marked decline in death rates from cardiovascular diseases (MCD) in the United States [4,7,43]. In 1960s, among all industrialized countries the United States had the second highest death rate from ischemic heart disease [IHD], a major component of cardiovascular diseases. However, by 1990 the United States ranked 18th for men and 14th for women in ischemic heart disease deaths and was among the lowest-ranked countries in stroke deaths [43]. If the rise in mortality rates of the 19505 had continued until the present, annual deaths from ischemic heart disease would be nearly 1.5 million; instead they are now half that number [43]. The decline in MCD death rates are related both to improvements in medical care and to decreases in cardiovascular diseases risk factors [43, 16]. Despite these remarkable achievements, cardiovascular diseases remain the leading cause of death in the US (and most other western countries). According tothe American Heart Association in 1996 the yearly cost of cardiovascular diseases in the United States was estimated to ber $259 billion. In 1996 according to statistics provided by American Heart Association [1], the number of deaths from 5 groups of cardiovascular diseases are listed below: Major Cardiovascular Diseases 960,592 Cerebrovascular Disease 157,991 Diseases of Heart 737,563 Hypertensive Heart Disease 39,981 Ischemic Heart Disease 481,287 About 16 percent of MCD deaths occurred among those under 65 years of age [43]. The epidemiological pattern of mortality and morbidity from cardiovascular disease is generally characterized by differences due to age, race, and gender. Before age 60, heart disease rates are substantially higher in men than in women. However, heart disease increases markedly in women after menopause, with women accounting for nearly half of the heart disease deaths annually [31]. Mortality rates from some of the cardiovascular diseases tend to be higher in the black population than in other ethnic groups in the United States [20]. Furthermore, there was great disparity in the rates of decline between blacks and whites for most categories of cardiovascular diseases. Among white males, death rates from cardiovascular diseases declined 42 percent between 1950 and 1988. In comparison, the decline among black males for the same time period was only 31 percent. On a national level, it has been observed that more than 7,000 fewer blacks would have died of cardiovascular diseases in 1988 if their death rates had equaled those of whites [43]. The American Heart Association reported that compared to whites, blacks have more hypertension related morbidity, a higher mortality rate from stroke, more frequent left ventricular hypertrophy by electrocardiographic criteria, and strikingly more end-stage renal disease [3]. In Michigan, Cardiovascular Disease is the largest cause of adult deaths. According to American Heart Association, 36,070 Michigan residents died in 1995 from cardiovascular diseases. That figure represents 43 percent of all deaths, 31 percent more deaths than cancer, and forty five times as many deaths as were due to AIDS. The objectives of this study were to a) determine the overall trends in mortality rates in cardiovascular diseases among the race/gender groups in Michigan during the period 1945 to 1995 and b) determine the effect of three interrelated factors - age, calendar period of death, and birth cohort - on mortality from cardiovascular diseases. Knowledge of cardiovascular disease mortality rates by race, gender and age as well as temporal changes in mortality rates within Michigan are important for health planners in continuing and implementing programs aimed at awareness, prevention, and treatment of cardiovascular diseases. Age-period-cohort analysis can be used to make future predictions of the impact of disease in various race and gender groups. Cohort analysis has not been widely applied previously to cardiovascular disease mortality with the exception of stroke. This analytical approach could, however, provide valuable new insights into the epidemiology of cardiovascular diseases. 11. Trends of specific cardiovascular diseases An overview of ICD-9 codings for Major cardiovascular diseases and Diseases of heart is provided with an in-depth description for 3 specific subgroups (Ischemic heart diseases, Cerebrovascular disease, Hypertensive heart disease). These specific subgroups will be presented in order of the importance of burden of their mortality in the population. Major cardiovascular diseases (MCD): Major cardiovascular diseases (MCD) refer to the disorders of the heart and blood vessels that are included in the rubric of ICD-9 Nos. 390-448. Within this category are several major clinical groupings. The first two, Diseases of the Heart and Cerebrovascular Disease are examined separately in this study: Major cardiovascular diseases (ICD-9 Code No. 390-448) consists of: [CD-9 Nos. 390-398, 402, Diseases of Heart 404, 410.429: ICD-9 Nos. 430-438: Cerebrovascular Disease ICD-9 No. 401: Essential Hypertension ICD-9 No. 403: Hypertensive Renal Disease ICD-9 No. 405: Secondary Hypertension ICD-9 Nos. 440-448: Diseases of Arteries, Arterioles, and Capillaries Diseases of heart: Diseases of Heart is the largest component of Major cardiovascular diseases. Nationally, diseases of heart started to decline after 1950 and the rate declined at a more rapid rate during each successive decade. From 1950 to 1960, the age adjusted death rate for heart disease fell by an annual average of 0.88 percent. During 1960-1970, it dropped an average of 1.14 percent annually; from 1970-1980 it dropped 2.03 percent; and from 1980-1990, 2.21 percent annually [59]. In 1950 heart disease caused 37 percent of all deaths compared with 32 percent in 1988. The pace of decline accelerated steadily for white and black males, while among females of both races, especially blacks, it accelerated until the early 19805, and then slowed. In 1986, age adjusted rates were 91% higher in men than in women and 37% higher in blacks than in whites. Differences in mortality rates in blacks and whites were higher in younger than in with older persons [12]. Diseases of heart (ICD-9 Code No. 390-429) consists of: ICD Code No. 390-398: Acute rheumatic fever and chronic rheumatic heart disease ICD Code No. 401 Essential hypertension ICD Code No. 402: Hypertensive heart disease (Section C) ICD Code No. 403: Hypertensive renal disease ICD Code No. 404: Hypertensive heart and renal disease ICD Code No. 405: Secondary hypertension ICD Code No. 410-414 Ischemic heart disease (Section A) ICD Code No. 415 Acute pulmonary heart disease ICD Code No. 416 Chronic pulmonary heart disease ICD Code No. 417 Other diseases of pulmonary circulation ICD Code No. 420: Acute pericarditis ICD Code No. 421: Acute and subacute endocarditis ICD Code No. 422: Acute myocarditis ICD Code No. 423: Other diseases of pericardium ICD Code No. 424: Other diseases of endocardium ICD Code No. 425: Cardiomyopathy ICD Code No. 426: Conduction disorders ICD Code No. 427: Cardiac dysrhythmias ICD Code No. 428 Heart failure ICD Code No. 429 Ill-defined descriptions and complications of heart The category of cerebrovascular disease is encompassed in International Classification of Diseases, 9th Revision (ICD-9) codes 430-438. Cerebrovascular disease ICD -9 Nos. 430-438 (Section B)] ICD Code N o. ICD Code No. ICD Code No. ICD Code No. ICD Code No. ICD Code No. ICD Code No. ICD Code No. ICD Code No. 430 431 432 433 434 435 436 437 438 Subarachnoid hemorrhage Intracerebral hemorrhage Other unspecified intracranial hemorrhage Occlusion and stenosis of precerebral arteries Occlusion of cerebral arteries Transient cerebral ischemia Acute, but ill-defined, cerebrovascular disease Other and ill-defined cerebrovascular disease Late effects of cerebrovascular disease However, due to the difficulties (and incompleteness) of exact diagnosis, from a public health perspective, these codes are often combined into one group (430-438). A. Ischemic heart disease (IHD): A.1 Overview Ischemic heart disease (ICD-9 Nos. 410-414) accounts for more than 50 percent of all cardiovascular disease deaths in the US [43]. In 1995 a total of 481,287 persons died as a result of IHD. Ischemic heart disease (IHD) is a generic term for any type of cardiac diseases that produce inadequate myocardial oxygenation. This may result from several different types of diseases of the coronary arteries, including coronary atherosclerosis leading to thrombosis, arterial spasm and embolism. It is commonly manifested as a heart attack which results from ischemia to the myocardial muscle. Angina pectoris (chest pain) is a symptom of inadequate oxygenation which can result from any of the pathologies listed. The rubric for Ischemic Heart Disease included 5 codes :- ICD Code No. . 410 Acute myocardial infarction ICD Code No 411 Other acute and subacute form of ischemic heart disease ICD Code No 412 Old myocardial infarction ICD Code No 413 Angina pectoris ICD Code No 414 Other forms of chronic ischemic heart disease Mortality rates for Ischemic Heart Disease were at their peak in 1965, when age adjusted IHD mortality for the United States was approximately 240 per 100,000. After 1965, however, overall rates declined approximately 2% each year. Initially, this decline occurred approximately equally in men and women, blacks and whites, and across all age groups. However, after 1976, the secular trends diverged considerably. Rates of decline were steeper for whites than for blacks, and steeper for men than for women. From 1990 through 1994, age-adjusted Ischemic Heart Disease death rates for the United States population 35 years and older declined 10.3% i.e. from 416.3 deaths per 100,000 to 373.6 deaths per 100,000 [32]. However, the rate of decline again diverged by race and gender. The largest annual percentage decline occurred among white men (2.9% per year), followed by white women (2.5%), black men (2.3 %), and black women (1.6%) [32]. A.2 Etiology and Pathophysiology of IHD: The basic pathophysiology of ischemic heart disease is an imbalance between myocardial oxygen supply and demand [6]. Oxygen supply may decrease or demand may increase beyond the limits of coronary perfusion reserve resulting in ischemia. Myocardial oxygen supply can decrease due to decreased cardiac output, decreased diastolic pressure or by increased coronary arteriolar resistance. Myocardial oxygen demand can increase with increased heart rate, increased heart size and increased systolic pressure. Atherosclerosis is the first sign of IHD. Atherosclerosis is a pathologic condition of the coronary arteries characterized by a series of abnormal changes of the intima of the arteries [56], the earliest lesions of which are the development of fatty streaks. Fatty streaks may progress to fibrous plaques and eventually to lesions complicated by ulceration, hemorrhage, calcification and thrombosis. Fatty streaks are characterized by an accumulation of lipid filled smooth muscle cells and macrophages and fibrous tissue in focal areas of the intima [56]. The fatty streak is usually sessile and causes little obstruction and no symptoms. The lesion appears in various segments of the arterial tree at different ages. It is generally believed that the fatty streaks are reversible, but the evidence is inconclusive. Fibrous plaques, also called raised lesions, are palpably elevated areas of intimal thickening and represent the most characteristic lesion of advancing atherosclerosis. These plaques first appear in the abdominal aorta, coronary arteries, and carotid arteries in the third decade and increase progressively with age [56]. They appear in men before women. Typically the fibrous plaque is firm, elevated and dome shaped that bulges into the lumen. It consists of a central core of extracellular lipid and necrotic cell debris covered by a fibrovascular layer containing large numbers of smooth muscle cells, macrophages, and collagen. The plaque is much thicker than normal intima. The complicated lesion is calcified fibrous plaque containing various degrees of necrosis, thrombosis, and ulceration [56]. These are the lesions frequently associated with symptoms. With increasing necrosis and accumulation of gruel, the arterial wall progressively weakens, and rupture of the intima can occur, causing aneurysm and hemorrhage. Arterial emboli can form when fragments of plaque dislodge into the lumen or thrombi and can result from gradual occlusion as plaques thickens. Although knowledge of the etiologic events is incomplete, it is clear that no single factor is responsible for the development of atherosclerosis. Epidemiologic studies have established the association between certain factors and coronary atherosclerosis. Risk factors for the development of coronary atherosclerosis include a family history of premature coronary artery disease, cigarette smoking, hypercholesterolemia, hypertension, and diabetes mellitus [6]. Obesity, physical inactivity, and stress may also play a role in the development of atherosclerosis. A.3 Symptoms and Signs of IHD: There are three main types of symptoms of ischemic heart disease [6]. However, almost 30% of all myocardial infarctions are silent (without symptoms). a) Angina Pectoris Brief pain attributed to myocardial ischemia (angina pectoris) is typically a retrosternal discomfort brought on by exercise or emotion and relieved by rest. It usually lasts at least 3 to 5 minutes. The discomfort is not necessarily described as pain, but rather as a fullness, ache, pressure, tightness or burning. The discomfort may occur in the jaw, neck, throat, interscapular area, or arm; it may be on the left more often than on the right, and may occur with or without simultaneous chest discomfort b) Severe Angina Prolonged pain of myocardial ischemia has the same characteristics as brief pain, but the duration is longer (15 minutes to several hours). The pain may be more severe and may be associated with nausea and perspiration. It may occur at rest or during exertion. c) Other cardiovascular symptoms Other cardiovascular symptoms may be associated with brief pain or prolonged pain. Patients may have palpitation (rapid heart beat), dizziness or weakness. A.4 Diagnostic Tests of IHD: 13 a) Electrocardiogram of IHD : A normal electrocardiogram (EKG) does not exclude the diagnosis of ischemic heart disease, however, certain characteristic abnormalities tracings obtained at rest can confirm it. ST segment change and T-wave changes accompany episodes of angina pectoris and disappear thereafter [56]. b) Stress Test: The most widely used test in the diagnosis for ischemic heart disease involves recording the 12-lead EKG before, during and after exercise on a treadmill or using a bicycle ergometer [56]. The test consists of a standardized incremental increase in external workload while the patient’s EKG, symptoms, and arm blood pressure are continuously monitored. c) Coronary arteriography: This invasive diagnostic method is used to detect evidence of coronary atherosclerosis or to exclude the condition [56]. Coronary arteriography is indicated in 1) patients with chronic stable or unstable angina pectoris who are refractory to medical therapy and who are being considered for revascularization, and 2) patients with troublesome symptoms 14 that present diagnostic difficulties in whom there is need to confirm or rule out the diagnosis of coronary artery disease. d) Serum enzyme studies: Enzymes are released in large quantities into blood from necrotic heart muscle following myocardial infarction. The rate of liberation of specific enzymes differs following infarction, and temporal pattern of enzyme release is of diagnostic importance. SGOT and creatine phosphokinase (CK) rise and fall rapidly, while that of lactic dehydrogenase (LDH) rises later and remains elevated longer [56]. CK is more specific than SGOT since it is not present in significant concentrations in extra-cardiac stores. In myocardial infarction, the level of LDH rises during the first day, peaks at 3 to 4 days, and returns to normal in 14 days. Characteristic rises occur in serum enzyme concentration in more than 95% of patients with clinically proven myocardial infarction. The amount of enzyme releases also correlates with the size of infarct. B. Cerebrovascular Diseases (CD) B.1 Overview Cerebrovascular disease or stroke is an important form of cardiovascular disease both in terms of mortality and morbidity. Of all cardiovascular disease deaths, nearly 16% are due to cerebrovascular disease or [43]. CD is the third leading cause of death in United States. The death rates for CD are higher for non-whites than for whites up to age 75, reflecting mainly higher incidence and severity of hypertensive disease among blacks. At older ages Cerebrovascular Disease rates are higher in whites because of the increasing prominence of atherosclerotic thrombo-occlusive cerebrovascular disease [35]. It has an overall prevalence of 794 per 100,000 [56]. Five percent of the population over 65 are affected (mortality and morbidity) by it and more than 400,000 patients are discharged each year after an episode of cerebrovascular disease [56]. Blacks are twice as likely as whites to succumb to strokes. The extended hospitalization and post treatment rehabilitation that the affected persons require during recovery make the economic impact of this disease very devastating. In US, the age-adjusted mortality rate for cerebrovascular disease declined even more sharply between 1950 and 1988 than did Diseases of the Heart. CD declined at similar rates in all age groups [43]. Though in all race and sex groups the death rate for stroke declined at least 60 percent, the greatest decline was noticed among black women (68%). In the 19505, 19605, and 19705 these reductions accelerated for whites and blacks of both sexes. In the 19805 the rate of decline still continued, though slowed for all except white males [43]. The black/white variation, however, changed differently for males and females. In 1950, for example, the death rate for stroke among black females was 95 percent greater than among white females, but by 1988 the difference was 83 percent [59]. The difference of cerebrovascular disease mortality rate between black and white 16 males, however, increased from 68 percent to 93 percent [43]. Black males have a far higher CD death rate than any other group. B.2 Symptoms and Signs: The exact signs of cerebrovascular diseases depend on the type of cerebrovascular disease pathology and the area of brain affected. Brain damage from cerebrovascular diseases can cause: a. Paralysis - Paralysis of one side of the body (hemiplegia) is very common. However, sometimes the paralysis is more localized such as only a leg or an arm. b. Loss of sight - May affect reading, objects may appear closer or farther away than they really are. c. Aphasia and dysarthria - Victims may have trouble verbalizing. Aphasia affects the ability to talk, listen, read and write. Aphasia usually occurs when cerebrovascular disease affects the right side of the body. Dysarthria affects speech which can be slowed, slurred or distorted. C. Hypertensive heart disease: (ICD-9 code no. 402) C. 1 Overview When cardiac abnormality (for example, on the X-ray or electrocardiogram) is demonstrable in association with elevated blood pressure, the term Hypertensive heart disease [ HPD] is employed [32]. Hypertensive heart disease produces the third largest number of deaths from Disease of Heart, but only 4 percent as many as the ischemic form [59]. Moreover, death rates have declined more slowly for hypertensive heart disease than for ischemic heart disease. Hypertensive heart disease death rates are 1.6 to 13.5 times greater in blacks than in whites depending age and sex. The discrepancy is particularly greater in the ages 25-64, though it does diminish in old ages [59]. According to ICD-9 [21] Hypertensive heart disease (ICD-9 Code No. 402 ): It consists of several sub-categories. ICD Code No. 402.0 Malignant ICD Code No 402.00 Without congestive heart failure ICD Code No 402.01 With congestive heart failure ICD Code No 402.1 Benign ICD Code No 402.10 Without congestive heart failure ICD Code No 402.1 1 With congestive heart failure ICD Code No 402.9 Unspecified ICD Code No 402.90 Without congestive heart failure ICD Code No 402.91 With congestive heart failure The trend of mortality rates for hypertensive disease was steadily downward from 1940- 1960 for white males and females in all age groups [35]. The rate of decline accelerated in 19505 and the age adjusted rate of decline was 60 percent. 35 to 44 year old men and women were benefited the most. The declines were considerably greater for women than for men. In 1940, in 25 to 34 age group, the mortality rate for hypertensive heart disease was seven times higher in blacks than in whites. Mortality rates for hypertensive heart disease in blacks had not declined until 1950. For the decade 1950 to 1960, rates for nonwhites fell substantially; but at all ages up to 85, the rates of decline were not as great as those for whites. As a result of these trends, the relative position of non whites in comparison with whites worsened from 1940 to 1960. C.2 Etiology and pathophysiology of HPD The relationship between hypertension and heart disease is not clear, and the delineation of hypertensive heart disease as a pure etiologic entity, independent of other forms of heart disease, is ill defined. The causal relationship of hypertension to heart disease is 19 based on the frequency with which the two conditions are associated i.e. the high incidence of hypertension in cases of heart disease and predominant frequency of ultimate cardiac disease in hypertensive patients. Furthermore, physiologic evidence that hypertension increases the work and impairs cardiac function strengthens the concept of a form of heart disease due to hypertension [35]. However, the exact causal relationship between hypertension and heart disease is obscured by two types of data [35]: 1. There is a lack of correlation in most cases between the severity and duration of hypertension and the development of cardiac complications. 2. Hypertension is associated with a high incidence of coronary atherosclerosis. In 70 percent of coronary atherosclerotic heart disease hypertension is present. Conversely coronary sclerosis has been found in 90 percent of hearts of persons dying with hypertension [35]. Hypertensive heart disease is due to the excessive work load imposed by increased systemic pressure. It is first evident as ventricular hypertrophy. Ventricular hypertrophy is characterized by an increase in thickness of ventricular walls. Ultimately the functions of this chamber deteriorates, the cavity dilates, and the symptoms and signs of heart failure appear[56]. 20 Q3 Symptoms and signs: Palpitation and rapid heart action are the earliest symptoms of Hypertensive heart disease. Headache, fatigue, diminished exercise tolerance and shortness of breath are however, some of the other common symptoms [6]. On physical examination, a fourth heart sound is sometimes evident. C.4 Diagnostic tests of HPD The EKG and chest roentgenogram are exceedingly important diagnostic tools for determining Hypertensive heart disease [6]. a) Left atrial abnormality: The earliest sign of cardiac involvement following hypertension is the electrocardiographic finding of left atrial abnormality. The left atrial changes may be present without EKG based evidence of left ventricular enlargement. They do not reflect atrial disease, but rather the response of a dynamic atrium to the less compliant ventricles as hypertrophy is developing. Nevertheless, these patients demonstrate evidence of left ventricular hypertrophy by echocardiography. The ventricular septum thickens and ventricular mass is increased. The fiber shortening rate of the left ventricle is also impaired [6]. 21 b) Left ventricular hypertrophy: As clinical ventricular hypertrophy becomes obvious, it can be identified by EKG or roentgenographic criteria. At this stage of Hypertensive heart disease resting cardiac output is reduced. c) Left ventricular failure: Eventually if hypertension remains untreated, left ventricular failure takes place. D. Summary of trend for the three Cardiovascular Diseases The studies on trends of cardiovascular disease in US as described in literature is summarized in TABLE 1. Cooper et a1 analyzed mortality rates in the four major race-gender group from 1940- 1975. He showed that the age adjusted mortality rate rose in all race and sex group from 1948-1967. The rate of increase was greater in blacks than in whites. However, during the period 1968-1975 age adjusted mortality rates from IHD in the US declined significantly among all four race-gender groups (blacks and whites, males and females). Overall, the trend was more favorable for blacks than for whites. The data for this analysis was drawn from two sources. Moriyama et a1 provided the data for 1940-1966. The National Center for Health Statistics was the source of data for the 22 years 1967-1975. To minimize the effect of the changes in ICD revisions, the data were analyzed in four time sets, corresponding to the periods of the four ICD revisions. Sempos et al demonstrated the divergence in IHD mortality trends among the four major race/sex groups. He computed age adjusted mortality rates of US population from 1968- 1985 [49]. Age adjusted mortality rate declined almost equally in all four groups from 1968-1975. However, during 1976-1985, secular trends diverged considerably. The rates of decline for black males, black females and white females slowed down, but white males actually experienced an increase in their rate of decline. Cardiovascular mortality trends from 1980-1986 was studied in Harris County, Texas by Yoon et al. Mortality trends in four ethnic groups (blacks, whites, Hispanics and others) were examined [58] The black population, in general, showed significantly higher cardiovascular mortality rates than the other ethnic groups. Higher cardiovascular disease mortality rates were also found in the white and black populations with respect to the Hispanic and “other”. Furthermore males had higher cardiovascular mortality rates than females. 23 Age-Period-Cohort Analysis: Time trends of incidence and mortality rates for a particular disease often provide an epidemiologist with important clues for disease etiology. Three time factors which are often considered in such an investigation are i) age at death ii) date of death (period effect) iii) date of birth (cohort effect) [24]. An age effect is an intrinsic effect independent of period and experience, and dependent on differentiation and maturation of the organism [53]. It is present when the disease rate varies by age regardless of birth cohort and independent of the period (calendar time). Cohort effects are environmental effects attributed to the singular experience of each cohort antecedent to the time when the outcome of interest is observed. A cohort effect is present when the disease rate varies by year of birth, regardless of age. That is persons born in certain years carry with them throughout their lives a relatively higher (or lower) rate of disease. For example, children born during the years when diethylstilbestrol was prescribed to pregnant women in US might face a lifetime risk for certain type of cancer that differs from that faced by children born at another time [24]. Not only are the factors at the year of birth considered as cohort effect but any factor that affects disease incidence that is related to year of birth. For example, cigarette smoking is usually started in late teens, the effect of any large change in smoking habits depend on a person being of a 24 particular age when the change the in smoking habit occurs. Diseases like cardiovascular diseases that are strongly related to cigarette smoking might be expected to produce a birth cohort pattern even though there was nothing associated with birth itself that influenced disease risk [24]. Period effects are environmental effects attributed to and defined by a given time period [53]. The population at risk is thus all those living at the point in time when the observations are made. For example changes in levels of a pollutant in the air or water might be expected to produce the same change in disease risk for everyone in the population. Changes in medical technology also might produce such a period effect by ascertaining more cases due to improved diagnostic ability, thereby producing an artifact that may not represent a true important change in public health [53]. A secular trend is characterized by systemic change in age specific rates over calendar time. For any single age group, a secular trend may be due to period effect, a cohort effect, or some combination of the two [24]. The procedures used for analyzing longitudinal data is called cohort analysis [24]. Cohort analysis involves the retrospective collection of data from three or more observation periods for a single dynamic population. The approach may involve incidence, prevalence, or mortality data and the entire study period is usually spaced over a span of 20 or more years. For example, using vital statistics and census information between 25 1945-1995, we could estimate the age specific mortality rates for cardiovascular disease among white males, white females, black males and black females of Michigan. In cohort analysis, the rate of disease is displayed graphically as a function of age, by birth cohort (i.e. year of birth) or by period (i.e. calendar time). Frost in 1939 considered the implications of these three factors on mortality rates from tuberculosis in Massachusetts [9]. The procedure he developed was primarily descriptive, and graphs were used to examine patterns in disease rates over time. Using a graphical approach, he found that the age and cohort analysis provided a consistent pattern in the trends, which was not apparent for the age and period analysis. He demonstrated that tuberculosis mortality rate peaked at different age groups over time. He showed that in 1910, the highest mortality from tuberculosis was evident at the age of 30-40 . However, in the later period (1930), the highest rate of mortality came at the age of 50-60. The same group of people who were at higher risk in 1910 when they were 30- 40 years old carried with them this higher risk for tuberculosis in 1930 when they were 50-60 years old [Appendix A]. For past few years, age period cohort analysis (APC) has become a popular epidemiological tool. Regression models were designed by different investigators to quantify the separate effects of these three factors: age, period, and cohort [26]. These investigators studied several different diseases, including as breast cancer [33], cancer of 26 the cervix [2], prostrate cancer [22], bladder cancer [34,51] and lung cancer [51,52]. Each study has adopted a regression analysis approach in the treatment of incidence or mortality data; typically a three factor model (age at occurrence of disease or death, time of occurrence of disease or death (i.e. period and birth cohort) or a two factor model (usually age at occurrence and birth cohort), or some modification of these two models has been employed. Methodically the major objective of cohort analysis is the empirical separation of three time related effects that could provide alternative explanation for the observations; these are age, period and birth cohort effects. There is an inherent limitation of cohort analysis since each set of data has at least two etiologic explanations [24]. For example, if we know that a person is 50 years old in 1980, then we also know that the person was also born in 1930. This limitation is often called an identification problem in APC analysis [24] and is due to the fact that the factors age, period and cohort are mathematically related. When these factors are treated as continuous variables, this mathematical relationship (when mortality is the end point) is simply (year of birth) + (age of death) = (year of death) [24]. The correspondence among these factors is called a linear dependence. In regression analysis, when regressor variables are linearly dependent, it is not possible to attribute separate effects to each of these factors. Again, the net result of this identification problem is that any attempt to separate the three effects empirically depends on a priori knowledge [24]. For example, if cardiovascular disease mortality in a population is due to period effect, the variation in mortality rate may have been due to a 27 gradual increase in the prevalence of certain risk factors - affecting all age groups and birth cohorts equally. However, this assumption is not valid, specially since we know that there is an age effect. Thus, in general, when observing one age group, one period, or one birth cohort, there is no way to separate out the effects of the other two factors. Mortality data are usually organized as a set of age-race-sex specific rates for several periods of time. The interval widths for age and period are usually equal, typically i.e. both age and period are divided into five-year intervals. The birth cohort is defined by the age of a subject and the date of occurrence of the event of interest. Because age and period are expressed as intervals, birth cohorts are intervals also. Indeed birth cohort intervals are longer and may overlap to some extent. For instance, if we have five-year age and period intervals, then individuals aged 50-54 who died during the period 1960-64 were born sometime during the years 1905-1914. Similarly, individuals aged 55-59, who died during the period 1960-1964 belong to the 1910-1919 cohort, which overlaps the previous cohort. Since population based data are often tabulated in 5 or 10 year age group and it is not possible to obtain individual records, including date of birth. The interpretation of secular trends in terms of period, age and cohorts by Susser is illustrated by data on peptic ulcer mortality for England and Wales from 1900 to 1977 [53]. He examined the relationships between age effects, period effects and cohort effects. Susser adopted the cohort method that relies upon the recognition of patterns in sets of age specific rate curves. The interpretation may depend upon recognition of 28 patterns in sets of curves or on their quantified relation. Prior to Susser’s analysis, the prevailing view was that incidence of peptic ulcer was rising. Cohort analysis by Susser changed the interpretation of the overall pattern, and showed that the disease was probably in decline. Susser presented the data in semi-log graphs for duodenal ulcer in males. Period date contours for peptic ulcer (Appendix B. Fig. a) The sharply divergent age distribution of mortality for the different time period makes mortality pattern confusing and irregular. Cohort date-of- birth contours for peptic ulcer(Fig. b) The pattern in the ‘date of birth’ curves shows a steady rise for each successive birth year or generation between 1835 to 1885. For each successive birth year or generation after 1905, by contrast, a steady decline can be observed. For the birth years 1885-1905, the pattern is muddled by the overlap of curves. To avoid the problem of waxing and waning of mortality, the data was recast in the form of age contours. Period age contours for peptic ulcer (Fig. c) These age contours show a fan shaped divergence, increasing over time between the 29 mortality curves of young and old age-groups. The pattern, with relative small age differences in early time period and larger age differences at later times, is typical of waxing and waning rates among successive generations. Cohort age contours for peptic ulcer (Fig. d) In figure (1 the period age curves of figure c are reassembled with the abscissa indicating date of birth, instead of date of death. In this arrangement, beginning with the birth year on the abscissa, each generation can be followed vertically upwards across the curves for successive age groups. It becomes evident that for successive birth year cohorts, up to 1885, mortality rises steadily, and after 1895 declines steadily. Also an age effect is present for all cohorts throughout, excepting the beginning of the curve for the 75 years and over age group. Chapter II - Issues A. Accuracy of mortality statistics: The accuracy of mortality statistics depends on the method of data collection and reporting process. The factors affecting mortality rates are: 1) registration of deaths and 2) enumeration of population [35]. 1. Registration of Deaths: The annual collection of mortality statistics for the United States Death Registration States began with the calendar year 1900. However, complete nationwide coverage was not obtained by the death registration system until 1933 [35]. It has been generally assumed that registration of deaths is relatively complete, with possible underregistration of certain of segments of the population and the less accessible areas of the country. Underregistration of deaths is probably higher in the black than in the white population. It is also likely to be a problem in certain areas of the country where registration offices are not readily accessible for geographic reasons. Overall, it is unlikely that cardiovascular disease mortality data are affected to any great extent by incomplete death registrations. 30 31 2. Enumeration of P0pulation: It is likely that underenumeration of the population is a more important source of inaccuracy in death rates than underregistration of deaths. The effect of underenumeration would be to over inflate death rates - opposite to that of underregistration of deaths. The undercount of the population in the censuses has been shown to be unequal among different sex and race groups. The undercount rate has been higher for men than for women and higher for blacks than for whites. It was demonstrated in 1955 that almost 10 percent of the blacks were not counted in the 1950 census [35]. The percentage of understatement was smaller in the succeeding decades, and in 1990, 7.5 percent of all black males and 2 percent of all white males were estimated to not have been counted [25]. These levels of understatement of population and resulting overstatement of death rates should be taken into account in interpreting mortality rate. B. Comparability of Data: Comparability of data is affected by 1) diagnostic accuracy and 2) change in coding. 32 B.1 Diagnostic accuracy The chief problem in interpreting mortality statistics on cardiovascular disease deaths are uncertainty and inaccuracy of diagnosis, lack of uniformity in diagnostic terminology used, and selection and classification of the diagnostic term to be tabulated when multiple contributory causes of death are reported. The accuracy of mortality statistics depends on the method of data collection and reporting process. At one extreme, the disease is asymptomatic and undetectable for many years; at the other, it is just one aspect of multiple other problems that are present at the time of death. Thus, change in coding and change in diagnostic practice makes comparability of data difficult. Cause of death statistics are essential in evaluating public health problems and in measuring the progress of medical science in resolving these problems. Statistics on causes of death are derived from information entered on the death certificates. This portion of the death certificate is completed by the physician who attended the descedent or by the medical examiner in cases of death by violence and of death without medical attendance [38]. The cause of death information so reported is classified according to the International Lists of Causes of Death. This classification scheme provides a uniform basis for the assignment of the causes of death. When one cause of death is reported on the death certificate, the assignment is relatively simple. However, more than half of the death certificates in US report two or more 33 diseases as causes of death [38]. It is required to select only one underlying cause of death for statistical analysis. Hence the method of selection used has an important effect upon the resulting statistics. Before 1949, selection of a cause of death for tabulation was made by reference to a set of priority tables, published in the Manual of Joint Causes of Death [38]. By using these priorin tables when more than one cause was jointly reported, the primary cause for tabulation was selected. The new international rules adopted for use in 1949 made the medical practitioner responsible for indicating the underlying cause of death for tabulation. Underlying cause of death, is defined as the disease or injury which initiated the train of events leading directly to death. For example, Right lower limb of a man was paralyzed after stroke and he could not walk without help. While trying to walk he fell down from stairs and died of a head injury. He would be categorized as stroke death (underlying cause) after 1949 but as head injury death (primary cause ) before I 949. B.2. Change in coding: The causes of death are standardized by the World Health Organization (WHO) in its Manual of the International Statistical Classification of Diseases, Injuries, and Causes of 34 Death (ICD). ICD has been in use since 1900. The ICD describes the specific causes and groups them into major categories and specific subcategories at several levels. The ICD also includes rules for systematically identifying the underlying cause of death, which is relatively easy to report and understand. ICD has been revised approximately every ten years to reflect progress in medical knowledge [Table 2]. ICD 9, however, has been in place since 1976. Though the revisions introduce refinements and categories that correspond better to changing realities they create certain comparability problems over time. From 1945 to 1995 five different versions of the International Classification of Diseases (ICD) were used namely the Fifth, Sixth, Seventh, Eighth and Ninth. A systematic method for evaluation of classification changes in transition from one ICD revision to the next is essential in order to determine the net effect of any change in ICD version. The International Conference for the Decennial Revision of the International Lists of Diseases and Causes of Death recommended that a sample of deaths for a country as a whole should be coded according to the two consecutive Revisions [38]. This dual coding of causes of death (first introduced after 6th Revision in 1950) shows the net changes resulting from the two revisions. The comparability ratio has been in use after 1950 for all ICD revisions. The number of deaths assigned to a particular cause of death in accordance to the most recent revision divided by the number of deaths assigned to a comparable cause in accordance with the previous revision gives the comparability ratio (C.R.) [ Table 3]. The comparability ratio for the Eighth and Seventh Revision would be 35 calculated as (Number of deaths by Eighth Revision)/(Number of deaths by Seventh Revision). For example, 202,894 cerebrovascular disease deaths were recorded according to 8th Revision in 1966. Death by comparable cause (Vascular lesions affecting central nervous system) according to 7th Revision in 1966 was 204,841. Thus, comparability ratio for the Eighth and Seventh Revision for cerebrovascular disease was therefore 202,894/204,841 i.e. 0.9905. A list of comparability ratios for all five cardiovascular disease groups evaluated between 1945-1995 in this study are shown in Table 3. Comparability ratios give a valuable quantitative indication of the net change between revisions. Comparability ratios may also vary with age at death, race and sex. Age, race and sex specific comparability ratio was calculated for Fifth, Sixth and Seventh revisions. A comparability ratio of 1.00 indicates that same number of deaths was assigned to a particular cause or combination of causes in both the revisions. However, a ratio showing perfect correspondence (1.00) between the two revisions does not necessarily indicate that the cause was unaffected by the changes in classification and coding procedures because the changes may compensate for each other [37]. C. Issues with comparability of data (diagnostic accuracy and change in coding) for specific cardiovascular diseases: Many factors affect the interpretation of mortality statistics on cardiovascular diseases. These may result from method of data collection and from procedures employed in the 36 classification. The effects of diagnostic procedures and coding changes for the three important cardiovascular diseases are discussed. C.1 Ischemic heart disease: a) Diagnostic Accuracy: Many IHD deaths occur suddenly, without medical attention, and with no history of the existence of cardiovascular or other disease Because there are few diseases other than IHD which are frequent causes of sudden, unexpected death, it is reasonable to attribute such deaths to IHD. Available information indicates that one-third of all IHD deaths are certified by medical examiners [35]. Although medical examiners consult physicians or hospital records in a substantial proportion of cases and make postmortem examinations in some, the diagnosis of IHD is frequently a presumptive one [35]. b) Change in Coding: For approximate comparability of data on diseases of the coronary arteries, “arteriosclerotic heart disease, including coronary disease” (ICD-6 420) should be compared with “diseases of the coronary arteries and angina pectoris” (ICD -5 No. 94) and “Chronic myocarditis and myocardial degeneration, not specified as rheumatic” (ICD-5 No. 93d) in the fifth revision. The comparability ratio for the Fifth and Sixth 37 Revisions for IHD for all ages, races, and sexes combined was 0.77 [Table 3]. It ranged among adult subgroups from a low of 0.46 for nonwhite females aged 35-44 to a high of 0.93 in white males aged 45-54 [40]. The IHD categories in the seventh revision were the same as those in the sixth, but changes in coding rules resulted in somewhat different code assignments [39]. The Seventh Revision had no special provision for classifying arteriosclerotic heart disease (ICD-7 No. 420.0) jointly reported with Hypertensive disease. Consequently, some deaths attributed to this combination of diseases were classified by the Seventh Revision to hypertensive heart disease (ICD-7 No. 440-443). Age, race and sex specific comparability ratio ranged from 0.45 (35-44 year old black female) to 0.89 (35-44 year old white female). Dual coding of a ten percent sample of deaths that occurred in 1958 supplied the basis for comparability ratios that were than applied to deaths classified according to the Seventh Revision. The overall comparability ratio for the Seventh and Sixth was 0.98 [Table 3]. Adoption of the Eighth Revision of the ICD in 1968 created a break in comparability of the IHD trend data. The comparability ratio for all age-sex-race groups estimated by the National Center for Health Statistics provided a measure of the impact of the new coding procedure. An estimated 14.57 percent more deaths (totaling 83,500) were assigned by the Eighth Revision to this title than were assigned to Arteriosclerotic heart disease including coronary disease (ICD-7 No. 420) by the Seventh Revision [37]. In particular 38 the Eighth Revision deleted the category Other Myocardial degeneration, and gave preference to IHD in classifying heart disease related to hypertension. The largest single group of these deaths (41,228) was assigned by the Seventh Revision to Other myocardial degeneration with arteriosclerosis (ICD-7 No. 422.1). Of these 41,228 deaths an estimated 40,993 (99%) were classified by the Eighth Revision to Chronic ischemic heart disease (ICD-8 No. 412). Comparability ratio by age race and sex was not computed for Eighth and Seventh Revision. Comparability ratio combined for all age race and sex is 1.063 [37]. In the Ninth Revision, a new classification, cardiovascular disease, unspecified (ICD-9 No. 429.2) was used, which accounts for the reduction in chronic ischemic heart disease grouping. Chronic ischemic heart disease grouping in the Ninth Revision is termed as old myocardial infarction (ICD-9 No. 412) and other forms of chronic ischemic heart disease (ICD-9 No. 414). Comparability ratio (Ninth and Eighth) for ischemic heart disease for all age, race and sex was 0.8784 [36]. C. 2 Cerebrovascular disease: a) Diagnostic Accuracy: There are several problems with respect to interpretation of cerebrovascular disease mortality data, including the frequent appearance of cerebrovascular diseases as a cause 39 of death along with other causes in the death certificates. For example, cerebrovascular disease and hypertensive heart disease may appear on a certificate together. Whether or not the death is classified to the cerebrovascular category depends on the sequence reported by the medical certifier. A detailed analysis was made of multiple causes of death based on the US data for 1955. Altogether cerebrovascular diseases appeared 304,004 times on death certificates, and were coded as the underlying cause of death in 173,541 cases, (57.1%) [35]. This percentage varied with type of stroke. Clinically it is difficult to distinguish between the different types of stroke, and this produces difficulties in the interpretation of mortality data [35]. Change in Coding: Vascular lesions affecting central nervous system (cerebrovascular disease) was coded as Code No. 83 according to 5th Revision and as Code Nos. 330-334 according to 6th Revision. The comparability ratio between the two Revisions adjusted for age race and sex was 1.16. The Sixth Revision also created a discontinuity in the data for vascular lesions of the central nervous system. The 16% gain was caused by change in coding rule. Most of these ‘extra’ deaths were classified into the various categories of heart disease, arteriosclerotic kidney, and diabetes when the Fifth Revision was in effect [35]. 40 In the Seventh Revision vascular lesions of the central nervous system (cerebrovascular disease) was also coded as ICD-7 Code No. 330-334. The comparability ratio the Sixth and Seventh Revision was 1.00. Actually, 940 deaths (4.9%), were assigned differently by the two revisions Cerebrovascular disease gained 466 cases and lost 474 cases by the Seventh Revision, with the resulting comparability ratio of 1.00. The Eighth Revision title for vascular lesions affecting central nervous system(ICD-7 No. 330—334) was cerebrovascular diseases (ICD-8 Nos. 430-438) [37]. The comparability ratio between these two titles of cerebrovascular disease for the Seventh and Eighth Revisions was 0.981 [37 ]. Although the ratio is close to 1.00, there were some important changes in coding procedures, which in part compensated for each other. About 4,516 deaths in 1966 assigned to ICD-7 Nos. 330-334 in the Seventh Revision were transferred by the changes in coding procedures in the Eighth Revision to categories other than cerebrovascular diseases [37]. On the other hand 2,569 deaths not assigned to cerebrovascular diseases by Seventh Revision were assigned to cerebrovascular disease by Eighth Revision. Introduction of Ninth Revision did not bring much change in cerebrovascular disease category. Same ICD coding 430-438 was used. The comparability ratio of Eighth and Ninth Revision was 1.0049 [36]. 41 C3 Hypertensive heart disease: Diagnostic Accuracy: Physicians frequently certify hypertensive heart disease on the death certificate as a contributory cause rather than as a underlying cause of death [35]. This inevitably results in underestimation of the impact of hypertensive disease on mortality. Moreover, the coding rules tend to give certain major causes of death precedence over hypertensive heart disease. When a physician reports both cerebrovascular disease and hypertension (without mention of heart) on the certificate, for instance, the death is coded to vascular lesions affecting the central nervous system. Thus, coding a single underlying cause of death in death certificates results in loss of critical data concerning the role of hypertension in mortality. The definite diagnosis of hypertensive disease must be made before death. Several studies show that in the general population a sizable percentage of hypertension is undiagnosed. When catastrophic illness like heart attack or stroke occur in such persons, leading rapidly to death, the diagnosis of hypertensive disease may not be made and therefore may not appear in the death certificate [35]. b) Change in Coding: Prior to the introduction of the Sixth Revision of the ICD in 1949, hypertensive heart disease did not appear as such in the classification. A11 deaths from this condition was 42 coded at that time to other cardiovascular disease category. In the Sixth Revision hypertensive heart disease was coded as ICD-6 Code Nos. 440-443. Hypertensive heart disease (ICD-6 Code No. 440-443) was compared with Fifth Revision ICD-5 Code Nos. 131a and 93d [40], the combined comparability ratio was 0.29. This low comparability ratio is due to changes in procedures for selecting the cause for primary tabulations; and also because the sum of the causes described by the Sixth Revision categories is not precisely the same as that described by the Fifth Revision title with which it is compared. While the examination of the inclusion terms under the titles of both revisions and of the transfers of cases from one cause under one revision may suggest some of the reasons for differences, it is not possible to determine quantitatively how much of the change in a ratio results from the incomparability of the title and /or from each of the factors which may be involved [40]. The Seventh Revision of ICD coding did not change HHD coding very much. The adjusted comparability ratio for all age, race and sex groups combined was 1.11 [39]. In the Eighth Revision cause hypertensive heart disease with or without renal disease (402,404) was compared with the Seventh Revision cause hypertensive heart disease (440-443), with a resulting ratio of 0.398 [37]. An estimated 60 percent of all deaths assigned in the Seventh Revision to hypertensive heart disease (ICD-7 440-443) were transferred in the Eight Revision to chronic ischemic heart disease with hypertensive disease (ICD-8 412.0). 43 The Seventh Revision had no special provision for classifying arteriosclerotic heart disease (ICD-7 No. 420.0) jointly reported with hypertensive disease. Consequently, some deaths attributed to this combination of diseases were classified by the Seventh Revision to Hypertensive heart disease (ICD-7 No. 440-443). In the Eighth Revision arteriosclerotic heart disease was classified as an ischemic heart disease, and four digit subcategories were provided under the ischemic heart disease categories for jointly reported hypertensive disease [37]. When the Ninth Revision List of 72 Selected Causes of Death was compared with the Eighth Revision List of 70 Selected Causes of Death for the Annual Summary, the comparability ratio for hypertensive heart disease (ICD-9 No. 402) was 3.3022 [36]. D. Age-adjusted and Age-specific mortality rates: We used age-race-sex specific mortality rate rather than crude death rate or age adjusted death rates. Crude death rate is a ratio between the number of deaths in one year and the enumerated or estimated total population in that year. It gives only an overall idea of mortality levels regardless of age, sex, race or other characteristics. Use of crude death rates for black and white comparison may give the impression of lower cardiovascular disease death rates in blacks because the black population is much younger than the white population. Age is the most crucial variable in mortality analysis and no index that fails to account for it is very useful, especially in comparing population with markedly 44 different age profiles. The common practice is to compute age adjusted death rates. The age-adjusted death rates distill age-specific death rates into a single figure and account for changes in the age structure of the population [59]. They indicate what the death rate would be in an actual population if it had the age distribution of a standard population. Any census year could be used for standard population, but the NCHS uses 1940 to compute age adjusted rates for the US. However, this common practice of age adjustment is no substitute for the age specific rates, because the selection of the age range for inclusion and the stande population can drastically influence the results obtained. The age adjusted death rate conceals any variation of trend by age or cohort. Moreover, changes in age adjusted rates by age and sex provide less insight into detailed causes than do changes in age specific rates because cardiovascular diseases declined more as a cause of death for some groups than others. Use of age adjusted rate obscure the fact that racial differences are often reversed at different extremes of age, with upper ages having much more impact on the adjusted rate than the lower age E. Race: Mortality analysis of cardiovascular diseases by race is complicated by the following problems [1 3] 45 El Mortality for blacks and other nonwhite groups not being documented separately: The past failure (prior to 1975) of the National Center for Health Statistics to report mortality for blacks and other nonwhite groups separately [1 1]. E.2 The different population age structures for blacks and whites: The different population age structures for blacks and whites making the published generalizations of black white trends using age-adjusted rates misleading. For example, age adjusted IHD rates for black male aged 35-74 years were higher than those for white males in 1968, whereas age adjusted rates for black males of all ages were lower than white male rates [59]. E.3 The greater inaccuracy of death certificate diagnoses in blacks than whites [13]. A greater percentage of IHD deaths in blacks than in whites below age 65 occur out of hospital and emergency rooms [25]. The cause of death for out hospital deaths is often based on minimal or non existent information. The death certificates are often signed as due to IHD because of lack of better diagnosis. More blacks, especially in younger age groups die outside the hospital and therefore are at greater risk of being classified as IHD deaths for the lack of better information [25]. The majority of out of hospital deaths 46 among black men, if carefully studied, are not due to IHD. In Pittsburgh, Pennsylvania, for example, only 33 percent of sudden, natural, out-of—hospital deaths among black men, ages 35-44, were due to IHD. 25% of other causes of death were misclassified as IHD deaths. Misclassification of 25 percent of the other causes of sudden death as due to IHD would increase the IHD deaths in black men by close to 50 percent at least in the younger age groups [25]. According to a 1970-1972 Baltimore study a much higher percentage of deaths certified as arteriosclerotic heart disease among black men were misclassified, 33 percent compared with 18 percent for white men [25]. The percentage of IHD deaths classified as sudden was higher in black men - 69 percent compared with 54 percent among white men. Misclassification of sudden death may also contribute to the higher IHD mortality rates among black men compared with white men [25]. EA The greater inaccuracy of census data for blacks than whites: Another problem with death rates among nonwhites is inaccuracy of population data. Greater undercounting of nonwhites in censuses compared with whites has been reported. In 1990, 7.5 percent of black males were undercounted compared to 1.5 percent of white males [25] and poor intercensal estimates would tend to inflate nonwhite rates [13]. Trends toward reduced undercounting could produce apparent decline in mortality. Consistency of reporting of age is more of a problem among nonwhites than whites [35]. 47 According to a study there was 90 percent agreement between the age recorded on death certificates and census record of whites 65 to 74 years who died of cardiovascular disease [35]. The corresponding figure for non-whites was 70 percent. For ages over 74 years, ages reported on death certificates as compared with census record were understated in about 16 percent of the deaths in nonwhite males and 18 percent in nonwhite females. For whites the net difference was only 2 percent [35]. E5 Change in ICD code: Change in ICD code from ICD-7 to ICD-8 created a large discontinuity in IHD rates in blacks between 1967 and 1968. In the Eighth Revision ischemic heart disease when jointly reported with hypertensive heart was coded under ischemic heart disease. Since Seventh Revision did not have provision for classifying ischemic heart disease in association with hypertensive heart diseases, a large number of these deaths were classified under hypertensive heart disease. Furthermore, in Eighth Revision the old terminology “myocardial degeneration” was eliminated and most of the deaths due to myocardial degeneration was coded in ICD-8 under ischemic heart disease [39]. Change from ICD-7 code to ICD-8 code, in which IHD diagnosis takes precedence over hypertensive heart disease and the old terminology “myocardial degeneration” has been eliminated would affect nonwhite rates much more than whites, because (1) hypertensive heart disease is more prevalent in blacks and (2) myocardial degeneration is more frequently used in blacks to designate IHD [13]. 48 In ICD-9 however, when hypertensive heart disease and ischemic heart disease was jointly reported, it was coded under hypertensive heart disease. This was also found to have more affect on blacks than whites. Thus, the changes from ICD-7 to ICD-8 and ICD-8 to ICD-9 were associated with substantial artifactual upward and downward shifts in the IHD rates in the black population [13]. E6 The greater impact of influenza epidemics on CHD in blacks compared to whites. Influenza epidemics cause more mortality in blacks than whites. Since, respiratory infection is often a complicating cause in cardiovascular deaths, influenza epidemics run in parallel trends with heart diseases [13]. Cardiovascular disease mortality rose sharply in blacks during December 1968 to January 1969 epidemic. Racial differences in percentage decrease in cardiovascular mortality were less after the epidemic. Chapter III Methods: A. Data sets: 1. Data sources: The number of cardiovascular disease deaths in Michigan for adults 20 years or older was collected for the years 1944 to 1995, categorized by race (black or white ), age (by 5 year category) and sex. The number of deaths due to the five major categories of cardiovascular disease (Major cardiovascular disease, Diseases of heart, Ischemic heart disease, Cerebrovascular disease and Hypertensive heart disease) was tabulated from the vital statistics of the United States, Natality and Mortality Data for the United States, Tabulated by place of residence. All data were tabulated by place of residence rather than place of occurrence because in most cases, both will be the same and because risk of death from chronic diseases is usually related to long term risk factors prevailing in resident populations. We grouped age at death in five year intervals and restricted our analysis to 11 age-groups, from 20-24 years through 70-74 years prior to 1970. We further analyzed our data for 13 age groups from 20-24 years through 80-84 years after 1970. The lower age restrictions avoided highly variable estimates due to small number of events that occur at very young ages. We could not include higher age groups because 49 50 population figures in 5 year age groups were not available for older ages prior to 1970. We decided not to include 75+ and 85+ groups in our analysis because analyzing data with age 75 and over or 85 and over as one age group will result in too much lumping of data, and will not allow us to separate mortality rate in different age groups. 2. Count Data: The number of deaths for each disease category was calculated for each five year period (i.e. 1945, 1950, 1955 ...... 1995). We used an average of the number of deaths of three consecutive years around each 5 year period. For example, number of cerebrovascular disease deaths for 1960 was calculated as an average of the number of deaths of 1959, 1960 and 1961 [Table 4]. Three year averages reduced the effect of annual fluctuation in mortality and the possibility of significant misreporting in any one year. Since 1996 mortality data were not, yet available, for 1995 we calculated only the average of 1994 and 1995. 3. Population Data: Population data from 1971 to 1995 were obtained from Michigan Department of Community Health (MDCH). These data were tabulated by age (5 year age group), race (black or white) and sex for every year. Intercensal population estimates were provided to MDCH by the state department of Management and Budget. Prior to 1970 population 51 data by age, race and gender was not available for every year. Population data classified by age, race and gender for 1940, 1950, 1960, and 1970 were obtained from decennial census. Intercensal estimates were then made using linear interpolation. Decennial census data were obtained from MDCH. An example, of linear interpolation between 1950 -l960 for white males is shown in Table 5. Prior to 1975, reporting of deaths for all races other than whites were grouped as non- whites. For Michigan, non-white rates were probably a fair approximation of rates for blacks, in as much as blacks comprised greater than 85% of the non-white population in those years [11]. Furthermore, results of linear interpolation are based on the assumption that there has not been any large migration of population in and out of Michigan. Large migration in and out of Michigan from 1940 to 1970 may have affected both the number of events and actual population size but may not be immediately reflected in the population estimates. We inserted the count data (i.e. number of deaths for each disease category and the population data) in Excel. Underlying cause of death was categorized by the Ninth Revision of the International Classification of Diseases. We calculated age-race-sex specific death rate for each of the 5 disease categories by constructing simple fractions, where the numerator is the number of deaths during one calendar year (estimated as an average of three consecutive years), due to the cause of interest in a given age-sex-race group and the denominator is the estimated size of the relevant population. The fraction 52 is reported as the observed mortality rate per 100,000 persons at risk. Thus the formula for determining IHD death rates in black women aged 20-24 in 1970 is given by: (Number of deaths due to IHD among black women aged 20-24 in 1970)/(Total number of black women aged 20-24 )* 100,000 The period (1945-1995) under review encompasses five revisions of the International Classification of Diseases (ICD): the Fifth (1939-1948), Sixth (1949-1957), Seventh (1958-1967), Eighth (1968-1978) and Ninth ( 1976-1996) (Table 3). Each of these revisions of ICD codes altered the coding rules and resulted in lack of comparability with the previous code. To make the rates of 1945 comparable to 1995 rates we have to take into account of these changes. For example, to make the mortality rate due to IHD in 1945 to be comparable with the 1995 rate, the 1945 rate must be multiplied by the relevant comparability ratios, i.e. 1945 rate * OR. of 5th and 6th Revision for MCD deaths * OR. of 6th and 7th Revision for MCD * C.R. of 7th and 8th Revision for MCD * OR. of 8th and 9th Revision for MCD. For example to make 1945 MCD rate for black male 50-54 year old comparable to 1995 rate we applied the following procedure: 53 917(1945 rate) * 1.17(C.R. 6th, 5th) * 1.00(C.R. 7th, 6th) * 0.991(C.R. 8th, 7th) * 1.0069(C.R 9th, 8th) =1079. The rate, for 50-54 year old black males for 1945 (917) when adjusted to be comparable to 1995 rate was 1079. The 1945 data from the Michigan population did not code ‘chronic myocarditis (93d)’ as a separate group. Instead, it was included in one category with several other conditions in diseases of the heart (other forms) [other pericarditis (90b), acute endocarditis (91), diseases of aortic valve (92a), other chronic endocarditis (92d), endocarditis unspecified (92e), acute myocarditis (93a), myocarditis unspecified (93b), other chronic myocarditis (93d), other myocarditis unspecified (93c), functional diseases of heart (95a), other diseases of heart, not rheumatic (95c)] in the category diseases of heart (other forms). Using the national data we were able to estimate chronic myocarditis (93d) mortality by examining the proportion of deaths due to chronic myocarditis relative to total mortality due to diseases of heart (other forms) in the national data. By applying this proportion to the disease of heart in Michigan population we estimated the chronic myocarditis (93d) mortality in Michigan for 1945 for each age, race and sex group. [Table 6]. The formula used was to estimate 93d was {93d in US/diseases of heart(other forms) in US}* diseases of heart other form in Michigan. For 1950, onwards mortality data for IHD did not pose any further problems. 54 B. Analysis Data analysis was done in two parts: 1. Period analysis 2. Cohort analysis 1. Period analysis: Age-specific mortality rates for major cardiovascular disease were presented by gender and race in five year age groups [See Data 0001 to 0020]. In these Data Sets, the rows represents categories of age at occurrence and the columns defines categories of year of occurrence. The rates were based on 5-year age intervals and 5-year period intervals. Data were plotted with the rate per 100,000 as the ordinate and the age at death category as the abscissa. Each graph was standardized for each disease cause. The linear trend graphs were superimposed in lower age groups and some were not even evident . So, mortality rate was plotted on a log scale to accommodate the wide range of rates. The same process was repeated for other disease entities. A separate graph was plotted for each race and gender group [See Figure 0001 to Figure 0004]. 55 The average annual percentage change in age-race specific mortality rate for each 5 year period (for each of the cardiovascular diseases category) was calculated. For example, average annual percentage change for IHD deaths in white males for the period 1950- 1955 was calculated as the 1955 rate (WM age 50-55 age) minus 1950 rate (WM 50-55 age) divided by 5, divided by 1950 rate and multiplied by 100 [Table 7(i)]. 2. Cohort Analysis: To examine mortality trends in a different way, we computed mortality rates in 5-year birth cohorts from 1873 to 1973 for each of the 5 disease groups. Age, race and sex specific mortality in these cohorts aged 20 to 84 is shown in Data Table 0021 to 0040. However, in birth cohorts prior to 1888 we could not include 75-79 and 80-84 age groups, since denominator data were not available prior to 1980. We took the central year of age group i.e. for 70-74 year age group we took 72. We deducted 72 from 1945 to identify the birth cohort and to give us the mortality rate of persons born in 1873 (1945-72 = 1873) and died when they were 70-74 years old. For example, IHD mortality rate in 1945 for white male 70-74 year age group was 946 per 100,000. To construct a cohort chart this mortality rate will be for 1873 (1871-1875) birth cohort of white males aged 70-74. Similarly, DOH mortality rate for 50-55 year old white males in 1970 was 350 per 100,0000. This group of white males will constitute the (1970-53 = 1917) 1917 birth cohort of white males aged 50-54. 56 We plotted mortality rate per 100,000 as the ordinate, year of birth instead of year of death as the abscissa and stratified it by age. To accentuate the difference in mortality rate in lower age groups we plotted mortality rate per 100,000 in log scale. This was particularly helpful to show the differences in lower age groups where the number of events was very small. For cohort graphs there were fewer data points in the two extreme ends of ages. For example for persons born in 1873 we can only include 70-74 years old and for people born in 1973 we can only include 20-24 years old. For each disease group a separate graph was plotted for each race and sex group [Figure 0021 to Figure 0040]. Chapter IV - Results A. Major Cardiovascular Diseases: White Males: Figure 0001, Data Set 0001, and Table 7a each presents age specific mortality rates in white males over 5-year successive period over a fifty year span beginning in 1940 and ending in 1995. An overall trend of consistent secular decline in mortality rates emerges from a review of this data set. In a handful of instances, in particular years certain age groups reported mortality rates that are inconsistent with the overall trends i.e. instead of a decline, they in fact report a slight increase. These inconsistencies may be derived from extraordinary events impacting on a relatively small sample size. From 1945 to 1965 the rate of decline of mortality rate was relatively small. In 1945, 673 per 100,000 white males died and in 1965, 599 per 100,000 white males died of MCD. The average rate of annual decline was only 0.57 percent. The linear graphs (1945-1965) also show a series of almost parallel straight lines, indicating very little change in the rate. The rate of decline accelerated after 1965 and continued to do so until 1990. Mortality rate in 1990 was almost half that of 1965. However, the rate of decline in MCD-related mortality substantially slowed in the period between 1990-95 as compared to the earlier periods studied in this review. 57 58 Although the data reveals a constant pattern of overall declines in mortality rates, age- specific mortality rate for every period increased with age, particularly for over 50 year age groups, suggesting what has now become obvious -- that aging population groups carry a higher risk of cardiovascular diseases. Data Set 0021 shows that in 65-69 year age group 1888 birth cohort had the highest mortality rate (2577). In 60-64 year age group however, 1893 birth cohort had the highest rate. Men born in 1888 were 65-69 years old in 1955 (1953-1957) and men born in 1893 were 60-64 years old in 1955. So, a period effect and not cohort effect is evident. If a cohort effect would be present a particular birth cohort would have the highest rate in all ages. White Females: Table in Data 0002 and Figure 0002 show age specific mortality rates in white females aged 20-84 from 1945 to 1995. Mortality rate in white female declined steadily from 1945 to 1965. During 1945 to 1965 average annual rate of decline in 50—54 year olds white female was 2 percent. From 1965 to 1995 the average annual rate of decline was also 2 percent. The rate of decline was consistent among all age groups. However, the decline of mortality rate was highest during 1980-1985, and it was more pronounced in the older age groups (55-84). For example, 80-84 year old women experienced an 59 average annual decline of 4 percent during this period compared to 0.5% among women of 50-54 year age group. Figure 0002 shows a consistent decline in mortality rate through out our study period. The cohort graphs do not show any cohort effect. If strong cohort effect would be present, it would not be evident in the period graph. Black males: Age specific mortality rate in black males showed a pattern similar to white males, except in the period 1990-95 when black males registered an increase in mortality rate while the overall declining trends continued for all other groups. As Data Set 0003, Figure 0003 and Table 7c demonstrate age specific mortality rate declined steadily in all age groups from 1945 to 1965. The rate of decline accelerated after 1965 and continued until 1995. However, death rates in age groups 55 to 69 increased during 1990-1995. It increased at an annual rate of 2 percent among 55-59 year old men, at 0.8% among 60-64 year old men, and at 0.7% among 64-69 year old men. The cohort Data 0023 and Figure 0023 also do not indicate any cohort effect. Black males aged 60-64 had the highest rate in 1883 birth cohort. and 65-69 year old men had the highest rate in 1888 birth cohort. No particular birth cohort showed very high rates. 60 Black females: Overall mortality rate in black females declined during the study period (1945-1995). The rate of decline of mortality among Black females of all age groups reached accelerated in the period between 1970-75. During this period, 60-64 year old black females experienced an annual reduction of 8.2%. In the same period, Black males of the same age group, registered a decline of just 2.6%. The rates continued to decline throughout the 19805 and the 19905. In 19905 the rate declined in all age groups except 40-44 age group (6.3% increase) and 50-54 age group (5.6% increase) [Table 7d]. This increase in mortality among these two particular age groups of Black females mirrors a similar trend of increase of mortality rate among Black men during this same period. The Cohort Table Data 0004 and Figure 0004 do not show any cohort effect. Black females aged 65-69 show highest mortality rate in 1888 cohort. 70-74 year old black females showed highest mortality rate in 1873 birth cohort. No particular birth cohort was at higher risk for cardiovascular disease death. Cardiovascular disease mortality rate was higher among black males at ages 20-69. In age 70 and higher age groups white men had higher rates. White females had the lowest cardiovascular diseases death rates. Larger percentages of white males than white females die in every age group from cardiovascular diseases, although the gap is narrower in some ages than others. The mortality differential by sex varies considerably with age. 61 In 1975 death rate of white males 70-74 years was 1.9 times that of white females [Table 7b]. Cardiovascular disease mortality rates are 2 to 3 times higher in black females than in white females. This mortality gap due to racial difference widens in middle ages and converges in older age. However, the death rate for black females were well below those of black males. Death rates also declined faster in black females than black males. B. Diseases of Heart: White Males: Data Set 0005, Figure 0005 and Table 7e present age-specific mortality rate in white males in Michigan from 1945 through 1995. DOH death rate decreased in virtually every age group for every time period after 1955. However, the greatest declines have been since 1985. Again, during this period (1985-1995) 55-59 age group has most benefited. They experienced an annual decline of mortality rate of 4.7 percent [Table 7e]. During 1950-1955 DOH declined in all age groups except 60 to 69 year old men. In 1950, 1305 per 100,000 white males 60-64 years old died. In 1955, 1362 males per 100,000 white males of the same age group died. The rate of decline accelerated after 1970 and was evident in all age groups. The death rates decreased most during the years of 1985 to 1995. However, not all age groups benefited equally from this declining trend. For example, during 1985-1990, 40-44 year old age group experienced the most 62 reduction in rate (6% decline) and in 1990-1995, 70—74 age group experienced maximum decline (5% decline). Cohort Table ( Data 0025) and cohort graph (Figure 0025) do not show any evidence of cohort effect. 50-54 year old white male had the highest mortality rate in 1903 birth cohort. Similarly, 65-69 year old men had highest death rate in 1888 birth cohort. White Females: Data 0006 shows age specific DOH mortality rate in white females in Michigan from 1945 to 1995. DOH mortality rate decreased in all age group during this period (1945- 1995). The rate of decline, accelerated after 1970 and was more pronounced after 1985. Average annual rate of decline was three times more during 1970-1995 (3.5 %) than in during 1945-1970 (1.1 %). DOH mortality rates do not show any cohort effect. In Data 0026), 70-74 year old white females of 1893 birth cohort had higher mortality rate than 1888 birth cohort (1647 versus 1621). However, in 65-69 year old group 1893 birth cohort has lower mortality rate than 1888 (921 versus 1027). 63 Black males: Table in Data 0007 and Figure 0007 presents age specific death rates in black males in Michigan from 1945 to 1995. The rate of decline showed a very complex, almost bewildering pattern with declines for one period followed by increase in the next. Different age groups in the same period also showed very different rates. Consistent decline were seen only for 1970-1975 and 1980-1985. During 1990-1995 death rates increased in black males except in ages over 70. In Cohort Table Data 0027 and figure 0027 shows no cohort effect in black males aged 20-84 during 1945-1995. Highest mortality rate in 65-69 age group was in 1898 birth cohort, in 60-64 age group it was however in 1888 birth cohort. No consistent pattern was evident in cohort graphs. Black female: Data 0008 denotes mortality rate in black females in Michigan from 1945 to 1995. The DOH mortality had an overall declining trend, but this declining trend was not evident in every period. The rates consistently declined only after 1980. From 1975-1980, however the rate increased in all age groups. Cohort Table Data 0028 and Figure 0028 shows no consistent pattern in the mortality rates of the birth cohorts. No particular birth cohort experiences higher death rate in all age groups than the other birth cohorts. In age 60-64 black women born in 1918 had higher mortality rate than black women born in 1913 (649 versus 605). In age group 55- 59, 1918 birth cohort shows lower mortality rate than 1913 birth cohort (397 versus 493). C. Ischemic Hear Disease: White Males Table in Data 0009 and Figure 0009 present age specific mortality rate in white males from 1945 to 1995. Ischemic heart disease in white males increased steadily from 1945 and reached a peak in 1965. After 1965 it started to decline and continued to do so until 1995. During 1945 to 1965 the rate of increase was not same in all age groups. It was higher in older age groups (age 65 and over). For example, mortality rate for IHD in 70- 74 year old men was 1350 per 100,000 in 1945. In 1965 it was it was 2257 per 100,000, representing a 70 percent increase. During the same time period mortality rate increased only 30 percent in 50-54 year old men. After 1965 mortality rate started to decline in all age groups. This decline was most pronounced since 1980. Data Table 0029 and figure 0029 shows age specific mortality rate by birth year. Cohort effect is not evident in white males. 70-74 year old white males show highest mortality 65 rate in 1893 birth cohort. Similarly, in age 65-69, 1898 birth cohort shows highest mortality rate. Examining cohort data it is evident that highest mortality rate in age groups follow a diagonal pattern. If a cohort effect would be present a particular birth cohort would show largest mortality rates throughout all age groups. The diagonal pattern of the cohort table indicates a strong period effect rather than a cohort effect. White Females: Data 0010 and Figure 0010 depicts age specific mortality rate in white females from 1945 to 1995. The mortality rate for White females was unchanged in all ages from 1945 to 1965, except those 65 and over. Women 70-74 year old experienced almost 40 percent decline in that period. After 1965 the rate began to decline for all groups and has continued to show substantial consistent declines ever since. Like white males white females also do not show any cohort effect [Data Set 0030 and Figure 0030]. In 70-74 year old women highest mortality rate was in 1893 birth cohort. Highest mortality rate in 65-69 year old women was not in 1893 birth cohort but in 1898 birth cohort. Like white males cohort analysis shows a diagonal pattern rather than vertical pattern indicating absence of cohort effect but a strong period effect. 66 Black Males: Table in Data 0011 and Figure 001 1 indicated age specific mortality rate in black males during 1945 to 1995. Mortality rate of black males increased rapidly and almost equaled that of white male by 1965. For example the average annual increase in 70-74 year old men during the period 1945-1965 was 3.9 percent. During 1965 to 1970 IHD rates began to show decline among men 50 years and older. Between 1970-1990 it declined in all age groups. However, since 1990 IHD rates have increased in men less than 50 years. Black males also showed no cohort effect [Data 0031 and Figure 0031]. Mortality rate in 75-79 year old men was highest in 1893 birth cohort. In 70-74 year old men it was highest in 1898 birth cohort. If a cohort effect would be present highest mortality in 70- 74 year old men would also be in 1893 birth cohort. Black Females: Table in Data 0012 and Figure 0012 shows age specific mortality rate in black females during the study period (1945-1995). Age specific mortality rate peaked in 1965. After 1970 it decreased in all age groups, particularly among the younger age groups. Declines have been consistent until 1990, though they appear to be larger in younger age groups. During 1990-1995 evidence shows that the rates have increased in younger ages(<55 years). 67 Cohort analysis in black females do not show any cohort effect [Data 0032 and Figure 0032]. Highest mortality rate in 60-64 year old black females was in 1903 birth cohort. However, highest mortality rate in 55-59 year old women was in 1908. No particular birth cohort was at risk higher risk of IHD deaths. D. Cerebrovascular Diseases: White Males: Table in Data 0013 shows age specific mortality rate in white males from 1945 to 1995. With few exceptions age specific mortality rate declined steadily from 1945 to 1995. The rate of decline accelerated after 1970. The average annual rate of decline from 1970 to 1975 was 3.5 percent. Thus, the rate of decline almost tripled during the second half of the study period. During 1990 to 1995 the rate of decline diminished in older age groups (55-84). And it even increased in some age groups (35-49) [Table 7m]. Cerebrovascular disease death rates in the birth cohorts 1873 to 1973 did not show any cohort effect [Data Set 0033 and Figure 0033]. In 70-74 age group mortality rate was highest in 1878 birth cohort. However, in 65-69 age group it is highest in 1888 birth cohort . If a cohort effect was present mortality rate would be highest consistently in all age groups in a particular birth cohort. 68 White Females: Cerebrovascular disease started to decline in white female from 1945. The rate declined steadily in all age groups. The rate of decline was accelerated in 1970 and continued to do decline until 1995. Death rates declined more than 50 percent between 1945 and 1965. For example, in 50-54 year old men it was 99 per 100,000 in 1945 but only 44 per 100,000 in 1965. Death rate declined more than 60 percent during 1970-1995 [Table 7m]. Cerebrovascular disease in white females did not show any cohort effect. 70-74 year old white females had the highest mortality rate in 1923 birth cohort but 65-69 year old women had highest mortality rate in 1918 birth cohort. No particular birth cohort had highest mortality rate in all age groups. Black Males: Mortality rate in black males started to decline in all age groups after 1960 and continued to do so until 1990. The rate of decline accelerated after 1965. During 1970-1995 the rate increased in most age groups. For example in 55-59 year old men experienced 4 percent annual average increase in mortality rate. 69 Black males did not show any birth cohort effect [Data Set 0035, Figure 0035]. 55-59 year old black males had highest mortality rate in 1893 birth cohort. However, in 1898 cohort, 50-54 year old black males had the highest mortality rate. If a cohort effect would be present a particular birth cohort would have highest mortality rate in all age strata. Black Females: Cerebrovascular disease in black male started to decline in 1950, particularly for the younger age group. The rate of decline was higher during 1970-1975 and 1980-1985. The rate of decline diminished and even increased in some groups during 1990-1995 ( 40- 59, and 70-74 age groups). A cohort effect is not evident in 20-84 year old black females born during 1873- 1973[Data Set 0036, Figure 0036]. Largest death rate in 65-69 year age group is seen in 1893 birth cohort. In 60-64 age group, however, highest mortality rate is in 1888 birth cohort. E. Hypertensive Heart Disease: White Males: 7O Mortality rates in 30-84 year old white males for hypertensive heart disease from 1950- 1995 is showed in Data 0017 and Figure 0017. Mortality trend was studied from 1950 and 30 year old was the youngest group included since, we did not have data under 30 years of age. It was also very difficult to compare 1945 data to 1995 data because of drastic changes in ICD codes. The rate declined in most age groups from 1950-1990. The rate of decline accelerated in 1970 and continued to decline until 1990. During 1975-1980 the rate of decline was more for the older age groups(age 60-84). For example, the rate of decline was 12.1 percent in 75-79 age groups during 1975-1980 but only 2.4 percent for 50-54 year old men [Table 7q]. White males did not show any cohort effect for HPD mortality rates in birth cohorts 1878-1963. For 75-79 age group mortality rate was higher in 1918 birth cohort than in 1913 birth cohort[Data set 0037]. In 70—74 age group, however mortality rate was higher in 1913 birth cohort than in 1918 birth cohort. If a particular birth cohort would have higher risk of CD death mortality rate in that birth cohort would be higher in all age strata. White Females: The Data Set 0018, Figure 0018 and Table 7r, presents age specific mortality rates for white females from 1950-1995. Mortality rate for white female declined from 1950-1985 in all age groups (with few exceptions). In 1950 mortality rate in 60-64 year old white 71 female was 153 per 100,000 and in 1965 it was only 40 per 100,000 i.e. that is almost 1/4 the rate of 1950. Mortality rate continued to decline until 1985. The rate continued to rise in most age groups during 1990-1995. White Females did not show any birth cohort effect. In 55-59 age group, 1913 birth cohort had higher mortality rate than 1918 birth cohort. However, in 50—54 age group 1918 birth cohort had higher mortality rate than 1913 birth cohort. Mortality rate was not consistently higher in a particular birth year. Black Males: There was an overall decline in mortality from 1950-1995 [Data Set 0019, Figure 0019 and Table 75]. However, black males showed complex pattern, marked increase and decrease depending on age group and time period. Mortality rate decreased from 726 to 88 per 100,000 in 65-69 age group during this period, representing an overall decrease of 87 percent. Black males aged 30-84 in birth cohorts 1878-1963 did not show any evidence of cohort effect in cerebrovascular disease mortality [Data set 0039 and Figure 0039]. 50-54 year old black men showed higher mortality rate in 1913 birth cohort than in 1908 birth cohort. In 55-59 year old black males, however, higher mortality was in 1908 birth cohort than in 1913 birth cohort. 72 Black Female: Mortality rate from hypertensive heart disease is presented in Data Set 0020, Figure 0020 and Table 7t. Mortality rate of black females had more of a consistent pattern than black males. .Mortality rate from hypertensive heart disease in black females declined rapidly from 1950-1995. Age specific mortality rate in 60-64 year old women dropped from 963 to 41 per 100,000. Black females experienced the largest decline during 1970-1975. In 60-64 age group it declined more than 60% during this period. During 1990-1995 rate increased in 50-54 age group (13.5%) and in 55-59 age group (9.2%) [Table 7t]. No cohort effect is evident in 30-84 year old black females born during 1878-1963 [Data set 0040 and Figure 0040]. In 50-54 age groups higher mortality is observed in 1917 birth cohort than in 1913 birth cohort. The trend reverses in 45-49 year old age group, mortality rate is more in 1913 birth cohort than in 1918 birth cohort. A consistent cohort trend is not evident. Chapter IV 1. Discussion In this study we demonstrated variation of cardiovascular mortality in Michigan by time, race and gender. An overall declining trend for all cardiovascular diseases was evident throughout our study period (1945-1995). However, if we analyze each period separately, we find that the rates of decline have not been the same during successive periods of analysis. From 1945 to 1970, the rates declined slowly but steadily for all race and sex groups. During this period decline in mortality can be attributed almost entirely to decreases in fatalities from cerebrovascular diseases and hypertensive heart diseases. During 1970 to 1975 the rates dropped substantially in all race and sex groups. The fall in mortality rate was almost equal among the five- year age groups. However, the decline in mortality rate was disproportionately high for blacks, particularly for black females. After, 1975, the downward trend continued but now in a slower pace in blacks than in whites. During 1990-1995 cardiovascular mortality increased in black males (age 40-44 and age 45-49). Ischemic heart disease (IHD) deaths increased in Michigan in all four major race and sex groups from 1945 to 1965. The rise in mortality was most pronounced among black males. Beginning at a much higher initial level than the black males, the rates of white 73 74 males also rose moderately. The first downward trend in IHD mortality in Michigan became evident in 1965, and we saw a sharp decline during 1970-75. This decline among black females was double that of white males. After 1975, trends for blacks and whites began to diverge, with a deceleration in the annual fall in rates for blacks. This diverging trend continued, and in 1990 for the first time in Michigan more black men than white men aged 45 to 55 were dying from IHD. However, the death rates for black females were more than that of white females since 1960. During 1990-1995 the rate of decline in mortality decreased in all race and sex groups. The mortality rate even increased in 30 to 54 year old blacks. Death rates from cerebrovascular disease have been steadily declining in Michigan throughout our study period. Black males had the highest mortality rate. In 1945 data we observed the decline of mortality in white males, white females and black males. For black females the rate started to fall after 1950. The rate of decline accelerated in whites during 1975 to 1980 and in blacks during 1970 to 1975. The rates continued to decline thereafter.. During 1990-1995 the rate of decline decelerated in all race and sex group, and even increased in some groups. Mortality rate for cerebrovascular disease increased in 40 to 50 year old black men and 40 to 54 year old black women. Hypertensive heart disease started to decline in 1950 and the greatest reduction in mortality rate was observed during 1975-1980. However, during 1990 to 1995, death 75 rates from hypertensive heart disease increased in all races and genders except black females. The interpretation of divergent trends of mortality is difficult since our ability to identify clear cut cause and effect relationship is limited. However, a number of factors could potentially have contributed to the slower rate of decline in cardiovascular disease deaths among blacks. Decrease in cardiovascular mortality is accompanied by l) a decrease in the prevalence of risk factors, 2) a decline in the incidence of disease and 3) increased access to medical care. 1. Trends in the prevalence of risk factors: Hypertension , high serum cholesterol levels and smoking have been shown to be independent risk factors for cardiovascular diseases. The best sources of data on risk factor levels in the US population is the National Health and Nutrition Examination Survey. A. Hypertension Three National Health and Nutrition Examination Surveys were carried out in the United States. 76 National Health and Nutrition Examination Survey N HANES -1 (1971-1975) [47] National Health and Nutrition Examination Survey NHANES H (1976 to 1980) [48] National Health and Nutrition Examination Survey NHANES III (1988-1991) [50] From NHANES I to NHANES II the estimated prevalence of definite hypertension (BP greater or equal 160/95 mm Hg and /or currently taking antihypertensive medication) was unchanged among whites, while a modest decrease was noted in black males and females(3-4%) . However, prevalence of hypertension increased from NHANES II to NHANES III for black men ages 65 to 74 years. This national data is in consistent with our findings that hypertensive heart disease, and cerebrovascular disease both declined more in blacks from 1975 to 1985. The increase in rate in hypertensive heart disease and cerebrovascular disease in black males during 19905 is also in agreement with the national data. B. Serum cholesterol level: Between NHANES I and NHANES II serum cholesterol levels declined by approximately 3-4 mg/ 100 ml for blacks and white males and 1 mg/100 ml for white females. Total cholesterol levels fell from NHANES II to NHANES-III in all four race and sex groups. The decrease in the percentage of persons with high cholesterol levels was similar for blacks and whites. 77 C. Obesity Both black men and women had much greater increases in prevalence of overweight. than did whites. Distribution of body fat also varies among blacks and whites. Central obesity and diabetes is a stronger predictor of cardiovascular disease. A 28 year follow-up study of US women showed more central obesity and diabetes in blacks [43]. D. Smoking : According to 1991 data 29% black males smoke compared to 26% white males. Among black women, prevalence of during smoking was slightly higher in 1991 than 1990 [43]. Based on the 13 year ischemic heart disease data from the First National Health and Nutrition Examination Survey Epidemiologic Follow-up study, cumulative incidence rates were 24 percent for white men, 22 percent for black men, 14 percent white women and 15 percent for black women. Overall, it would appear that earlier data from 19605 and 19705 showed lower incidence rates for blacks, while in mid 19805 the rates are similar in blacks and whites. E. Socioeconomic status: 78 Socioeconomic forces within a community can raise or lower the risk associated with cardiovascular mortality. The adverse trends in cardiovascular disease mortality among blacks have occurred due to the overall worsening of the social and economic condition of blacks. The disproportionate fall in rates among the educated and rich strengthens this nofion. F. Access to medical care: A striking consistency has emerged in the studies of racial differences in access to care for cardiovascular diseases. In data sets obtained from private hospitals, Medicare and Veterans Administration indicate that blacks are half as likely to have bypass surgery with similar angiographic findings [25]. 2. Conclusions: The factors that have led to the decline in cardiovascular mortality in Michigan did not affect all four major race and sex groups equally. As the result of the divergent trends among blacks and whites, Michigan vital statistics data show IHD mortality of blacks now exceeds that of whites for both men and women. Although the rates continue to decline in all groups in Michigan, the total benefit of the major public health advances achieved against the coronary epidemic has not been realized among blacks as it has among whites. Furthermore, the rate of decline of cardiovascular diseases in all race sex 79 group declined in Michigan during 1990 to 1995. In Michigan, after 1990 ischemic heart disease increased in black males age 35 to 44. Cerebrovascular disease and hypertensive heart disease also increased in 19905 in almost all ages in black males. Black females also showed a marked increase in cerebrovascular disease in some age groups. The results emphasize the need for increased efforts aimed at primary and secondary prevention and access to appropriate treatment in Blacks. However, since 19905 have been the decade of less improvement in all groups. Efforts must be directed also towards improving the gains already made on whites. 80 mBano E 8:: 5:9; um; am a Em .30 mod 5 was :5 .29; um; .22, w: ho mtma .650 can“ own. own. new m_nE:_oO c_ 559; 9m; mom: =2< 382.998 .52. .o mucoomm 5381 do .550 *0 wm_mE Emoxo 5.3... .o >EmtoE 5583223 EwEtmamD 8me E0: 5 8:08 .02 .85 __< 89 9% 8:267. .55 megs; 8,: 385 E .29; mm; E09 >ocm9mEo new .2230... *0 So 835m 5.81 .2 .250 ucanoo mémmovoow .20sz E0: Ema mozmzfim 5.8: .2 3.58? .250 .95sz E0: So: 85.8w can 2% 2:89: m3 752 8-0va E0: dm xom new mEmEoE mom. __m .2 $2885 6 naEmocoE on. E0: 86. 2:986 or. 98 Euros 8-2.2 mmEE 9.53 3098 3305 __m E 33905 $5.03 .0 9m. moan? E0: whmomr E0: 3:06 85sz 5.81 __m E m: E 85.08 .2 .250 .98sz E0: 88. £985 9... 58 £955. magnet Sac do 858 8:295 3:030 .m m_mEmn_ w Ems. xogm 36.89»: .mBano m_mEo..._ w 222 223 2:04. an? a men men. .550. new oEmawf mmwmomfi. 8me 5.92m .222, $385,228 08 782 5.58 ea: 520 new 98> mm m_mEmn_ w 222 xom_m Emma Emcoaoo 28.8 a was. 232, 58.5 82-83 866 8:5 seem“. a 922 x85 81.50835 228a w was. 932, :8: 2852 m3 79.2 865 8%: 2man a was. xomE SI: oedema QmEou. a 222 933 two: 2828. mam 789 33.6 nose: coaom cozmsaom bzcw omega bored Egg—.102 mwoam<0 “.0 wnzmm... 20 mmazhm F m.._m<._. 83 “mm mom? gm 8 .mo 95203 .mo 988% m... =8. .8. “mm 23% a .32 2238 .m< «688. <6 80> .3 _ ”mm 82.2.2.5 .a 922m E “mm £2 ”a a .< .96 4. EESw .m .8000 :9; “mm mam? gm 6 .02 .93.. .m .8000 .0 manmw 8:335 do ..> .552 81 0000 .00.0 0. .030. >..00.0.0E 0. .00 0000 .00000> .0 02.0.5 .2 .0... 2 0.0.0 0. 20. 9.9.... 2.0%.: .000... 000 00000.. 0. 00.003 0. 3:0..0E 00.020 0000xm. 000? 2 000. 000.200 00.003 .2 0.0E .00 0000.0 :0 0. 005.000 020. 5000 00.0000 00< 020. .000... 0030.. 00.005 00000.0 0300035200 .2 oat. 00.... 00022.0 2.0.5... 0.332000 020. .000... 00.00 00.005 .3..0..0E CI. .2 0000 00; 00022.0 2.023.005 00-2.2.2 .0 00000 500 00000 000000 0.0E 000.. w 003 000.200 0 00-2.2.2 0.> or 000. 00000 000000 0.0E2 000, w ommp 000.200 02.0000 .620 0502.83 .000002 E02 0.00 0000020 5.00: .2 .2000 000.02 E02 0.00 8.0.0.0 5000 .2 .2000 00000.2 8.0.0.0 5000 .2 .2000 00000.2 0... 9.000 0.0E0u. 000 0.02 02000.... 0.0.000 0 0.02 0.00.0 00000.0 0.0E0u. 0 0.02 2...>>_0_0000>0.00.00 000 7000p 020% 02.0: .00. 000 .000? ..0 .0 .0 0.000 .0. 000.000< .0 0.02.01 .0: 000 52 0.3.0 .0 0.06.0 00.50 0 0.0.2 0.000 22.00 0 20.2 0.22. 00000.0 000.. 2.2.2.8.... 000 .002 00.0.0 02.5 .0. 000 ..0000.2 000' 0000220 .0 0000. 0.0E0n. 0 0.02 0.005 00000.0 200.00 0 00.2 9.52080085900 52-002 .0. 000 00000.2 0009 0000.520 .0 000m 0.0E0n. 0 0.02 v.00.m 0 00000.0 200.00 0 20.2 2.55 000.. 0.0.2.8. 50 :02 00000.0 030005000 .0 00.. 00000 003 0 80F .80? 0. 0.00.. 00-00 0000 00.000 0000005 000000000002 000. ..0 .0 .< .0000.0m 005.000 3:0..0E 000.00 00.0 000... E0000_E0.n. 0.0E0n. 0 0.02 2.03 .0.0000>0.0.00 00.0 7000' .E0000.E0.n. .00_.000<.0 .—->—< ICARDIOVASCULAR DISEASES IN MICHIGAN I I I I I I f I 1 I I ICHANGE OF MORTALITY RATE FOR MCD IN WHITE MALES IMAJOR CARDIOVASCULAR DISEASE IIN WHU E MALES I I I I I I Ag: 1945 I 1950 1955 I 1960 1965 1970 1975 = 1960 1965 I1990 [1995 30-34I 43 I 42 I 34 I 30 I 23 I 23 19 16 17 ‘ 16 ‘15 35-39I 66 I 66 I 62 I 75 T69 I 63 48 49 41 31 36 40-44 166 161 I 166 I 172 172 143 127 111 99 72 I 70 45-49 366 359 345 336 327 295 262 220 179 145 I133 50-54 673 631 619 603 599 515 462 365 335 256 I233 55-59 1100 1024 976 952 945 696 780 657 572 . 439 I366 60-64 1611 1602 1643 1534 1541 1466 1261 1113 946 747 I661 65-69 2534 2469 2577 2432 2365 2336 2066 1767 1505 1226 I1065 70-74 3631 3937 3647 3656 4056 3613 3164 2635 2529 2002 1775 75-79 I . I 5679 5093 4466I 4019 3266 2794 60-64 I f 6944 7994 71514: 6425 5332 4935 ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD. A_ge 1945- 50 1950-5195561 1960 6:1965fl7q1970-751975 6(I1960-6I1965 90 1990- 95 30-34 -1.0 -6.2 -3.6 -7.5 0.4 I -4.5 0.6 -15 0.9 0.5 35-39 0.3 -4.9 -6.4 -6.4 -5.6 I -14.8 1.0 -7.6 I -10.4 5.5 40-44 -7.2 5.1 -13.5 0.6 -28.4 -16.2 -16.1 -11.5 -27.7 -1.8 45-49I -9.1 I 43.9. -7.0 -11-02 -31.0 I -33.6 42.0 -40.3 -34.7 -12.0. 50-54 ‘ -41.4 I-11.6I -16.2 -4.6 I -83.5 ' -53.2 -77.1 -49.7 -79.4 I -22.5I 55-59 -75.9 I-46.4' -25.6 -7.4 -49.2 -115.7 -122.5 -65.6 -132.2 -53.2I 60-64 -9.1 40.6 -106.6I 7.1 -75.1 -165.6 -167.5 -167.5 -196.2 -66.5I 65-69 -65.0 107.6 -144.7 I -47.7 -46.7 -252.5 -316.6 I-262.0 -277.0 -163.2 70741 106.7 i-90.3 11.4 I197.5 -442.7 -449.2 -326.9I-305.9 -527.5 I-226.4 75-79I 0.0 0.0 JD ’ 0.0 5679.1 -586.5 -624.5 I-449.1 -752.6 -472.1 60-64 0.0 0.0 I 0.0 0.0 ‘8944.3 -950.6 -642.6I-725.9 -1092.7 0975 ~ I E I I I I I I I I I ! I I Y I ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945 50 1950- 1955- 6 1960 611965-701970 751975-601960-61965-90I1990-95 30-34 0.5 -3.9 -2.1 -5.0 0.3 -3.9 0.6 -1.7 -1.1 I -O.65 35-39: 0.1 I -1.1 I -1.6 I -1.7 -1.7 -4.7 0.4 -3.2 -5.0 ' 3.6 40-44 0.6 I 0.6 -1.5 0.1 -3.3 -2.3 -2.5 -2.1 . -5.6 0.5 45-49 0.5 ’-0.6 0.4 0.7 -1.9 -2.3 -3.2 -3.7 -3.9 -1.7 50-54 -1.2 0.4 I 0.5 I 0.2 -2.8 -2.1 -3.3 -2.6 -4.7 -1.6 I 55-59 -1.4 I 0.9 I 0.5 I 0.2 , -1.0 I -2.6 L -3.1 I -2.6 I -4.6 I -2.4 I 60-64 0.1 I 0.5 ’ 4.3;. 0.1; -1.0 I -2.5 ; -2.6;-3.OT; -4.2 I -2.3 I 65-69 0.5 0.9 I 4.17 0.4 . 0.4 I -2.2 I -3.1 I -3.0 I 37 I27 ; 70-74I 0.6 I 0.5I 0.1 I 1.0 I -2.2 I -2.5 -2.1 I -2.2 I -4.2 I-2.3I 75-79I I I . I I -2.1 -2.5 I -2.0 I -3.7 I -2.9 I 80-84 64I I : ' -2.1 I -2.1 1 -2.0 I -3.4 I -1.5 I 88 TABLE 7(b) 1 I CHANGE OF ITY RATE FOR MCD IN WHITE FE MALES MORTAL f MAJOR CARDIOVASCULAR DISEASE IIN WHITE FEMALE I I ‘ ' i ‘ i t A 1 - Age 1 1945 1950 19554 1960 ' 1965 1970 = 1975 1960 ' 1965j 1990§1995 30-34' 31 20 17 7 17 15 13 8 10 6 6 I 7 35-3 63 42 34 26 30 . 26 22 16 13 12 15 40-44 100 85 66 56 59 ’ 46 %42 35 34 23 26 45-49[ 191 156 122 106 106 93 ‘ 67 A 65 62 A 46 1 44 50-541‘ 363 262 I 221 201 199 176 155 I 124 121 I 103 I 69 55-59; 591 470 425 366 335 306 266 242 214 I 1927 154 60-64’ 979 641 603 700 614 561 501 446 417 I 323j 305 65-69 1609 1465 1440 1293 1177 1069 690 799 732 599 3 523 70-74 3064 2779 2524 2332 2317 1971 1679 1436 1306 1069 [1020 75-79 3702 3174 2596 2435 194111616 60-64 E 4 6573 5561 4941 4466 3601 3501 L ' E I I ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD 593 1945-501 950-551 955-6c1960-651 965-7019707511 975-60 1980-8511985-9011990-95 30-34 -11.3 -2.5 -0.2 -2.3 -1.6 -4.6 T 1.1 -1.4 ‘ -0.5 -0.7 1 35-39 -21.0 -7.2 -6.5 3.7 -1.6 -6.5 -5.1 -3.1 -1.5 2.9 40-44; -15.5 -16.7 -10.5 2.9 -10.2 -6.2 -7.0 -1.3 -11.1 3.1 45-49’ -35.2 -33.6 -16.4 0.5 -13.2 -5.5 -22.21 -2.9 143.9 -4.2 50-54 -61.2 -61.5 -20.0 -1.2 -21.6 -22.6 -30.6‘ -3.1 -16.6 -13.4 55-59; -120.9 -44.9 -59.6 -30.4 -29.1 -38.4 -26.0 -26.0 -22.1 -37.4 60641 -137.6 -37.6 -103.2 -66.5 -32.6 -80.6 -54.6 -26.4 -94.5 -17.4 65-69 -323.7 -44.6 3147.4 -115.6 -107.7 -179.2 -90.6 -67.1 1433.1 -75.9 70-74 3046 2555-1914 -14.9 -346.3 -291.7 -243.5 -130.2 -237.0 -48.4 75-79 -527.6 -576.6 -163.0 -493.3 -123.1 60-64 2. -1012.3 -619.7 -453.0 1887.0 -100.1 {I I I I ' T L I I I . . I 1 I I ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Ag 1945-5 1950-51955-611960-631965-7 1970-75 1975-80T1980-85L1 965-91} 990-95 3034 -7.3 -2.5 -0.3 -2.7 -2.4 -7.0 2.7 -2.9 I -1.2 -1.7 I 35-39 -6.7 -3.5 -4.9 2.9 -1.1 -4.6 -4.7 -3.8 E -2.2 4.9 ' 40-4 -3.1 -4.4 -3.2 1.1 -3.5 -2.6 -3.3 -0.7 L-6.6 2.6 45-49 -3.7 -4.3 -2.7 0.1 -2.5 -1.2 -5.1 -0.9 i -4.5 -1.6 50-54 -4.5 -4.4 -1.8 -0.1 .L -2.2 -2.6 -4.0 -0.5 ’ -3.1 i -2.61 55-59. -4.1 I -1.9 I -2.6 -1.7 i -1.7 ; -2.5 i -1.9 -2.3 -2.1 I -3.9 * 60644 -2.6 1 -0.9] -2.6 -2.5 7. -1.1 I -2.6 ' -2.2 -1.3 -4.5 I -1.1 65-69 -3.6 § -0.6 ’ -2.0T -1.6T -1.6 I -3.4 -2.0 I -1.7 -3.6 j -2.5 70-74‘ -2.0 I -1.6 I -1.5 -0.1 T -3.0 -3.0 -2.9 -1.6 -3.6 I -0.9 75-79 ’ L A -2.9 1 -3.6 -1.3 I -4.1 s -1.3 I 60-64 I T -3.1 I -2.2 -1.6 . -4.0 I -0.6 f 89 TABLE 7(c) I V ICHANGE OF MORTALITY RATE FOR MCD IN BLACK‘MALESI I MAJOR CARDIOVASCULAR DISEASETIN BLACK MALES I I. ; J t I I I Age 1945 I 1950I 1955I1960I 1965 1970 I 1975 I 1960; 1965 I 1990 I1995 30-34 71 I 95 ' 63 , 71 63 77 ; 43 I 55 I 76 I 32 I54 35-39 176 I 157 120 135 127 . 149 I 106 I 106 122 I 114 '104 40-44 366 316 . 296 266 266 310 226 I 252 227 I 164 221 45-49 604 540 514 434 420 476 4 372 I 446 356 I 370 392 50-54 1079 901 . 636 . 762 766 672 ‘ 636 I 639 621 I 595 I563 55-59 1620 1364I 1235 i 1063 1063 1035 696 I 926 . 674 I 615 I696 60-64 2462 2064 1643 I 1796 1975 1651 I 1435 I 1396 1402 I 1256 11305 65-69 - 4474 . 2659 2617 2716 2699 2463 1910 I 1671 1916 1666 I1744 70-74 I 5116 I 4064 3341 3426 3614 3465 3039 3 2751 2964 2666 2427 75-79 4666 3769 : 4262 4092+ 3654 3344 60-64 . , 6290 5741 6365 6237 ' 5257 $4923 I I I T T 1 ‘ 1 I I I. I I I ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945-50 1950-5f1955-6I 1960-619654 1970-75 1975-60 1980-8: 1965-9 1990-95 30-34 24.0 -32.2 6.2 -6.4 14.5 -34.5 11.6 23.5 -45.9 22.0 35-39 -21.4 . -36.6 14.7 -7.4 21.2 -42.1 1.5 13.9 -7.5 -10.6 I 40-44 -49.3 -20.6 -31.2 -0.7 44.2 -64.2 26.3 -25.0 «43.3 37.0 I 45-49 -63.8 -26.0 -80.1 -13.5 55.4 -104.3 76.7 -92.4 14.1 21.6 50-54 -177.9 -65.4 -53.9 4.5 -113.9 -36.4 2.7 -17.6 -25.4 -13.0 55-59 -256.2 -126.6 -151.6 -0.3 -46.0 -136.7 31.2 .-53.4 -59.6 63.4 60-64 -397.4 -221.7 -44.9 177.3 -324.4 -215.7I -37.1 I 4.1 -145.9 46.7 I 65-69 -1614.6 156.2 -99.2 -19.2 -215.6 -572.9‘ -36.7 I 46.6 -232.1 56.4 I 70-74 -1052.1 -723.2, 65.4 167.7 -146.9 -426.1 -267.7 212.7 -276.1 2565 75-79 ' -1117.6. 513.2 -190.3 -437.4 -310.1 60-64 I I 5469 I 644.0 -147.6I -960.1 -334.0 I I 1 L i . I I ; I ~ I . 'ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945-50 1950-5g 1955-61 1960-6I 196517197075 1975-6g 960-6I 1 965-91I1‘990-95- 30-34 6.6 -6.6 I 2.6 -2.4I 4.6 I -6.9 5.5 I 6.6 I-11.6I 13.7 35-39 -2.4 -4.7 I 2.5 ' -1.1 I 3.3f-57 0.3 I 2.6 I -1.2 I -1.9 40-44 -2.7 -1.3 I -2.1 I 0.0 I 3.3 -5.4 2.3 -2.0 I -3.6 4.0 45-49 -2.1 -1.0 I -3.1 I -0.6 2.6 -4.4 4.1 I -4.1 I 0.6 2 1.2 .. 50-54 -3.3 -1.5 -1.3 f 0.1 -2.9 -1.1 0.1 I -0.6 I -0.6 I -0.4 I 55-59 -3.2 -1.9 -2.5 I 0.0 I -0.9 -2.7 I 0.7 I -1.2 I -1.4 ‘ 2.0 I 60-64I -3.2 -2.1 -0.5 I 2.0 I -3.3 I -2.6 -0.5 0.1 I -2.1 0.6 I 65-69I -6.1 I 1.2 I -0.7 I -0.1 I -1.6 I 4.6% -0.4 0.5 I -2.4 0.7 : 7074! -4.1 - -3.6I 0.5 I 1.1 -0.6I -2.5 I -1.9 I 1.5 ITI -1.9 75-79I I f I I 4 -4.6 I 2.7 ‘ -0-91 -2.1 I -1.7 60-64I ~ I I -1.7 I 2.2 ' -0.5I -3.1 I -1.3 90 TABLE 7(d) I I I I I 4 I I I I ; ICHANGE OF MORTALITY RATE FOR MCD IN BLACK FEMALES IMAJOR CARDIOVASCULAR DIISEASE IN BLACK FEMALES I I T I 1 I. -¥1I—‘ —< B I I . I Age 1945 I 1950 1955I1960I1965I 1970 1975I1960 1965! 1990 1995 30-34. 113. 76 I 65 I 63 I 63 ; 49 . 29 37 33 31 I 32 35-39I 227 I 176 I 160 133 154I 110 I 59 67 59 I 59 I 56 40-44I 526 I 395 L313 222 245I 160 I 113 . 130 66 I 66 I113 45-49 I 729 I 641 I 456 422 314 I 279 I203 I 216 I 179 I 166 j 172 50-54 I 1145 I 1041 I 662 633 535 491 I 361 326 I 326 253 324 55-59 f1329 I 1343 ' 1196 990 653 679 527 . 512 516 462 466 60-64i 2252 I 2224 1596 ! 1607I 1479 1291 764 616 776 723 726 65-69 2033 I 2040 2156 '2040' 1665 1697 1254 1271 1167 1104 1066 70-74 3652 I 3056 3065 3332 2694I 2730 2126 2033 1911 1710 1619 75-79 I I 3612 3116 . 3160 3047 2661 2350 60-64 = I I § 7: 5566 I4399I 5103 5113 4325 3692 i if L ' I I I I fl I I I I I I ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945-50119505111 955-6CI1 960-6I 1965-7I 1970-75 1 975-6 1960-6 1985-9' 1 990-95 30-34 -36.8I 9.5 -1.9I-19.9I -14.3 -20.0 6.1 -4.6 -1.6 1.0 4L 35-39 -50.4 I -16.5 -26.5 I 20.9 I -44.0 -51.3 i 7.6 -6.0 0.7 -1.7 I 40-44 -131.4I -61.5 -91.2 23.1I-65.3 -67.0 17.6 -42.4I -1.81 27.1 A; 45-49 -66.2 I-164.7 -34.6 -107.4 -35.3 -75.8 12.9 -37.6I 7.0 -13.6 fl 50-54 -104.1 -178.7 -229.4 -98.2 -44.1 -109.1 ~55.6 2.7 -75.5 71.0. 55-59 14.6 -147.3 -205.9 -137.3 -174.4 -151.1 -15.7I 4.4 -34.3 -16.2I 60-64 -28.9 -625.5 8.5 -127.1I-188.3 -526.8 52.0 -40.3 -52.8 42 65-69 7.4 117.8 -117.6 -155.1r-188.6 -442.6 17.2 -104.4 -62.5 -36.1 70-74 -596.3 29.3 246.6 -637.9 36.1 -603.6 -92.6 -122.6;-200.7 -91.1 75-79 . I . -693.8 I 42.3 -112.9I-366.1 -311.4 6064 I I w ‘ I-1167.1I 703.7 10.4II-766.6I -432.3I - - 4. I I I ‘ I I I I; I I I = I I I I I ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOI: Age 1945-50 1950-551955-6 1960-6I1965-7I1970-75I1975-6 1960-619659199095 fir 5.6 -2.5 -1.0: 0.6 . 2.6 -2.4 0.2 -0.6 I 3.1 -6.5 -0.4. 6.3 I 30-34 -6.5 2.5 0.4 -4.8 -4.5 i -8.1 35-39 -4.4 -1.9 -3.3 3.1 -5.7 I -9.3 40-44I -5.0 -4.1 -5.6 2.1 I -5.3 I -7.5 45-49I -2.4 -5.6 I -1.5 II-5.1 i -2.2 -5.4 I 1.3 -3.5 0.6 I -1.5 I 1 4 50-54; -1.6 ; -3.4 I -5.3 : 3.17 -1.6 -4.4 I -2.9 0.2 -4.6I 5.6 I I I I . I I 60-64 I -0.3 I -5.6 ~ 0.1 -1.6 ; -2.5I -8.2 I14 -1.0 -1.4 I I . I I 55-59II 0.2 I -2.2 I -3.4 -2.6 I-4.1 * -4.5 -0.6 0.2 -1.3I -0.7 I . . , . I 0-1 65-69I 0.1 .I 1.2 I -1.1 -1.5I -2.0 -5.2 0.3 I -1.6 -1.1 1 07? 70-74I -3.3 I 0.2 I 1.6 -3.6I 0.3 . -4.4 -0.9 -1.2 -2.1 I -1.1 4 75-79I I I I II -3.6 0.3 -0.7 -2.5 I 2.3—I 60-64I ' I -4.2 3.2 I 0.0 I -3.1_I -2.0 I 91 TABLE 7(a) 1 I 1 1 CHANGE OF MORTALITY RATE FOR DOH IN WHITE MALES 41 1DISEASE OF HEART IN WHITE MALES I -__.4 1 1 £991 19751 ._W .. 4 71995 11945 1 1950 1 1955 1960 1 1965 1 1970 1980 1 19851 1990 30-34‘ 30 1 33 25 1 22 17 j 17 15 1 14 1 13 1 13 1 11 35-39 62 1 73 59 T 65 57 1 52 1 39 1 43 1 36 1 26 1 27 4044 139 158 164 I 151 147E127 1 114 98 1 90 1 63 52 45-49 285 316 300 1 297 290 262 236 1 199 166 1 130 -103 50-54 513 539 540 1 530 528 1 453 415 1 350 1 301 1 223 1184 55-59 845 876 835 1 815 1 817 781 678 ‘ 590 ' 513 391 1302 60-64 1206 1 1305 1 1362 1273 1300 1242 1089 981 836 659 1502 65-69 1807 19581 2036 1932 1920 1916 1735 1520 1287 1046 802 70-74 2622 2889 2862 2878 3046 2824 2487 2319 2079 1651 1303 75-79 1 4261 3815 3520 3240 2632 1978 80-84 1 1 6344 5749 5326 5026 4180 3429 1 ‘1 1 1 1 1 * ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD _A_g§ 1945-50 1950-55795560 1960-65 1965-701970-7511975-81 1980-8.1985-9 1990-95 30341 2.6 -8.0 1 -2.6 -5.0 0.1 -2.1 L11 -0.7 -0.7 -1.91 35391 10.7 -4.2 1 -4.0 -8.0 -4.6 -13.5’ 4.7 -7.6 -9.5 0.4 1 40-44' 18.7 5.5 -12.8 -3.7 -20.4 -12.3 -16.8 -7.3 -27.2 -11.4 45-49 31.0 -15.7 -3.4 -6.8 -28.5 -25.5 -36.7 -33.8 -35.3 -26.8 50-54 21.2 1.1 -10.2 -2.3 -74.8 -38.0 -65.3 -48.21-73.2 -44.7 55-59 31.1 -40.6 -20.8 2.9 -35.9 -103.0 -88.0 -77.3 ’-122.1 -89.3 60-64 98.9 56.6 1 -89.0 1 26.6 -57.1 -153.7 -107.6 -145.5 -177.1 -157.0 6569 151.3 77.6 ' 403.81 -12.0 1 -3.4 -181.5 21491-2333 2412 .2431 70-74 267.1 -27.5 16.8 16731-2221 -336.4 -168.21-240.2 -427.3 -348.9 7579 1 1 -446.3 -294.7 -280.7 -607.2 -654.2 8084 1 1 ‘-594.7 -422.8 -300.1 -846.2 351.21 1 . 1 , - 1 - I . 1 1 ‘ 1 L 1 i 1 I 1 1 I J - L '. ‘ 1 . ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD I Age 1945-551950-551955-6011960-651965-701970-7 1975-811980-851985-911990-95 30-34 1.7 -4.9 1 -2.1 -4.5 0.1 -2.5 -1.5 -1.0 1 -1.0 -3.1 35-39 3.4 1 -1.2 -1.2 1 -2.5 -1.6 -5.2 2.4 -3.5 1 -5.3 0.3 , 4044* 2.7 1 0.7 -1.6 1 -0.5 -2.8 -1.9 -2.9 -1.5 g -6.0 -3.6 : 45-49‘ 2.2 -1.0 -0.2 1 -0.5 -2.0 -1.9 -3.1 -3.4 1 -4.3 -4.1 ‘ 50-54 0.8 . 0.0 -0.4 -0.1 -2.8 -1.7 -3.1 -2.8 1 -4.9 1 -3.9 1 55591 0.7 1 -0.9 -0.5 1 0.1 f -0.9 -2.6 -2.6 -2.6 1 -4.8’ -4.6 ; 60641 1.6 1 0.9 . -1.3 ‘ 0.4 I -0.9 -2.5 - -2.0 -3.0 1 -4.2 -4.8 ' 65-691 1.7 1 0.8 1 -1.0 -0.1 = 0.0 4.941 -2.5 -3.1 1 -3]. -4.7 70-74? 2.0 1 -0.2 1 0.1 1.2 -1.5 1 -2.4 ‘ -1.4 -2.1 1 -4.1 1 -4.2 ' 75-797 1 ‘ ; 1 -2.1 -1.5 -1.6 1 -3.7: -5.0 1 80-841 1 1 1 1 -1.9 1 -1.5 -1.1 1 -3.41 -3.6 1 92 TABLE 7(f) I I I ICHANGE OF MORTALITY RATE FOR DOH IN WHITE FEMALES I DISEASE OF HEART IN WHITE FEMALE I I I I I I I 1 I Age I 1945I 1950 1955 . 1960I 1965 I 1970 I1975 1960 I 1965I 1990 '1995 30-34 21 ; 13 I 11 I 10 7 6 4 7 5 I 5 7 4 35-39 42 I 29 I 24 I 16 16 16 I 13 11 10 I 6 9 40-44I 65 I 56 ' 43 ' 34 I 35 31 27 24 25 I 14 15 45-49I 110 j 104 64 I 72 I 73 63 62 48 44 I 6I2 50-54I 207 I 169 - 152 140 142 125 I 116 96 97 I 61 T 56 55-59 350 323 I 305 269 251 I 242 ‘ 210 197 177 152 I 106 60-64 567 590 I 563 520 467 I 451 393 365 340 266 I216 65-69I 1072 1034 ' 1027 . 921 679 I 617 675 641 599 467 365 70-74I 1655 1654 1733 1621 1647 ' 1465 1243 1115 1053 636 I 669 75-79‘ I 2611 2252 1964 1917 1497 - 1233 60-64 I 4534 3636 3566 3399I 2726 2309 I I I T . I i I L I I I ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945-501950-551955-601960-651965-701970-751975-60 1980- 6 1965- -9I1990-95 30-34 -7.8 -2.1 0.4 -3.5 0.7 -3.6 2.6 -1.7 0.1 I -1.5 I 35-39 -12.7 I -5.4 -7.7 1.3 -1.5 I -2.7 -2.7 -1.0 -1.7‘ 0.7 40-44 -9.2 I -13.1 -9.3 1.5 -4.3 I -4.4 -2.3 1.2 -11.2 1.2 I 45-49 -5.3 -20.6 -11.4 0.5 -9.4 -1.0 -14.0 43 -6.0 -9.1 I 50-54 -16.6 -36.7 -12.4 2.4 -16.7 -9.1 -20.1 0.4 -16.1 -22.5I 55-59 -27.1 -16.3 -35.9 -16.0 -9.2 -31.7 -13.0 -20.1 -24.6 -46.6I 60-64 3.1 -7.4 -63.0 -52.1 -16.7 -57.8 -27.5 -25.7 -71.4 -52.1I 65-69 -38.3 -6.7 -105.4 -42.1 -62.6 -141.8 -33.5 -42.6 -112.1-121.8I 70-74 -0.5 -121.5 -111.6I 26.3 -182.5 -221.7 -126.2 -61.9 -216.9 4466 75-79: I - -358.7 -268.2L66.4 -420.1 264.11 80-84I I I I I 6955 -252.0 ‘-167.0 -672.7 -417.4 I : I ' I I I I f T I I I I I I I IANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD 1970- 751975 60 _ge I1945-501950- 55 1955 6019606519657 1960 61196591990 95 30 34I -7.5 I -3.2 -0.7 -6.6 2.0 -9.4 13.6 -4.9 0.5 5.9; 35-39; -6.0 I -3.7 - -6.4 1.6 -1.7 -3.4 -4.1 -1.9 -3.5I 1.6 I 40-44} -2.8 E -4.7 i -4.3 I 0.9 -2.4 . -2.9 I -1.7 I 1.0 -8.8 1.7 ‘ 45-49I-1.0 -4.0 I -2.7I 0.1 -2.6 L03 I -4.5 ’ -1.8 -3.7 -5-0I 50-54] -1.8 I -3.9 I -1.6 I 0.3 I -2.4 I -1.4 I -3.5 0.1 -3.3 -5.6I 55-59I -1.5 I -1.1 I -2.4 I -1.3 I -0.7 I -2.6 I -1.2 . -2.0 . -2.6 . -6.1 I 60-64I 0.1 I -0.3 I -2.2 I -2.0 T -0.7 I -2.6 I -1.4 ’ -1.4 I -4.2 -3.9] 65-69I -0.7 I -0.1 I -2.1 I -0.9 -1.4 I -3.5I -1.0 -1.3 I -3.7 I -5.0 I 70747 0.0 I -1.3 I -1.3 I 0.3 I -2.2 I -3.0 = -2.1 . -1.1:I 411-3.5 75-79 7 i I -2.7 I -2.4 I -0.7 I -4.4 -3.5I 60-64I I I -3.1 -1.3 I -1.0I-4.0* -3.1 I 93 TABLE 7(9) l 1CHANGE OF MORTALITY RATE FOR DOH IN BLACK MALES1 DISEASE OF HEART IN BLACK MALES 1 1 l r 1 Age + 1945 1 19501 1 1955 19601 1965 1970T1975 1960 11985f19901 1995 30-341 49 I 62 1 44 1 44 36 57 25 40 1 61 264 48 353911321 95 1 65 1 65 1 66 127 71 1 66 j 964 68 I 66 40-441 253 1 252 208 1 168 169 217 165 f 196 178 1 139 173 45-491 427 1 401 379 1 330 315 369 1 261 1 380 292 307 313 50541 777 1 639 634 579 587 543 510 514 527 470 477 55-59 11521 1003 876 782 824 811 726 779 753 705 752 60-64 1176 1492 1400 132111419 1316 1156 1120 1170 1057 1107 65-69 1505 1680 1957 1809 1919 1676 1471 1479 11602 1429 1443 70-74 3195 2720 23162266 2567 2642 2311 2152 2406 222011961 7579 1 1 1 1 3582 2899 3323 3260 3046 2673 60-64 1 ‘ 1 1 4571 4120 4666 5119 4386 3726 1 i L 1 L 1 1 1 1 1 J 1 1 1ABSCLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945-51195051195561960-61965-711970-7 1975-80 1980-85 1965-911990-95 30-34 12.7 1 -18.6 1 0.6 -8.9 21.1 1 -31.4 15.1 20.2 -34.4 ' 21.5 35-39 -37.5 -10.11 0.4 3.5 38.7 -56.6 15.0 10.4 -7.8 -1.9 40-44 -0.4 1 -44.6 -39.9 21.6 27.1 -51.8 31.3 -17.7 -39.4 34.1 45-491-262 217490 -15.3 54.2 -67.7 98.8 -86.5 14.8 6.1 50-541-1380 -5.1 154.9 8.4 -44.0 -32.9 3.8 12.6 -56.5 6.7 55-59 -148.3 -1251-96.3 42.3 -13.5 -84.8 52.6 -26.3 -47.0 46.7 60-64 313.9 -91.91-78.9 96.1 -102.6 -160.0 -35.9 50.1 413.9 50.0 65-69 374.3 77.1 1-146.1 110.5 -43.6 -405.1 8.1 123.6 -173.6 14.4 70-74 -475.3 -401.91-51.31300.6 74.9 -331.0 -158.9 253.7 -166.3 -236.3 7579 . 1 1 -683.2 423.7 -62.8 -213.3 -373.5 8084 1 1 -450.8 747.5 251.2 -732.7 -660.3 1 5 4; 4 1 1 1 1 1 1 1 1 1 1 1 1 1 ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIC Age 1945-511950551195561960-61965 7 1970-71 1975- 80 1960 85 198591199095 30341 5.1 1 -6.01 0.4 -4.0 11.9 -11.1 12.0 10.0 -11.14 16.4 35 39 -5.7 ' -2.1 ' 0.1 0.6 8.8 -8.9 4.2 2.4 ~ -1.6 -0.4 40-44 0.0 -3.5 -3.8 2.6 2.9 -4.6 3.8 -1.8 44 4.9 45-49 -1.2 -1.1 -2.6 -0.9 3.4 -4.8 7.0 -4.7 1.0 0.4 50-54 -3.6 -0.2 1 -1.71 0.3 -1.5 -1.2 1 0.1 0.5 1 -2.1 0.3 55-59 -2.6 -2.5 1 -2.2 1 1.1 -0.3 -2.1 1 1.5 -0.7 1 -1.2 1 1.3 60-64 5.3 -1.2 1 -1.1 1 1.5 -1.4 -2.4 1 -O.6 0.9 1 -1.9 1 0.9 65-69 5.0 0.6 -1.5 '1 1.2 -0.5 -4.3 1 0.1 1.7 1 -2.2 ' 0.2 70741 -3.0 1 -3.0 1 -0.4 i 2.7 1 0.6 1 2.51 -1.4 1 2.4 1 -1.5 -2.1 75791 1 F ~ 1 -3.6 1 2.9 ' -0.4 1 -1.3 1 -2.5 80-641 1 -2.0 3.6 1 1.0 1 -2.9 1 -3.0 94 TABLE 7(h) lCHANGE OF MORTALITY RATE FOR DOH IN BLACK FEMALES 4 1 V t l 1D|SEASE OF HEART INELACK FEMALES 1 ' TWT Age 1 1945119501 1955 1 1960 1 1 i 1 196511970 W975 19801 1985 1 1990 11995 1 134123118121124 20122 @341 71 1 39 =1 56 41 i 35-391 138 120 94 76 . 87 ' 59 1 37 1 49 1 36 35 1 38 40-441 301 1 259 208 129 148 109 L70 83 1 67 53% 78 45-49 470 1 422 278 266 189 170 ’ 132 154 135 1 146 1124 50-54 700 ‘ 698 . 55-59 809 878 874 647 ‘5601 493 397 3861 404 378i 361 >————*>-—«a1>—.-—1>.——1-— 4 1 1 582 T 406 323 L353 266 250 246 1 206 1227 1 1 60-64 1470 1644'109511112 10351 937 605 649 621 601 582 % 65-69 1313 1344 1392 111—1292 1190 1 1198 920 947 932 908 839 70-7i 2175 1899 2087 1 2265 1785 1867 1552 1499 1460 1359 1221 75-79 I 1 .2595 2304 2360 2376 2079 1884 80-84 ‘ 1 1 13668 2914 3768 4105 3405 2936 1 1 . 1 1 1 1 1 ; I 1 1 1 1 ‘ 1 ‘ iABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 11945-50 1950-51195560196065 1965-7 1970-751197580 1980-81 1985-90 1990-95 30-341-319 17.0 -14.8 -7.2 -11.1 -5.1 2.7 2.9 -3.4 2.4 35-391-18.3 -25.6 -18.61 11.7 -28.2 -21.9 11.7 -13.3 -0.5 2.6 40-441-41.9 -50.7 -79.71 19.1 -38.6 -38.9 13.0 ~16.2 -13.9 24.6 45-49 -48.6 -143.3 -12.4 -76.9 -19.3 ~37.6 21.8 -18.6 11.0 -22.2 50-54 -1.9 -116.8 -175.21-83.8 30.6 -87.0 -16.0 -4.4 -39.5 20.8 55-59 68.4 -3.3 -227.41-86.8 -66.5 ~96.7 -10.5 17.31-25.9 -17.0 60-64 174.2 -549.2 17.0 -76.7 ~98.3 -332.3 44.8 -28.6Y -20.1 -18.3 65-69 30.4 47.9 I -99.2 -101.91 7.7 1 -278.0 26.9 -15.1 -23.8 -69.2 1 70-74 -275.71187.4 178.8 -480.3 82.0 -315.4 -52.5 -39.1 -101.6 -138.0 80-84 -753.3 853.8 336.4 -699.5 -469.2 75-79 1 1 -291.0 55.1 16.4 -296.4 -195.5 I I *f 1 ‘ 1 1 T 1 1 1 1 1 . ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945-501950-5111955601960-6511965-71970-761975-8 1980-811985-9 T1990-95 A 30-34 -9.0 8.7 -5.3 -3.5 1—6.5 -4.5 3.0 1 2.8 1 -2.9 2.3 35-39 -2.6 . -4.3 -3.9 3.1 1 -6.4, -7.4 ' 6.2 -5.4 1 -0.3 1.5 40441 -2.8 -3.9 -7.7 1 3.0 -5.21 -7.1 3.7 ’ -3.9 -4.2 1 9.3 r Y 1 | 1 T -1.2 -0.4‘ -3.2 2.0 i 1 1 1 1 1 45-491 21 -6.8 1 -0.9 1 -5.8 -2.01 -4.4 1 3.3 1 -2.4 1.6 -3.0 ' 50-54: -0.1 -3.3 -6.0 -4.1 1 1.9 -4.91 1 1 55-59 1.7 1 -0.1 -5.2 #27 -2.4 -3.9 1 -0.5 1 0.9 -1.3 -0.9 1 % 60-64 2.4 -6.7 0.3 s -1.4 1 -1.9 1 -7.1 ’ 1.51 -0.9 . -O.6 1 -0.6 65-69 0.5 1 0.7 1 -1.4 1 ~1.6 1 0.1 1 -4.6 0.6 1 -0.3 1 -0.5 1 -1.5 70-741 25 1 2.0 1 1.7 1 -42 1 0.91 -3.4 -0.7 1 -0.51 -1.4 ' -2.0 75-791 1 ' 1 1 1 -2.2 0.5 1 0.1 1 -2.5 -1.9 1 80-841 1 7 ‘ 7 1 -4.1 . 5.9 1 1.8 1 -3.4 -2.8 L 95 TABLE 7(1) ICHANGE OF MORTALITY RATE FOR IHD IN WHITE MALES ISCHEMIC HEART DISEASE IN WHITE MALES . I A I Age I 1945I 1950I 1955 I 1950 I 1955I 1970 1975 1980I1 95 5I 1990 1995 3034I 12 I 14 12 I 11 I12 I 10 I 10 7 I 5 I 5 5 35-39I 31 I 37 I 41 I 40 I41I 35 I 25 . 30 23I 15 I18 40-44I 91 897 105 I 99 109I 88 80 I 75 4 51 I 41 I42 45-49 170 188I 195T: 198 218] 184 171 151 I 120 I 92 I 99 50-54 301 . 325 357 ,- 352 398 I 328 . 301 I 274 I 225 I 155 I154 55-59 502 I 535 I 550 555 I 514 I 585 490 I 452 I 388 293 240 50-54 707 I 770 878 ; 854 I 977 I 901 . 781 I 784 I 838 504 353 55-59 935 ' 1101 1245 I 1297 I 1432 1397 ‘ 1250 1225 ' 989 795 548 70-74 1350 1528 1582 1888 2257 2054 1793 1855 1522' 1284 1170 7579 I . ‘ 3117 2752 2814 2495 2028 1790 80-84 I I I 4591 4137 4193 3788 3184 3285 I I I 2 f I I I I I » I I I I I I I I I I ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD I __A_ge I1945-5q1950 581955 501950 551955 7 1970751975 801980- 881985901990 95 30-34 2.2 I -2.2 -1.0 I 0.9 I -2.0 -0.1 -2.4 -1.5 -0.2 -1.1 I 35-39 5.7 I 4.3 -1.4 I 1.0 -5.5 ' -10.3 4.9 I -5.7 -8.3 3.3 I 40-44' -1.5 15.4 . -5.5 9.7 ’ -20.3 -8.6 -4.5 I-14.1 ' -19.7 0.5 45-49 18.1 5.8 3.5 19.3 -33.8 . -13.3 -19.1 -31.7 -27.7 5.5 50-54 23.5 32.41 4.8 35.9 -59.9 -25.7 -27.5 -47.3 -50.0I —2.8 55-59 32.9 15.2 4.7 I 58.5 -48.3 -74.8 -28.5 -74.1 -95.3 -52.1 50-54 53.0 108.4 -14.4 112.8 -75.1 -119.5I 2.4 -145.1 -133.5 -150.8I 55-59 155.0 144.5 51.3 134.3 -34.3 -135.9 -35.0 ‘-235.1 -193.0 -147.9I 70-74 177.7 154.4 205.51 358.4 -192.5I -271.3 52.4 -233.1 -338.5 -114.01 75-79 I I I I -355.2 51.7 -319.01 -455.9I-237.5 80-84 T I . -453.1 55.7 -405.5I -504.0I 101.9 I I I i I . I I i I I I * I I I I . PERCENT CHANGE OF MORTALITY CHANGE DURING SUCCESSIVE PERIOD I Age 1945-501950-551955-581950-55I1955-781970-751975-801980 881985- 901990-95 30-34 3.5 I -3.1 -1.7 I 1.7 I -3.4 I -0.3 -4.9 -4.1 I -0.5 -4.0 I 35-39r 3.7 I 2.3 I -0.7 I 0.5 I -2.7 I -5.9 4.0 I -4.5 I -7.2 I 4.4 I 40-44 -0.4 I 3.5 I -1.1 I 2.0 I -3.7 I -2.0 I -1.1 I -3.8 1 -5.5 I 0.3 I 45-49 2.1 I 0.7 . 0.4 I 1.9 L-3-1 I -1.5 I -2.2 I -4.2 -4.5 1.4 I 50-54 1.5 I 2.0 0.3 . 2.0 I -3.5 . -1.5 - -1.8 I -3.5 -5.3 w -0.3 I 55-59. 1.3 I 0.5 0.2 I 2.1 I -1.5 I -2.5 I -1.2 I -3-2: -4.9 - -3.5 I 50-54I 1.8 2.8 I -0.3 I 2.5 s -1.5 I -2.7 I 0.17 -3.7 ; -4.2T-5.0 I 55-59I 3.5 2.5 - 0.8 I 2.1 I -0.5I -2.0 I -0.5 I -3.9 I -3.9 I -3.7 I 70-74; 2.5 i 2.0 2.5 . 3.9 -1.7 I -2.5 ’ 0.7 i. 2.5; -4.2 , -1.8 I 75-79I : I L ‘ -2.3 0.4 -2.3 -3.7 I -2.3 I 80-84I ’ I T -2.0 0.3 -1.9 I -3.2fl 0.5 ; 96 TABLE 70) ‘CHANGE OF MORTALITY RATE FOR IHD IN WHITE FEMALES I I ISCHEMIC HEART DISEASE IN WHITE FEII/IALES ‘ ‘ I I T Age I 1945 I 1950 I 1955 F1960I 1965I197oz197sI 1960 L1985 I199011995 30-34I 5 I 2 I 2 2 I 4 I 3 I 2 g 2 I 1 I 1 I 1 35-39I12I 7 I 6 5I10 7I7I 4 T 4 I3j5 40-44I 23 I 16 17 13 ' 23 16 I 15 I 14 15 I 6 I 12 45-49I 42 36 33 34 49 L37 f 37 32 . 26 22 I 22 50-54? 99 76 72 70 100 I 76 I 73 66 I 64 51 I 47 55-59 194 143 146 143 179 ‘ 156 I 138' 137 125 . 101 ' 60 60-64 312 271 317 302. 335 296 267 274 I 236 I 165 142 65-69 505 467 546 554 ‘ 641 569 I 467 466 ’ 436 ' 346 . 261 70-74 657 623 696 936 1167 1046 673 I 654 779 611 559 75-79 , . . I 1657 1605 1534 1411 1069 1046 60-64 I I f - I 3247: 2721 2749 2472 2003 1930 I . I I ' ‘ i I I . i T L I I l 1 I i ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD I Age 1945-50I1950-55 1955-60 1960-E 1965-71‘1970-7 1975-3196065 1965-90 1990-95 30-34 -2.4 I 0.0 -0.4 2.1 -1.1 -1.3 I 0.7 -1.1 0.0 0.1 T 3539 -4.6 T -1.1 -1.4 5.6 -3.5 - 0.0 -2.6I 0.2 - -1.4 I 1.6 40-44 -6.6 0.6 -4.0 9.6 -6.3 ‘ -1.0 -1.1 1.0 I -7.2 {4.2 45-49 -5.7 -3.1 1.2 14.7 -11.7 -0.2 -5.4 -4.1 T -5.4 -0.3 50-54 -23.1 -4.1 -2.2 29.6 -21.8 -4.9 -7.3 -1.5 -13.0 -4.3 55-59 -51-3- 5.0 -4.1 35.7 -22.7 -18.5 -1.4 -12.0 -23.7 -21.4 60-64 -40.1 ’ 45.6 -15.5 33.6 -39.4 29.2: 7.4 . -36.2 -53.5 -42.2 65-69 -36.5 76.6 6.3 86.9I -72.2 -1020 21.0I -51.6 -66.3 665 70-74I -34.7 . 73.1 42.1 I 24921-1411 -173.1 -18.2 -75.7 1 -167.9 -51.7I 75-79I‘ I - I -251.4I -71.5 -122.4' -322.0 41.3 80-84 I i I 1 I .5254]: 27.2 -276.7 -469.3 I-72.1 T 7 j ' I . I : I I ; I i i i . : = I 1 I I I I I I I i I ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOE Age 1945-50I1950-55 I1955-60 1960-e 1965-7I1970-7I 1975-6 1960-65 1965-90 1990-95 30-34 -10.4 -0.3 I -3.3 423.4 -5.6 I -6.9 6.6 -10.0 0.3 2.5 35-39 -7.8 -3.2 T -4.6 I240 -6.8 I 0.1 -7.7 0.6 -6.5 I11.1 40-44 -5.9 I 0.7 -4.7 I146 55 I -1.2 . -1.4 I 1.4 -9.4 T104 4549 -2.7 I -1.7 0.6 I 6.5 I -4.6 I -0.1I-2.9I -2.6 I -3.9 -0-31 50-54I -4.7 I -1.1 -0.6 i 6.6 -4.4 -1.3 I -2.0 I -0.5 -4.1 . -1.7j 5559? -5.3 I 0.7 -0.6 I 5.0 -2.5 I -2.4 I -0.2 T -1.8 -3.6 I 4.2; 60-64I -2.6 3.4 -1.0 I 2.2 -247 -2.0 I 0.6 I -2.6 2 -4.5 I-4.6I _ 65-69I -1.5 T 3.4 I 0.3 I 3.1 I -2.3 I -3.6j 0.9 1 -2.1 : -4.1 ;-5.0‘ 70-74 I -0.8 I 1.6 ‘ 0.9 I 5.3 I -2.4 i -3.3 f 04] -1.8 7 -4.3 -1.7i 75-79 f f T -2.7 T-0.9 I -1.6 : -4.6 ‘ 06? 80-84 - ‘ T . -3.2 I 0.2 -2.0 I -3.6 -0.7= 97 TABLE 7(k) [CHANGE OF MORTALITY RATE FOR ED IN BLACK MALES ‘ . i ' L . - ISCHEMIC HEARTY DISEASE DEATI-I RATES IN BLACK MALES Age a 194511950I 195?I19601 1965 I 1970 I 1975 I 1960? 1965 I 1990I1995 30-34I11I21‘1016I20'26’776'7’517 -— L 3539: 37 I26I 26 41 - I55 61 34I20I25423i27 40-44I 59 76I 67 63 I 134 134 105 76 . 52 40 I 50 1r-—7> 45-49I 137 147I 179 ' 174 214 . 235 194 166‘ 110 I 94 : 111 50-54T 263 269 290 338 41 8 365 351 216 193 177 I 220 A 55-59I 391 404 419 426 564 545 507 376 327 292 I 253 60-64 336 543I 666 776 995 910 808 I 564 542 442 341 65-69 426 676' 691 1047 1374‘ 1339 1042I 601 614 I 660 590 70-74I 946 I1077 1110 L1334I 1879 1863 16347l 1222 1302 I 1161 1067 7579’ I I I I 2596 2064I 2063 I 1976 I 1641 : 1463 T 80-84I I - . I 3336 I 2916 T 2627 . 2664 .5 2519 I 2319 I L I I ' . i I ' I I I L I 1 I ‘ - T ' r ' ' T I I, V . ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD I Age 1945- 50 1950-I 1955- 60 1960- 6‘1 1965- 7C 1970- 75 1975- 80I 1980- 8 1985- 9C 1990- 95 30-34 9. 6 4-11 .0 -2. 0 12.7 5.1 -18.3 -0.9 0.2 -1.9 2.3 I 35-39 -11.1 I 2.4 I 13.1 13.7 6.1 . -26.6 -13.9 4.6 -2.5 4.7 I 40-44 17.2 10.6 -4.1 50.6 -0.2 -29.1 -26.2 -26.4‘ -12-31 9.7 I 45-49 10.7 31.4 -5.2 40.6 20.4 -40.7 -26.1 -58.1 -15.6I 17.0 .T 50'54. 6.7 I20.1 48.6 80.2 -53.2 -14.1 -135.6 -22.9 -16.0 43.8I 55-59I 13.0 I 14.2 7.8 I157.4T -36.9 -38.1 -1306 -48.6 I -35.0 . -39.7 60-64I 206.6 123.1 110.1 219.0 -84.6 -102.5 -244.0 -22.1‘ -99.7I-100.5I 65691 246.1 215.5 155.4 327.0 -34.6 -297.1 I-240.6I13.1 -134.5I -90.4I 7074? 130.7 32.9 224.6 544.9 -16.0 -229.6 I-412.2I 60.9 -141.6I -93.7I 75-79I 0.0 0.0j 0.0 0.0 I2596.4 -534-5I —O.6 I-67.2 6350-1777? 60-64I 0.0 I 0.0I 0.0 I 0.0433356: -417.6I -91.2 I 37.3 3345.3 -2001? —Y Y *— vi L T I I I I I I I I I _'L ¢ I . I . I I I; I I I i I I t I I I I I . - L . I 4 TANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD J6 I1945 50I1950-I1955-6I11960 6i1965-7C1970-75 I1975-6CI1960-6II1965-9c1990-95 3034I 17.4 I.-107I -4.1 I 33.3. 5.0 -14.3 . -2.5 I 0.6 I -5.7 100: - . J E 35-39I -6.0 I 1.9; 9.3 I 6.6 I 2.2 -8.8 -8.1 I 4.7 I -2.0 4.2 I 40-44I 5.6 I2.6 -0.9 I 12.2 0.0 -4.4 -5.0 : -6.7 . -4.7 I 4.9 45-49' 1.6 I4.3I -0.6 I 4.7 1.9 I -3.5 -2.7 I -6.9 ‘ -2.9 I 3.6 . 50-54 0.5 I 1.5I 3.4 I 4.7 I -2.5 I -0.8 -7.7 I -2.1 -1.7 ; 5.0 I 55-59; 0.7 I 0.7I 0.4 7.4 I -1.3 I -1.4 , -5.2 I -2.6I -2.1 . -2.7 I 60-64I 12.3 I 4-5. 3.3 5.6 3 -1.7 I -2.3 -6.0 -0.6 I 37 I45 I 65-69I 11.6 I 6.4I 3.5 , 6.2 -0.5 I -4.4 -4.6 I 0.3 I -3.3 I -2.7 ‘ 70-74T 2.6 I0.6I 4.0 I 6.2 I -0.2 I -2.5 I -5.0 I 1.3 . -2.2 -1.6 I g . 6 I . 75-79I . I , I -4.1 I 0.0 I06 : -3.4 I -2.2 T4 60-64? I T I . -2.5 I -0.6 I 0.3 . -2.4 I -1.6 . 98 TABLE 7(1) 1 I. I I I I 0 L I I ICHANGE OF MORTALITY RATE FOR IHD IN BLACK FEMALES ISCHEMIC HEART DISEASE INTBLACK FEMALES ~ I I T I , Ag I19457 1950 I 1955419604 1965 I 1970 I 19751960 I 1965I 1990,1995 30-34L10L 9 I 12 I 9 I 16 I 9 ' 9 i 3 7 5T 2 4 35-39I 25 I’ 24 I 22 25 I 49 I 27 16 I 9 I 10 I 5 I 6 40-44 757 70 I 63 42 I 96 61 36 I 25 I 11 #15? 19 45-49 116 103 67 106 131 - 101 I 63 I 54 I 56 I 47 47 50-54 146 226 199 199 216 I 214 174 126 I 109 77 104 55-59 196 216 260 332 392 337 254 193 I 216 196 I 116 60-64 349 529 406 561 ‘ 690 639 397 325 320 309 I 194 65-69 295 426 496 651 646 , 611 609 535 520 496 I 393 70-74 . 414I 694 I 773 I 1196 1256 1341 1071- 916 777 763 I 646 75-79 ’ I ‘g I . 1656 . 1596’ 1501 1406 1226 I 1019 80-84 I I I 2649 T2073 2309 2491 I 2173 I 1966 I I ' E I I I ' I I I I I ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945-5195055195560 1960-6 1965-7 1970-75 19756196065 1965-91 1990.95 3034 -0.9I 2.7 I -3.2 9.0 -9.2 0.3 -5.5 1.5 -2.7 1.9 35-39 -1.9 -1.0 2.4 24.1 -21.8 -9.5 -8.5 I 0.7 .4.7 0.9 40-44 -5.6 -6.3 -21.9 56.0 -36.9 -22.7 -12.6I -14.4 3.6 4.2 45-49 -15.0 I -15.9 16.4 25.1 -29.5 I -16.5 -29.0I 2.6 -9.2 I -0.5 2 50-54 79.2I -26.9 , -0.1 I 17.3 -1.7 I -40.2 I-46.3I -16.2 -32.3I 27.2 . 55-59I 19.9 62.0 I 52.2 I 60.0 I -55.9 I -62.6 I-61.2I 23.4 I 49.67-77.71 60-64 179.6 -122.6 175.1 106.5 -50.9 I-241.3 -72.7' -4.6 1-11.1I-115.2I 65-69 132.7 70.0 153.6 196.6 -37.3 I-201.9 -74.2 -15.0 ’ -21.9 I-104.7I 70-74 2609 76.2 425.2 59.9 63.3 I-270.3 -153.2E-140.9 6.6 435.51 7579 0.0 ‘ 0.0 0.0 0.0 1657.9 -262.4 -94.4 -93.6 -160.0 2065 80-84 0.0 0.0 0.0 I 0.0 2649.2 -575.6 235.9 161.6 -317.7 -165.2 L I . I I A I. I L. I I i I I i I I I I I. I : L I I IANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age ' 1945-5195055 1955-60 1960-6 1965-7 1970-7519756 1960-651965-9I1990-95 30-34 -1.8 I 5.6 ' -5.4 - 20.6 -10.4 0.6 L425 9.0 I -11.0: 17.5 I 35-39 -1.5 I -0.9 I 2.2 ' 19.4 -6.9 +70 I -9.6 I 1.6 -9.4 3.3 I 40-44 -1.5 -1.6 -6.9 27.0 -7.6 -7.5 -6.6 I -11.3 7.0 5.7 I 45-49 -2.5 -3.1 4.2 4.7 I -4.5 -3.7 -7.0 1.0 -3.3 -0.2 I 50-54 10.7 -2.5 0.0 1.7 I -0.2 -3.8 -5.6 I -2.6 . -5.9 7.0 I 55-59 2.0 - 5.7 3.7 3.67 -2.9 -4.9 I -4.8 2.4 I -1.6 -7.9 I 60-64 10.3I -4.6 6.6 I 3.7 ' -1.5 -7.6 I -3.7 -0.3 I -0.7 -7.5 65-69 9.0 I 3.3 6.2 I 6.0 -0.9 -5.0 2.41 -0.6 -0.8 -4.2 I 70-74 13.6I 2.3 11.0 I 1.0 1.3 I -4.0 -2.9I -3.1 0.2 -3.5 I 75-79 I I I I -2.6 . -1.2 I -1.2 I -2.6 -3.4 I 80-84 I I I -4.3 2.3 I 1.6 I -2.6 I -1.7 I 99 TABLE 7(m) I ICHANGE OF MORTALITY RATE FOR CD IN WHITE MALES l I ICEREBROVASCULAR DISEASE IN WHITE MALESI I I I I L I I F - I I I l Age I 1945 1950 1955AILI1960I1965 1970 I 1975 1980 1985 I1990I 1995 30-34 5 5 5 5 I 4 5 I 2 I 3 2 2 I 2 35-39 9 I 9 10 7 I 8 8 I 7 I 4 I 5 3 I 4 40-44- 17 I 17 17 I 16.I 18 13 I 11 I 11 I 7 7 I 8 45-49I 40 32 33 I 31 I 28 28 I 22 I 15 11 11 I 11 50-54' 77 72 58 51 T51 I 49 I 37 I 25 24 ' 20! 15 55-59 140 ' 129 113 97 I 90 83 75 I 46 41 31 30 60-64 235 241 225 184 168 I 166 , 138 92 69 54 59 65-69 414 418 437 366 322 ' 309 259 165 134 113 102 70-74 729 829 765 733 710 595 486 354 309A224 193 75-79 I 1122 960 652 538 427 407 80-84 I 1906 1704 1254 952 804 810 I 5 I I I I L I ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945- 1950 1955-61960-1965-7 1970-751975-811980-85I1985-90 1990-95 3034 0.2 0.4 -0.7 -0.4 0.8 -2.7 w 0.7 -0.8 -0.3 -0.5I 35-39 0.5 0.6 -2.6 0.6 0.4 -0.7 I -3.4 0.5 -1.6 0.6 I 40-44 0.3 -0.2 -0.6 2.3 -5.0 -2.6 F00 -3.8 -0.4 1.1 I 45-49 -8.0 1.3 -2.4 -3.1 0.6 -6.8 I -6.5 -4.1 I -0.4 0.7 50-54 -5.5 -13.1 -7.5 -0.1 -1.5 -12.6 -11.8 -0.9 -3.7 -5.8 55-59 -11.2 -15.9 -15.7 -6.6 -7.2 -8.2 -28.5 -5.8 -9.2 -1.0 60-64 5.7 -16.0 -40.6I-15.9 -2.6 -29.4 -44.8 -22.5 -15.0 4.5 65-69 3.8 19.0 -71.1I«13.5 -13.8 -49.9 -93.7 -31.4 -20.0 -11.4 70-74 99.8 -63.9 -32.3I-23.0 -114.3 -109.6 -132.1 -45.2 -84.7 -31.1 75-79 0.0 0.0 0.0 0.0 1122.3 -162.5 -308.1 -113.6 -110.8 -20.2 80-84 0.0 0.0 0.0 0.0 1906.1 -202.4 -449.4 -302.8 -147.6 6.2 I : 4‘ I I l I I I I I I L T I I . 2 I I ANNUAL PERCENT CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age 1945-8195504 1955-6 196E1965-7 1970-75497581198085 1985-90 1990-95 3034 0.9I1-2.4 19: 3.6 -10.5 6.1 -4.9 -2.1 .-4-2 35-39 1.3 I 1.4 -5.3 . 1.8i 1.0 I -1.7 -9.0 2.6 -6.9 I3.6 40-44 0.3 I03 07 ' 2.8 -5.4 -3.9 -0.1 -7.0 -1.2 ' 3.3I 45-49 -4.0 0.8I -1.4 -2.0 0.5 -4.8 I -6.1 -5.5 -0.7 1.4I 50541 -1.4 -3.7 -2.6 3-0-1- -0.6 -5.1 I -6.4 I -0.7 -3.1 1-5-7 55-59 I -1.6#-2.5I -2.8 I -1.4 -1.6 I -2.0 I -7.6 -2.5 I -4.5 I-0.6 60-64I 0.5 I -1.3 I -3.6 ' -1-7- .099 -3.5 I -6.6 49 L44 I 1.7 ; 65-69I 0.2 I 0.9 L-3-3 -2-4~. -0.9 I -3.2 I .7.2 I -3.8 I -3.0 I-2.0I 70-74= 2.7 ;-1.5I-0.8 I -0.6I -3.2 I -3.7 . -5.4 -2.6 I -5.5 T2.81 75-79I I I I ‘ -2.9 : -6.4 -3.5 I -4.1 ‘-0.9I 80-84I I T21 : -5.3 I -4.8 I -3.1 . 0.2I 100 TABLE (n) 1 ‘ . 1CHANGE OF MORTALITY RATE FOR 00 IN WHiTE FEMALES 1 ,CEREBROVASCULAR DISEASEYIN WHITE FEMALES 1 1 1 1 1 ‘ T1 1 1 r . 1 . Age 1 1945 1 1950 1 1955 1 1960 1965 1 1970: 1975 11980 19851199011995 1 30-341 4. 1 4 4 1 4 5 1 5 4 1 2 ‘ 2 1 2 ; 2 35-39: 10 1 8 ; 7 1 6 1 9 10 1 7 5 1 3 1 4 L4 40441 18 1 22 1 16 1 17 1 15 15 14 1 10 If 7 1 7 1 7 45-49 51 1 45 1 32 1 26 26 25 22 1 14 16 1 10 1 10 50-54 100 1 81 1 59 1 50 1 44 . 45 1 34 1 25 19 ’ 1F 18 55-59 144 L125 1 102 76 T 61 61 1 47 I 37 29 30 25 60-641 232 1 221 1 188 1 142 1 111 107: 92 65 1 60 44 41 65-691 434 1 385 ‘ 355 ' 295 231 209 176 1 128 103 80 73 70-741 712 1 756 664 565 516 - 419 362 249 194 173 179 75-79 1 1 1 1 882 747 1 479 388 340 330 80-84 1 1 1596 1066 D017 1 830 675 709 7 1 1 : 1 1 ; 1 1 2 1 1 ; i 1 1 1 4 1 1 1 1 1 1 1 1 ABSOLUTE CHANGE OF MORTALITY RATE DURING SUCCESSIVE PERIOD Age ' 1945-5011950-55 1955-60 1960-611965-7011970-711975-80 1980-81198591199095 30-34 -0.1 0.2 -0.3 1.1 0.2 -1.31-1.3 0.1 0.51-0.31 1 I 35-39 -2.1 1 -0.9 -1.3 2.8 0.9 -2.61 -2.0 ' -1.7 0.4 10.51 1 1 l —<>—--——oamOEE> Z. ww< ...02 (.55 8.00 5.8.0 - 8.8 888 888 888. 888. 888. -. -..-1111:; 11111111; 1 11 1.1.8.8 -888 888 888 888 888 88..--1111111.111111- - 1.11-881.8811- 188. 888 888 888 888 888 8.8 888 1 888 888 8 888 88-88- 88. 8.. 88. 88. 888 888 .1 888 888 888 8.8118.8 - 1888.18181- 88 88 88 88 88. 88. - 88. 88. 888 .88 88-81- 818.88 - - 88 .8 .8. 88 88 88 88 88 8.. 88. 118-81.11.88.881- 181.1|l8.81 88 88 88 88 .8 .18.- 88 ,1. I88- 88 88-881 1 .. .. .. - 8. 88 8.8 1.188. .8. 88 88 1 .881- 88-88 1 8 8 8--1-11.81111.. 8. -88--1181.111881- 181.11.- 8..‘ 88-88- 8 1811118 8- 8 .- 8 -18 -8 1 8... 1- .8111 8181188881 8 --8.--111811 8 8 81 8 - 81.1 81:8,- 8 ..8-881 8 . 8 8 8 .m- 8 8 8 8 -m-----8w81w81m1-. 8 . . - . 8 8 8 8 8 8 8 88-88 888. 88881 888. 888. 888. 888. 888-8, 888. 888. 888. 88811118? - 1111-1114111-- 8m._<_2 88.12. 2.88585 8580885888me ---:1 - 8 ..oo lommm1m11w1m1 . 11 -1-111 1-; -1 ..1-1 1L1- ;1-111-11-1 1m11- - 1 - 8 .88 <28 123 88 88. .8. 1|-.- L111.--- 11-1 :11 11 18181.88 88 88 11- 8.. - 88. 181818.: 118-818-1111 .818811- 88.8181 88 1.188 1 88 88 8t- -111181.m 11- .88-m1 .88--88 11 8.1 1T- 88 1.1-M818- 88 ..8-.- -1181.181:11 - 8m. 88-881 1188.1. I81. -- 118-8 11118.8. 1-8-m -11-111881111 Imp - @8818- 1 m 1 - -1-18 118-.11 118.. 188 1- 1.88 1.11818.- - 818888.8- - - 81 11 -8. - m 11.8. - 81 -18. 11mm- . 188-.88- 11.8.11- 18. 11 18. .1- 118- 1 mi 8 1118... -1 88-8-8- ’ ‘1 I:||:.|I1|'11||||.|. mum-.52 m._._I>> Z. mm_wzw._.mmm->I11 1. +l||1ii|-1 - 124 w Foo «...-<0 .8. 1-881.--.- 88 1 88 - 1.811: -18.- 1.811 .-.-81.1 118. 88885288. 8. .12. .2. 88590858188282 wkmwm>I 1 11'1“; 8311111111718 Valllnl 1:1 [I‘ll ‘rlt-il. 11.88.- - 1 88. 1-1- 88.8w-- 125 m 50 «8..-<0 11mm11 our mm... .80.. Nmm man 1 1111- 1 1 I11 11 11111 111 111 1Nmuwm11 1 mm mm m:- m: v:- mvm- 1 111111-11 111111 .11-1.- -. -. 1 |1 mh-mm :- 11mm|1111Nm—. [mm mm mm: Pmm1 v2.” 1 vmm 11- JM©1m1111 16m 1%1Wh-105 111mm-Ii-l-1-mhl1111i1m-W1 .815 - we. 11me mmm 1 mam. mm.“ mmh 1-1-g 188 1- 18811-188111 81811-885188 1888111818181 .8818.- -888 18188.88. 15$ 1 mm 1, av mm #5 cm --mmw Bar -1 imrm 1111 mmm 181mm1m1m1 - 88 111-88 -.8 88 188-111.88.1- 818..- .8. - 888 1.88.1 88.-88 1881- -- 8811 8.11 88 8. 8.81- 88 88. 88. -88.-1: 88..-88 . 28.8-11.8... -1 -8. 88 .81 88 -.88 88 -11-8.81-1 881.11-88.88 111R v N or w vm 11pm 11111| NF f- mm mm ..... 1 mW-MM1 1 m m 1.11-IF h 111$ mm B 1 m..-i. mp - 0.. 18 waom1 188181888... 888. 888.. 888. 888. 888. 8881.---11888.111 888. 1 88,81. 111-1 . mw4<2¥0<4m2_ mw_m2m._.mwn_>l 1 11 1 - 126 owoo <._.-m Zw>_0 < $0.... Z>>OIm w_ m I._-m_m 48$...sz Fmoo «...-<0 8888 8.88 8888 888.. -888 888- 128 Nmoo <._.oam .888 1-81. 11 1 _ - 1 1 . :é11... -- .Eoroov m 8.88.8 >8 8.8-.5288. 8658 2_ 88.8885 858088855886 8.02.2 - . 11 - , 8 8 - 11-111- 111... - . -18111-1. mmoo <._. 588 -- --1 11-. I: --11--1,1-1.11-1 _ 1.1 1.11- ,--;.-1..--1-.- -1- 1: 1.1.811 % 88801008828; 188.8 >8 88.2.2 88.12, 2.858.: ”.0 888888611--.- 1-1-+11. -11-81-1;1-11- 1.81- - 11- 1. 111- 11-..- 111 8 8 mwoo (.55 132 1- - - . 1-1 11- 1- 1- .1 1 11; 8mm 8% mmmm,-mmm1m mmmm 5.31 1- 1 i - - 11 -1 -1111;- 111- 8253 22 «m2 Wmmw-mom W111..- .1: 1 -- - - 1 1 1- 1 1 -1 1111 -m1mm1 one 82 3: $3 Wmfi-wvm-rimmrmmfi $2 -- 1 - E 1 - 1:- 111,-11m1m-m1-mmw mam Sm m5 561 m3 ,wmlm-1-1mw-ow-1vmmmwm8: 1 -1 - 1.. 9m 8m 8m. how 8m 16w. 8v ham 1QO .1 bay-m1- Bmf - - -- -- -..-i we NWT-RF Br Em New -E.o. -81m-1mom-1mmm1-1bmmg:1 - - -- -11.- mm 6! 3 8 @511.me «3 8w. mel-Imm-r Mow- - 1- - 117 -T-1R 8 3V 3 mm mm 2 -wW-Hm-Lo-Tbrw- - - - 1 - ---.e 3 & miklmzmePsfmwm37- - i-:- -- - @f-1-mI-E : 2 or 8 8 1% mma-rmw-_1-1- - - - - 1- - w1w1h!mlw m1: £1LTMVLW1-1H-s---1I-- m11w N N m v n m b: 3 111.11.- -1 1- 1-. 1 -1- .m--w1:--1-m- w v v w ww-a-111-1111-:-11-11--1;--1-.MJ:_11-111- ,8? $2 82 $2 $2 82 82 mm? 82 22 23 82 82 88 mmf mm? «Spam-mm- 1 11 5? cam-1:- -1 11111-11 11 1- 1 .- 1.11 ,1 M -- - fi 31. -1 fl - i- 1 1 5.0109 mzE_m>mwm._ 5.5 - 1 AHEOMOQ J mica—E m.__ Iii >m _m4<2_w:.-_>+> z_ mwfimfi E<_m: 0455mm -4111 L IIL11. 1.-I 136 once (.55 137 500 <._.m 322-NE; 2. $520. .m51pom<>0mmmmw& - - 7 I 11 .11 I11 .. .1I1Ill 1 #111; 11111111] 11 1.1 1- 11 . 111 111 111 +1. . 4 11 . v .. 1 #1 . | 1. .1 .1 .111 %I...11-1I #1 .I- 11%11 -1 11.11 111 1 1 1 11l*11-.1 11 i 1 1 1L 1 1 1111-1 1.11111 11 11f! - .1 - 1 l 1 11 11 T ’ + Vmco :E_N>m 3.25”. 5.132. mm0mmmmmo -11-N114 11 l4] mmoo <._. IE5 >m mmézmu xofim 2. $590 mfiaom<>ommmmmo E! I - .--: Eire 1E 250 (.56 143 - E--. II .I-.--.E.I.I---. E 8 cm Fm E L F8 :mmm fil. -.E-EIE- 9. -mm mm 9. «NF: FmF-EI- E-- i- -E- ., E .- . mm om R em 8 - oFF mt mom mmm I IE. II 3 NF 9 um I: we www-1- EmFImmw -.bWNI IE--- E. I: m :E-mFI-z-F-WI-mw-ImmEr» . .mIFFE-me .- IEEFI IFI-----mI-IN-F-I.mr-Imm mm a Imm- Imm-I -- I .-I -EmeE-F-I ~-...---mII-F ImF FF .2 ..IFm; 3--.: LI m.- Im I wI---.~--IFIF----mII.-m--w-L-..F-IIm- - I , -mI-I--.FI-.--w--. .N F F m :IMEWI :I-IIE - - o F FE- N F- m EWI F. m EEI- L .- E IEE F o OI m o o E FIIN I --.--I-EIEI-i- mmmF mvmw mva mm? mm? 89 mm? 93 99 we? momF-mMWMWEm-mmif- mmmFE mme IIIEE l-JI 23> cEm , -I EL. E IFEoIooF «am; Ema E 3.22 MEI; 2. $590 Easzmm>fl 1E- I E .EIEI - E :LIE ,- .- ,, 144 mmoo <._._mszmm>m- EV: IEIEL -Nmm - YIEI EI FEEE .. . - E EEEL . ..mww-MWIWOM mmoo <._. Ema ham-22 xofim 2. $590 Eda: mzmflEmnFF-z IIE , I .I I . . . - I- E E.I.IIE IIN-Em- EI - IE FIE LIL--II- “....-- -EIMI- E -... ovoo <._. I.E_m >m wmI_<_2wu 205m 2. wm_mzw._.mmm>IIEEI E ...... E II IE I I . E - EI. II: IIEI- E - I-IIE E EI IE-.E!II E II-I.-E IT .I.I-IE I: I- -IIII 147 VNION I O. I mNImN . . I . . vmém I D. I wmImm . Id. I .JYOVII&?II mVImv vmém . .... I mmImm EXII. voémlldl mmém . . 9 . . VNION l I. l mhhh ET va0m '0' m0 w0< mmm F ommF mam F 0mm F mhm F 58 mmDGE Ik 05a F mmm F omm F mmm F 0mm F mvm F F ...Omumm Dav—mum Fem-ow mws mm._> z. mm05m mum F Ohm F mom F omm F mmm F omm F mvm F d _I E F ooooF Fomtm 8.5.. 3.8 was mm._ mmm F omw F mam F cam F mum F cum F mom F com F mmm F 0mm F mvm F F F F F _ F E- F P 1 d F H I— T fi u u ...OmuEm DOE—mm Acmém mws mung: xofim z. mwOEE 0mm F mam F 0mm F mum F 93 F mom F com F mmm F 0mm F mwm F p p F — F b b _ _ « fi 4 F a F a fi - P hummum DOEmm SEAN mus mun-52m...— ¥O<._m z. mm05m<0 $0.45 OOO‘OOI- 83d aura 15] vmém I O. I mN-mN- . I . . vmémIUI mm-mm I £. I vvév IXI mv-mv vmém . I0. I mm-mm IXI Vmémlldl mokmm. - 9 . . 3.65 I I. I mnénlI vméw IDI m0 m0< mmm F 0mm F mom F 0mm F mooo NEDGE I._. mum F Ohm F mmmF 0mm F mmm F ommF mvm F F Fountm noEmn $98 was 3.2: mEEs E mmm F 0mm F mam F owm F mum F 05m F mom F owm F mmm F 0mm F mvm F F 152 vNéN I O. I awrmm. . I . . vmém I D. I mm-mm . II. I vvév IXI mv-mv vmém . I0. I mm-mm IXI vm-ooI.flI mobw . . 9 . . Fukuon I I. I mh-mh||§ll cooF wwém IUI 10 w0< Fomtm 92mm“. 3.8 was $.23“. mtzz E5: "_o .39me OOO'OOI- Had SLVH 153 vmém I O. I mmém. . I . . vmém I U. I mm-mm . II. I vvév IXI mvév vmém . I0. I mm-mm IXI vméwllfll m®-m®. . 9 . . «$-05 I I. I mnénll vwém IUI m0 m0< mmmF 0mm F mam F com F Soc mmDOE ...._. mum F chm F mow F com F mmm F ommF mvm F F I oF ooF I ooooF Swarm. 3.5.. 3.8 was 3.22 xofim E mum F Ohm F mma F omm F mmmF 0mm F mvm F _. ooooF Fomuum 00.5.. 5.8 was $.25”. xofim E mom. 8.: 82 82 mm: 2.2 $2 82 mm? 82 $2 to ..Vméulol F 33 I 9 I 5-8. . a. . - 8-8 IxI 4. . . vvévII: .....d ....... 4 ........ 4 ....... d. ....... 4 .............. A woF 9.-va xIIIIIIx/x a $8 . I . I ....... I. /xIIIIx/ I x x x xIIIIx $-mmlol ...... I ....... I ....... I. . . v0-00lx.l ....I ....... I ....... I ....... I ....... I ....... 00F 8-8:... ....... I.I.I.I IT E-2 I . . I ............................................. 3-2.3... ..... ....9.... .... x. ..... x ...... x ...... x , .Ix .... I? II xI EQIQI ...é ....... o ....... o ....... 0......I.H.Iu.... ...... o ....... o: 82 ....... 4- . . . . . IIIIII . ..... I ...... u.u.......u..... .... u ....... . mm 93 ....... .o oooop howumm DOE—mn— AemIch mm._<_2 m...=._>> z. mw mhm F Ohm F mmmF 0mm F mmm F ommF mvm F Fd ........ 4 IIIIIII ..82 . ..... o ...... I ....... 4 ....... « ....... < . .o ..... .0 ..... .0 ..... 6 0002 hommn—m DOEmn Aeméw mus mm._<_2mu_ m._._:>> z. mm mam F 0mm F mom F omm F mum F Ohm F mom F omm F mmm F cmm F mvm F o F F 0 \II ix d\ / \I/ \\ oooooo I \ I \ 4f IIIIIIIIIIIIIII a nnnnnnn q I I \\ III II \ II III a ::,.I.... .. S < IIIIIIx .. ....... < .. <22 XIIIIIIII) ) / <- X\\XIIIIIIX/X ...II.... 000000 I‘ll-on x/XI|.IIII-I‘unllx-|||I\..II||||\.I ' III-.- ...- ........... .. ....... I ......... 09—. I ....... I ....... I... U llllll x aaaaa ' It! ‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘ | III. - ...... 9 x ...... XI! .x ...... I”... II: oooF ooooF Fomumm 8.5". $98 was 3.2: gain 2. mm mum F Ohm F mmm F 0mm F mmm F ommF mva Fd Pomp—um DOEmn. Avméw mws mm._ mum F Ohm F mom F 9mm F mmm F ommF mvm F F.o . oooF oooo F #0975 09mm.“— Aeméu m0> z. mm0mmwmmo OOO‘OOI 83d BLVH 160 vméw I O. I mNImN . . I . . vmém . . U . . mmumm . II. I VVIOV IXI mVImv vmém . II. I mm-mm IXI vmémlldll mQImm. . 6. . . vNION I I. I mummh IOII VQIOQIUI no m0< mmmF 0mm F mam F owm F :8 meOE 1.530 to m mum F Ohm F mom F omm F mmm F 0mm F mva F.o ooooF 595 3.5“. 3-8 was mm._ommmmmmo OOO‘OOI» 83d SLVH 16I VNéN I O. I mN-mN- . I . - vmém I nv I mm-mm . II. I 370v IXI mvév vaOm . II. I mm-mm IXI vodmIdII mohm. . 9 . . EéhIII mNImn IT vmémln—I n10 m0< mmmF 0mm F mam F 0mm F mFoo mmDGE Ih mum F Ohm F mmm F omm F mmm F ommF mva F.o homuum 8.5; 23.8 was 3.2: xofim z. 35%. mfiaom<>ommmmmo OOO‘OOI- 83:! 5mm 162 VNION I O. I mNImN . . I . . v0-9.“ I D. I mMImm . II. I vvév IXI mvbv vmém . I0. I mm-mm IXI vmém Incl mmImm . . 6. . . whéh I I. I afimhlol vwémIDI mam F camF mam F omm F oFoo mEDOE :._. mum F Ohm F mom F com F mmmF 0mm F mvm F F.o OF 2: coo F oooo F 595 8.5.. $98 was mmismu gofim 2. $3me mfiaom<>ommmmmo OOO‘OOI- 83d ELVH 163 vmém . . .6. . . mmImm I .T I vvév In_.I mVImv I I. I vmém . . I . . mmImmI vmém I X. I mmfim . ID. I VNION ...... mNImh I I I vméw m0 w0< mam F ommF mam F owm F too mmDGE Ih mkm F Ohm F mom F 0mm F mmmF ommF Fd , oooF hymn—um DOE—mn— Avmucn was mun—<5. m._._I>> z. mm_mzm._.mmn_>_._ OOO‘OOI» 83d 31w 164 wFoo mmDGE I._. mmm F 0mm F mam F 0mm F mum F Ohm F mwm F 0mm F mmm F 0mm F . F.o \‘IIII'IT/ III oooF ...Umnmm DOE—mm €060 m0> Z_ mm_w2mhmma>1 00L 83d ELVH I 000 165 vmém . . Q. . . mm-mm I .T I VFIOVIIUI mvév I I. I vmém - . I . . mm-mm+ vmém I X. I mm-mm . ID. I Show ...... awmh I I I vméw no m0< mam F mFoo wIDGE Ih 0mm F mam F 00m F mhm F Ohm F mom F 0mm F mmm F 0mm F F.o I‘d; III III 11F ... \+. ... Av uuuuuuuu Q\\\\.\ // on: \‘T / 1; .K\ x .4. ........ 4.. a": / .- a II on I / \ I: II can: I OF . \I x \ \I. C. .. 4 ........ a ........ II I: +\ .\ / / coo \\\ \\+/ A \\ ‘.oo o/ \I\ at I/I<\ \\ I, \\ III on: on I \ / / .I.III‘K‘ I, \I ... ... . . .TII .... IIIII u ,- ........ . human—m 00.1mm 25.2" was mw.__mzm._.mmn_>_._ OOO‘OOI- 83d axvu 166 vmém . . .0. . . mMImm I .T I vvév IIUI mv-mv I I. I vaOm . . I - . mm-mm IT vmém I X. I mmém . ID. I VBION ...... wwmh I I I vmém m0 w0< mam F 0mm F mam F owm F omoo manor. 1.5mm no m mhmF Ohm F momF 0mm F mmm F ommF F.o TGF 2: coo F .Fomuum ooEma €98 m9: $.25". xofim z_ mm_mszma>: OOO‘OOI» 83:! 31V! I67 VNION I I. I mNImN . . I . . vmém I D. I mm-mm . I. I vvév IXI mVImv vmém - II. I mm-mm IXI voémIdll mmImm. . I . . FKION I I. I mNImhIII vwémIUI m0 w0< Fmoo wEDGE Ihm_muom 6 6 6 6 6 6 6 6 6 6 6 8 8 I. I. ummmmwmmmummmmmwmeeee F fl OIIOII OI. I...l-.I... .I.I...II?! I. .l. 6|) 43 a. u.: /‘ UIIDIID. ...... . 'nl .. II I.Inl..nrla I II I ‘I‘ i In], | w%%%%mmmmmmmmmmmmmmmm 8 8 Do 8 8 8 8 C 8 8 8 9 8 8 8 8 8 00 8 8 _ .I I I .I I I I. w .. I ._ I I I I I .. ._ I F f 1r! \I?! I. I ”I .I..I... ..IWI v~-omI9I n7 ...6... I 'D'nfl!’ Ila .I.I" mNmN..I.- U. I/ I In: .\Ir.._q. allilb/l II I I 0'. \|II I mmmm II N I.II\.IC.I.1./d vvévIXI mv-mv vm-om-II.I :ooF mm-mmeI vmémIldll m®-mm..I.. vnénlII afimnIOI :oooF vméwIDI ...; Q0w0< ooooF PUMP-hm ...:OIOO Avmuow m0> z. mm05m<0 m05<2 OOO‘OOI 83d 31‘!!! 169 vmém l O. l mmém. . I - . Vmém l U. I mm-mm . I1. I vvév IXI mvév vmém . :0. l mmbm IXI! vmémnlfil mwbm. . O . . whéh l I. l mh-mh IT vmém IDI 10 w0< 81.6 L 896 L 896 L 896L 896 L 876 L €176L 886 L 886 L mmoo mEDOE 1.520 “_O m 836 L 836 L 8L6L 8L6L 806L 806L 868 L 868 L 888 L 888 L 818 L 8L8 L Swarm #518 26.8 was 352 xoSm 2. 3.3.99 mfizom<>oam 836 L 836 L 8L6L 8L6L 806 L 806 L 868 L 868 L 888 L 888 L 8L8 L SL8 L -. 0.. .- cow 1 . ooow 0000 F ...Omn—um ...mOIOO Aemém m6OEm 6 6 6 6 6 6 6 6 6 6 6 9 % S 7 8 8 Z Z L L m w 6 6 8 8 I. I. 8 8 8 w W 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 f 816 L .... 1' .1. I? I ......I. II I..:il.l.ll|.i.ll....1 u/o/nIquIn ...Omuu—m PCOIOU $0.0N mus mung—z m._._...>> z. hm m m m 7g 7y I 9 8 8 8L6L 806 L 806 L 868 L 868 L 888 L 888 L 8L8 L 8L8 L : 0.. .- 00w .- 000w BEE. #5100 .38 was $3.2m“. mtzz 52!. ".0 $935 0000 F OOO‘OOL 83:! 31MB I73 vm-0m I O. I 0N-0N. . I . . v0-00 I D. I 00-00 . I‘. I 310v IXI 0v-0v v0-00 . .0. I 00-00 IXI v0-00llQI 00-00. . I . . vuén | I. I munmhIOI @000 III Q0 0.9.. 5N8 MCDGE IPEmn—Om mmwwmmmmmwmmmwmmwmwmm .. \O \ / 6. \6. xx \ I \ #0.. I”)-.. . u "If .. ...”..-- .- m/ .. .1- ... \ \ I.. /D. W ID.\ ‘II‘II‘I“ ll O\‘ll|‘|l' I‘ll Itoov .‘ OI..0.I¢I..OI.O..IOI.O.IoI.O IO! XI X X x IXIXI IXI . .0 x xix/x :82 . IIIIII ‘III.IIO 0 I. 1...:6 .6 .. .. I . ..I...l...l.l9.l IRIM... I I...- 0000.. ___.Umn—u—m #10100 .38 was 3.22 xofim E 886 L 836 L 8L6L 8L6L 806 L 806 L 868 L 868 L 888 L 888 L 8L8 L 8L8 L Q\.P. (I \l [.I.I-III ‘ : 00.. 0000 F Swarm onoo .88. 3.2—am". xofim E.3: ".0 $935 OOO'OOL Had SLVH 175 vm-omI-OI 00.00 I I. I vm-om - - 4. - . 00-00 IXI vv-ov. - I - . 0v-0VI1I v0-00 - I . I 0000+ .000 l X. I 00-00. . I - . #505 05-0.... . .1. . . V000 . I0. I 00 004. 816 L 896 L 896 L 896 L 896 L 876 L 8V6 L 886 L 886 L 0000 mmDGE ...—.50 no m 836 L 826 L 8L6L 8L6L 806 L 806 L 868 L 868 L 888 L 888 L 8L8 L 8L8 L 9.0 .- 0F .8 00.. .- 000w Puma—mm ...:OIOO $0-0m m0 I? L I? L L DIP In? I? II? I? II? I? I? If I? 'P I? If ..I' III 6 6 6 6 6 6 6 6 6 6 6 6 6 8 8 8 8 8 8 I I 8 I. I. wwmmwwmmuueemmwmmeae 9.0 -F :2 :00? .x:.. ....mm..:-....-..-..-..-............. d...0...¢...o...o...o...Q...o...o d-Ifl: 0.000F did-......- 0I.o..lo 4 .¢-I9:.-o 0009 ...Umuum ...mOIOO $0.0m mmvs mm-_> z. mm 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 8 8 8 I. 9 9 9 9 W W 8 8 Z Z L L O 0 6 6 8 8 I. I. 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 ...OmnEm PGOIOO A v0-0N was mm4<5 x050 z. mm m m m .6 a L 8 8 8 8L6L 806L 806 L 868 L 868 L 888 L 888 L 81.8 L 818 L —.0 -0F - 009 .- 000w 0000p POW-Em ...mOIOO $0-00 m6 15.1.0 8L6L 8L6L 8L6L 896L 896L 896L 896L 876L 8V6L 886L 886L 886L 886L 806L 806L 868L 868L 888L 888L 8L8L 8L8L fio 179 V0-00II 00-00. . I . . V0-00 I D. I 00-00. - I - - VV-0V+ 0V-0V I I I V0-00 - I . I 00-00 I X. I V0-00I 00-00. - Av . - V5-05 . II. I 05-05 V0-00IDI - 0F .- 00w .- 000.. mOmO< 0000 .. hymn—mm PEOIOO $000 mmvsmm-Es. 0.5:; 2. mm0mmmmmo OOO‘OOL 83d SLVH 180 V0-00 I 00-00. . I - . V0-00 I D. I 00-00 . . I . . VV-0V+ 0V-0V I O. I V0-00 - I - I 00-00 I X- I V0-00IIO-I 00-00. - O - - V5-05 - II. I 05-05 V0-00ID.I 0000< 8L6 L 896L 896L 896L 896L 8V6L 8176 L 886L 886L V000 memzu. 1.5I—.0 “.0 m 826L 836L 8L6L 8L6L 806L 806L 868L 868L 888 L 888L 8L8 L 818 L ...0 -- 0.. : 00F - 000w ...Ommmm PKOIOO €0-00 mmvs "Wm-2.20“. WEI; z. mw0mmmmmo 0000 P OOO‘OOL 83d BLVH 181 V0-00I 00-00 - . I - . V0-00 I n I 00-00 - . I . - VV-0V+ 0V-0V I I I V0-00 - I. I 00-00 I X. I V0-00IIT 00-00- . ¢ - . V5-05 - II I 05-05 V0-00IUI anG< 8L6L 896L 896L 896L 896L 8V6L 876L 886L 886L 0000 0.000.“. ...—Ema “.0 m 836L 836L 8L6L 8L6L 806L 806L 868L 868L 888L 888L 8L8L 8L8L 00 ... 00.. 0000 w human—m ...mOIOO $0.00 was 0m4<2 ¥0<-_m z. mwOmmmmmo OOO‘OOL 83d SLVH 182 V0-00I 00-00. - I . . V0-00 I U. I 00-00. . I . . VV-0V+ 0V-0V I O. I V0-00 . I. I 00-00 I X- I V0-00IO-I 00-00. - Av - . V5-05 . II. I 05-05 V0-00IDI 00 wO< 0000 meGE m ...—.50 mw0000m000000006689L4 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 F0 L-F ---> I ‘OII‘I ‘6‘ al'DHI‘OI D]; at 0 \ ..13..\ \ Ju.\ o. . n i: '0‘ l I! IO-II l-II.III. -\. .J-I. XII I ..I ..II I.- .-0 +02 0000p ...ownEm ..8—0:00 2000 max: mm-_0mmmmmo OOO‘OOL 83d BLVH 183 V0-00 . - I . . 00-00 . IX. I VV-0VIIDI 0V-0V - II. I V0-00 . . I . . 00-00 I V0-00 I X. I 00-00 I D. I V5-05 ...... 05-05 . I . I V0-00 00 00¢. 5000 0030.”. ...—.50 ...—0 m L L L L L L L L L L L L L L L L L L 6 6 6 6 6 6 6 6 6 6 6 6 6 8 8 8 8 8 9 0.. 9 w .7 8 8 8 70 I I 0 0 6 6 8 8 I. 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 P0 . . n V A ‘ ‘ .T or .- 00.. 000 F PUmu—um #00100 2.0-00 m0.mzm._.mmn_>1 OOO‘OOL 83d BLVH 0000 0000.“. 1.5—.0 ".0 m L L 6 6 7v L 8 8 896L 896L L 896L €76L 896L 886L BZSL 8L6L 806L 806L 868L €68L 888L 888L 8L8L X mm .V 8 I IX: 4 184 V0-00 . - I - . 00-00 . IX. | VV-0V I.DI 0V-0V . II. I V0-00. . I - . 00-00 I V0-00 I X. | 00-00 I U. I V5-05 ...... 05-05 . I . I V0-00 .- 0w -. 00w 0.0 wG< 000w hommmmhm0100 $0-00 00$ 00.2.20". 0.513 2. mm.mzm.—.mmm>1 OOO‘OOL Had SLVH 185 V0-00. . I . . 00-00 . IX. I VV-0VIDI 0V-0V - II. I V0-00 . . I . . 00-00 I V0-00 I X. I 00-00 I D. I V5-05 ...... 05-05 . I - I V0-00 n.0 0.06. 0000 00.00.... :hm.0n_0m L L L L L L L L L L L L L L L L L L 6 6 6 6 6 6 6 6 6 6 6 8 8 8 8 8 8 8 I. w m m w .... 3 III a a m m m m 3 III e e .- F0 .4. I. I? ...4. .x. 4 . z x . KI IrO—v I L.00F 000w Sunni onoo .38 was mum-Es. xoSm 2. $535 E.3: m>_mszmn_>: OOO‘OOL 83:! ans 186 V0-00 - . I . . 00-00 . IX. I VV-0VIUI 0V-0V - II. I V0-00. . I . - 00-00 I V0-00 I X. I 00-00 I D. I V5-05 ...... 05-05 . I - I V0-00 30 00¢. 0V00 0000.“. 05.0 no m L L L L L L L L L L L L L L L L L L 6 6 6 6 6 6 6 6 6 6 6 6 6 8 8 8 8 8 9 9 9 .V .7 8 8 8 a L L 0 0 6 6 8 8 I. 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 to - F 4. IXI. -\ K .- o. I . <. .. . .I. ..:/... --8. I Ill .. -/. 9..-bl. II ..E- J- I I. - .I. {I’D I /. ..:. l/InT.nl.-D .I JR. - coop human—0 ...mOIOo $0-00 00$ mun—(20.... 2050 z. mm_0zm._.mmm>_.. OOO‘OOL Had SLVH APPENDICES Appendix A 700 - thoa 600 ~ 500 - E; 3 h 400' Ybar 1880 8. D o E 300 - Your 1910 g x 8 200. Ybar 1930<__‘ 100 I- \ 0. 10 26 36 46 so 60 70 Age in your. ludunmhumflhscknuxraum honztnbennwmmr—au knnuk—by mag 1880,1910,1930. 187 Appendix B ‘0) P II I III I - I0 your «or... nun b) Qghng QgIg 991119ng , duInqu by control your oI «th «signaled by control your oI bum- I00 ;: IOO E : _ h b I o 2 I0 : 5 '0 I: > .. 2 : 3 : J h b O " .- 8. : §. . g _ 3 . c c u 1' | : z I E u '- I = = = - c - ‘ ' E .- 35 -. ‘ :- g I- O h : h .. '- d955 -0l 1 1 1 1 1 1 1 -0I 1 1 1 1 1 1 20304050007000 20304050007000 ‘6! AT DEATH 00! A? DELHI 5.0) Magnum - I0 you our... run 0) Qghgn Ag! anggg I! '-- «ugh-I“ by coma! nor 0! «.II ...-".1“ p, “on“ ”g, .4 gm. aa- -- too 0- - . E 75>+ - - I0 5 no : ’5‘ . Z : !§w' - 55-64 -' - 1 . '- I vs - ‘8’- - ' I; 45-54 § - I 45-54 ‘ I 5- : 1' I L- I I : U I— I- I | I 1: : 2 : l I ' g - 35-44 4 - , I : 1; ~ ‘ - . : - 2.- - 25 ~ : . : g I- :0 " I : I y 25‘34 O I I I 25‘34 f. I [01 : O'I '- I. : ' I- : , l 1 : : 1 I I h l5-24 " : I I [5.24 h- l' l h- I- ' I I .' I ' h l- I I ' I l , -0, I 1 1 1 1 1 1 .0. l J l 1 l l l l l 1 J 1 ISI 5 I035 I055 I075 I035 I053 I075 I005 IOI5 I035 I055 YEAR OF DEATH YEAR OF IIITII Death rates from gastric ulcer in males. England and Wales. 1900—1977. 188 LIST OF REFERENCES 10. LIST OF REFERENCES American Heart Association: 1995 Heart and Stroke Facts. Dallas, TX, American Heart Association, National Center, 1995. Barrett JC: Age, time and cohort factors in mortality from cancer of the cervix. J Hyg Camb 71: 253-259, 1973. Becker LB, Han BH, Meyer PM, et a]. Racial difference in the incidence of cardiac arrest and subsequent survival N Eng J Med 1329: 600-6, 1993. Broderick JP, Phillips SJ, Whisnant JP et a1. Incidence rates of stroke in the eighties: the end of the decline in stroke? Stroke: 20:5, 577-82, May 1989. Brownson RC, Smith CA. Racial differences in cardiovascular disease mortality and risk factors. Missouri Medicine: 89: 1, 42-45, January 1992. Chung EK. Quick reference to cardiovascular diseases 2nd Edition. J. B. Lippincott Company 1983. Cooper R, Stamler J, Dyer A et aI.The decline in mortality from coronary heart disease, USA, 1968-1975: J Chron Dis. 3:709-720, 1978. Davis DL, Dinse GE, Howe] DG. Decreasing cardiovascular disease and increasing cancer among whote males in the United States from 1973 through 1987. Frost WH. The age selection of mortality from tuberculosisin successive decades. American Journal of Hygiene, Section A, 30: 91-96, 1939. Gardner MJ, Osmond C: Interpretation of disease time trends: is cancer on the increase? A simple cohort technique and its relationship to more advanced models. I Epid Common Health 37: 274-278, 1983. 189 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. I90 Gillium RF, Liu CK. Coronary heart disease mortality in United States blacks, 1940-1978: Trends and unanswered questions. American Heart J Sept. 1984. Gillum RF. Cardiovascular disease in the United States: an epidemiologic review. In: Saunders E. Brest AN, eds. Cardiovascular diseases in blacks. Vol. 21. No. 3 of Cardiovascular Diseases in Blacks, 1991. Gillum RF: Coronary heart disease in black populations: 1. Mortality and morbidity. Am Heart J, 104:839-851, 1982. Gillum RF: Sudden coronary death in the United States. 1980-1985. Circulation,79:756-765,1989. Glenn ND. Cohort Analysis. Quantitative Applications in the Social Sciences. Series No. 07-005. Beverley Hills, California: Sage Publications. Havlick RJ, Feinleib M. Proceedings of the conference on the decline in coronary heart disease mortality. Bethesda , MD, US Department of Health, Education and Welfare, Publication No 79: 169, 1979 Holford TR. An alternative approach to statistical age-period-cohort analysis. J Chronic Dis, 38:831—40, 1985. Holford TR. The estimation of age, period, and cohort effects for vital rates. Biometrics , 39:311-24., 1983. Holford TR. Understanding the effects of age, period, and cohort on incidence and mortality rates. Annu Rev Publ Health , 12:425-57, 1991. Howard G, Anderson R, Sorlie P, et a1. Ethnic differences in stroke mortality between non-Hispanic whites, Hispanic whites, and blacks. The National Longitudinal Mortality Study. Stroke, 25:11, 2120-2125, Nov. 1994. International Classification of Diseases, 9th Revision. 3rd Edition. DHHS, March 1989. James IR, Sega] MR: On a method of mortality analysis incorporating age year interaction, with application to prostrate cancer mortality.Biometrics, 38: 433-443, 1982. KeiI JE, Saunders DE, Lackland DT, et a]. Acute miocardial infarction: Period prevalence, case fatality, and comparison of black and white cases in urban and rural areas of South Carolina. Am Heart J, 109:776-784, 1985. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 191 Kleinbaum DG, Kupper LL, Morgenstem H. Epidemiologic research: principles and quantitive methods. New York: Van Nostrand Reinhold Company, 1982. Kuller LH. Commentary on coronary artery disease in blacks: Public Health Reports. 1101570-579, 1995. Kupper LL, Janis JM, Karmous A, et a1. Statistical age-period—cohort analysis: A Review and critique. J. Chronic Dis , 382811-30, 1985. Lee MH, Borhani NO, Kuller LH. Validation of Reported Myocardial Infarction Mortality in Blacks and Whites: A Report from the community cardiovascular surveillance program. AEP Vol. 1. No. 1. 1-12., October 1990. Liao, Y, and McGee, D. : Population attributable risk of blood pressure among blacks and whites. [Abstract]. Am J Epidemiol 138:592 (1993). Ford, ES, and Cooper, R.S. : Racial/ethnic differences in health care utilization of cardiovascular procedure: a review of the evidence. Health Service Tes 30:237 252 (1995). Magnus MH: Cardiovascular Health among African Americans: A review of the health status, risk reduction, and intervention strategies. Am J Health Promotion. 5:282-290, 1991. Manson JE, Ridker PM, Gaziano JM, Hennekens CH et a1. Prevention of Myocardial Infarction. Oxford University Press. Oxford. 1996 MMWR, Vol. 46 / No. 7, Feb. 21, 1997. Moolgavkar SH, Stevens RG, Lee JAH: Effect of age on incidence of breast cancer in females. J Natn Cancer Inst. 62: 493-501, 1979. Moolgavkar SH and Stevens RG: Smoking and cancers of bladder and pancreas: risks and temporal trends. J Natl Cancer Inst 67: 15-23,1981. Moriyama IM, Krueger DE, Stamler J et a1. Cardiovascular diseases in the United States. Harvard University Press. Cambridge, MA. 1971. National Center for Health Statistics : Monthly Vital Statistics Report. Vol. 28, No. 11 DHEW, Rockville, MD. Supplement Feb. 29, 1980. National Center for Health Statistics: Comparability of Mortality Statistics for the Seventh and Eighth Revisions of the International Classification of Diseases, United States. Series 2. No 66, DHEW, Rockville, MD. 1975 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 192 National Center for Health Statistics: Comparability of Mortality Statistics for the Fifth and Sixth Revisions: United States, 1950. Vital Statistics Special Reports, Vol. 51, No. 2, DHEW, Rockville MD. Dec. 1963. National Center for Health Statistics: Comparability of Mortality Statistics for the Sixth and Seventh Revisions: United States, 1958. Vital Statistics Special Reports, Vol. 51, No. 4. DHEW, Rockville, MD. March. 1965. National Center for Health Statistics: Comparability Ratios Based on Mortality Statistics for the Fifth and Sixth Revisions: United States, 1950. Vital Statistics Special Reports, Vol. 51, No. 3. DHEW, Rockville, MD. Feb. 1964. National Center for Health Statistics: Health, United States 1990, DHHS Pub 26.No(PHS)91-1233. Public Health Service. US Government Printing Office, Washington, 1991. National Center for Health Statistics: Monthly Vital Statistics Report. Vol. 17, No 8, DHEW, Rockville MD. Supplement, October 25, 1968. National Heart, Lung, and Blood Institute Report of the Task Force on Research in Epidemiology and Prevention of Cardiovascular Diseases. US Department of Health and Human Services, Washington, DC. August 1994 Osmond C, Gardner MI, Age, period, and cohort models. Nonoverlapping cohorts don’t resolve the identification problem. Am J Epidemiol, 129:31-5, 1989. Osmond C, Gardner MJ, Age period and cohort models applied to cancer mortality rates. Stat Med, 1: 245-259,l982. Peltonen M, Asplund K. Age-Period-Cohort Effects on Stroke Mortality in Sweden 1969-1993 and Forecasts up to the Year 2003. Stroke, 27: 1981-1985, 1996. Plan and operation of the National Health and Nutrition Examination Survey, United States, 1971-73. Vital Health Stat [1] Nos. 10a and 10b. DHEW Publication No. (HSM) 73-1310. US. Government Printing Office, Washington, DC, 1973. Plan and operation of the Second National Health and Nutrition Examination Survey, United States, 1976-80. Vital Health Stat [1] No. 15. DHHS Publication No. (PHS) 81-1317. US. Government Printing Office, Washington, DC, 1981. Sample design. 1988 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 193 Sempos C, Cooper R, Kovar MG, et al; Divergence of the recent trends in coronary mortality for the four major race-sex groups in the United States. Am J Public Health, 78: 1422,: Third National Health and Nutrition Examination Survey. Vital Health Stat [2] No. 113. DHHS Publication No. (PHS) 92-1378. U. S. Government Printing Office, Washington, DC, 1992. Stevens RG, Moolgavkar SH : Estimationof relative risk from vital data : smoking and cancers of the lung and bladder. J Natl Cancer Inst , 62: 1351-1357, 1979. Stevens RG, Moolgavar SH: A cohort analysis of lung cancer and smoking in British males. Am J Epid , 119: 624-641, 1984. Susser M. Periodic effects, generation effects and age effects in peptic ulser mortality. J Chronic Dis., 35:29-40, 1982. Sytkowski PA, D’Agostino RB, Belanger A, et al. Am J Epidemiol 1432338-50, 1996. US Public Health Service: Health United Statesl996 Hyattsville,MD, US Dept Of Health and Human Services, DHHS publication No. (PHS)91:1232, 1991 Wilson JD, Braunwald E, Isselbacher K] et al.Harrison’s principles of internal medicine 12th Edition. New York: McGraw-Hill, Inc. 1991. Xu X, Laird N, Dockery DW et al. Age period and cohort effects on pulmonary function in a 24-year longitudinal study. Am J Epidemiol , 141:6, 554-566,1995. Yoon GY, Kapadia AS, Canfteld MA et a1. Cardiovascular mortality trends in Harris County, Texas: 1980 to 1986. Texas Medicine, October , 85:27-30, 1989. Zopf EP. Mortality patterns and trends in the United States. Greenwood Press, Westport CT. 1992. MICHIan STnTE UNIV. LIBRARIES IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 31293017065552