Tobacco Documents Online

I THE AUTHORS REUEL A. STALLONES (`The Rise nd Fall of Ischemic Heart Disease") is dean and professor of epidemiology at the School of Public Health of the Uni- versity of Texas. He is also James W. Rockwell Professor of Public Health at the university and adjunct professor of environmental science and engineering at Rice University. He received his un- dergraduate education at Visalia Junior College, Ripon College and the Univer- sity of Michigan. He got his M.D. at Western Reserve University in 1949 and his M.P.H. (Master of Public Health) at the University of California at Berkeley in 1952. In 1950-51 he was a battalion surgeon in the U.S. Army. From 1954 to 1956 he was assistant chief in the de- partment of epidemiology at the Wal- ter Reed Army Institute for Research. From 1956 to 1968 he was on the faculty of the University of California at Berke- ley. He joined the faculty of the Univer- sity of Texas in 1968. Stallones' princi- pal research interest, which is reflected in the theme of his article, is the epide- miology of cardiovascular disease. ~ received his master's degree in 1950 and his Ph.D. in 1953. He joined the staff of the Weizmann Institute in 1954. In 1977-78 he was Fogarty Scholar in Residence at the National Institutes of Health and visiting professor in the de- partment of biological sciences at the University of California at Santa Barba- ra. This year he is serving as chairman of the Federation of European Biochemi- cal Societies. The range of his scientific interests is best illustrated by the titles of the articles he has contributed to SctsN- 'rmc Absmucwrr: "The Bacterial Cell Wall" (May, 1969), "Glycoproteins" (May, 1974) and "Lectins" (June, 1977). Sharon is currently studying sugars on the surface of lymphocytes as markers for lymphocyte differentiation and mat- uration, and sugars on the surface of ep- ithelial cells that act as receptors for bacteria. DONALD E. PARKER ("The Ves- tibular Apparatus") is professor of psy- chology at Miami University in Oxford, Ohio. He received his B.A. at DePauw University in 1958 and his Ph.D. in ex- perimental psychology from Princeton University in 1961. In 1961 he was a postdoctoral research fellow in the Au- ditory Research Laboratory at Prince- ton. From 1962 to 1965 he was an exper- imental psychologist in the Aerospace Medical Research Laboratories of the Wright-Patterson Air Force Base. In 1965 he joined Hermann SchBne at the Max Planck Institute for Behavioral Physiology at Seewiesen in West Ger- many to study the psychophysiology of the vestibular system on a postdoctoral fellowship. Parker joined the faculty of Miami University in 1966. In addition to his research on the vestibular system he is investigating the influence of au- dio-frequency sound, infrasound and vi- bration on perception and performance. He is particularly interested in phenom- ena associated with temporary or per- manent hearing loss. He writes: "I have a long-standing collaboration with a group of resp5ratory physiologists at the University of Kentucky concerning the perception of respiratory resistance. (How hard is it to breathe through a filter from a Coke bottle?)" Parker also writes: "My favorite entertainment is flying. I enjoy using the complex Air Traffic Control System-people, com- puters and radar-to ensure a safe flight in marginal weather conditions, particu- larly if a good ballet or opera company is performing at the destination." 4 -o.i ~ ROBERT T. SCHIMKE ("Gene Am- plification and Drug Resistance") is pro- fessor of biology at Stanford University. He received his A.B. at Stanford in 1954 and his M.D. at the Stanford School of Medicine in 1958. From 1960 to 1966 he worked at the National Institutes of Health. He joined the faculty at Stan- ford in 1966 and became chairman of the department of biology in 1978. Schimke's principal research interests have been the mechanisms of hormone action in metabolic regulation and de- velopment, and the mechanisms of the control of protein turnover in animals. FREDERIC H. CHAFFEE, JR. ("The Discovery of a Gravitational Lens"), is the resident astronomer at the Smithsonian Institution's Mount Hop- kins Observatory in the Santa Rita Mountains near Tucson, Ariz. He got his A.B. in physics at Dartmouth Col- lege in 1963 and his Ph.D. in astronomy from the University of Arizona in 1968. He then joined the staff of the Smith- sonian Astrophysical Observatory in Cambridge, Mass.; he took up residence at the obscrvatory's new site in Arizona in 1970. Chaffee's main research inter- ests are in the high-resolution spectros- copy of stars, interstellar clouds and most recently quasars. NATHAN SHARON ("Carbohy- drates") is head of the department of biophysics at the Weizmann Institute of Science in Israel. Born in Poland, he em- igrated with his family to Israel (then Palestine) in 1934. He studied at the He- brew University of Jerusalem, where he T. H. GEBALLE and J. K. HULM ("Superconductors in Electric-Power Technology") are physicists with a com- mon interest in the theme of their arti- cle: superconducting electrotechnology. Thebaby savedfrom cerebral Pbeyour~sd d Your baby picked a fine time to be born. Through research, a baby's chances of having cerebral palsy are smaller than ever. Smaller, but too big. Doc- tors find some form of cerebral palsy in 1 out of 330 newborns. This is about 10,000 babies a year. This number could be halved in five years, and halved again in another five. All it takes is the money to fully use present medical knowledge, and to speed research that doctors are sure will pay ofL Help pay for research that fights cerebral palsy. Give to your local United Cere- bral Palsy association. Think what your money buys. United Cerebral Pa1sy. P-U w a..._..i .....~.».. r r - Ured fa,btl hb, A~. Is. K[. 3" S... N.Y.. N.Y. NOIi, w re O- l.qwrr Lttw. Os^ 1f-. ILT W.Wa~w As.. A&,. N.Y.. N.Y. t2t71. 15

SCIENTIFIC Established 1845 01E RI CAN November 1980 Volume 243 Number 5 The Rise and Fall of Ischemic Heart Disease In the U.S. the death rates attributed to heart attack and other results of the obstruction of the arteries that nourish the heart ha ve fallen since the 1960's. Why they ha ve is not understood by Reuel A. Stallones W henever a drastic difference is found in the frequency of a disease from one time to an- other or between disparate cultures or in different geographic areas, a special op- portunity is created for research into the causes of the disease. Epidemiologists refer to these opportunities as natural experiments because they are untainted by the intervention of an experimenter. The recent history of heart attack and other conditions embraced by the term ischemic heart disease presents evidence that the inhabitants of the U.S. are all participants in just such an experiment, conducted on a grander scale than any experimenter could have devised. For more than 30 years ischemic heart dis- ease has been the single greatest cause of death in the U.S. Of 1.9 million deaths in 1976, for example, some 650,000, or nearly a third of the total, were attribut- ed to ischemic heart disease specifically. Without fanfare, however, the mortality rates for ischemic heart disease reached a peak in the mid-1960's and have stead- ily declined since then, for a total reduc- tion that has exceeded 25 percent The terms ischemic heart disease, cor- onary heart disease and arteriosclerotic heart disease are more or less synony- mous; they denote the clinical manifes- tations of atherosclerosis, the obstruc- tion of the fiow of blood through the arterial network, when the vessels that are obstructed are the coronary arteries, which nourish the muscle of the heart. Although the factors that determine the progression of atherosclerosis are not fully understood, the process is known to involve the growth of a mass called an atheroma in the inner layers of the arte- rial wall. In its early stages the mass con- tains smooth-muscle cells and cholester- ol; later its structure is complicated by the growth of fibrous tissue and by de- posits of calcium. Atherosclerosis may begin early in life, even in childhood. It progresses slowly over decades, thicken- ing the walls of the arteries and restrict- ing the flow of blood. Eventually an atheroma may com- pletely block a coronary artery, thereby depriving a segment of heart muscle of its supply of blood in the med- ical crisis known as a myocardial infarc- tion, or heart attack. An atheroma also promotes the clotting of blood in an ar- tery; hence a clot may produce a sudden myocardial infarction long before the atheroma is large enough to do so. The likelihood of a heart attack is therefore determined by two processes, the slow course of atherosclerosis and the rapid, unstable clotting of blood. Acute myo- cardial infarction is often fatal; in about a third of all the cases in the U.S. the afflicted person dies before receiving medical attention. Those who survive are at high risk of subsequent attacks; they may be permanently disabled, and ultimately they may die of congestive heart failure. In spite of advances both in knowledge of atherosclerosis and in medical care for the victims of ischemic heart disease, the brightest prospects for the future lie in the effective prevention of the disease. Plainly a sustained de- cline in the death rate for ischemic heart disease commands attention and calls for explanation. Mortality data for the U.S. are de- rived from death certificates that have been coded for cause in accord with the International Classification of Diseases. Data in this form are available only since 1900, and for the first three dec- ades of this century they are available only for states that had a registration method of sufficient quality to warrant admission to the system. The coding has been revised at approximately 10-year intervals to reflect the evolution of med- ical concepts of diseases and their caus- es. All these circumstances sharply limit the comparability of data over long periods of time, and so interpretations must be offered with more than the usu- al caution. The changes in medical terminology preclude the direct examination of long- term trends in the mortality from ische- mic heart disease, but the broader cate- gory "diseases of the heart" has reflect- ed the changing patterns in spite of its lack of specificity. Between 1900 and 1920 the death rates attributed to diseas- es of the heart were relatively stable. An abrupt change came in 1920, when an upward trend was established that con- tinued for 30 years. The abruptness of the change was accentuated by a deficit of deaths in 1919, the year that followed the excessive mortality of the influenza pandemic of that time. Moreover, a de- tailed analysis of heart-disease mortali- ty by Theodore D. Woolsey and Iwao M. Moriyama of the U.S. Public Health Service demonstrates that in about 1920 the admission of southern and western states to the death-registration area added young people to the population base. This too accentuated the appar- ent break in the slope of the curve. An examination of the deaths by 10- 53

e year age groups shows, however, that the change in slope was characteristic of the mid-adult years. The death rates for diseases of the heart among peo- ple younger than 35 decreased steadily throughout the first half of the century, and the death rates increased steadily for people older than 75. If the change in slope were due to vagaries in the report- ing or the classification of the deaths, then the reason for the change to be so highly selective for the age groups be- tween 35 and 75 is obscure. Terence W. Anderson of the University of Toron- to presented additional evidence that the change was not artifactual when he computed the ratios of the male to the female death rates attributed to diseases of the heart and showed that from 1920 onward the ratios increased sharply. To put it another way, the male and the female death rates diverged as they both increased between 1920 and 1950 be- cause the death rates for males grew faster. A change in the rates by gender is no more likely to be adventitious than is a change in the rates for a particular age group. Within the diversity of diseases that constitute the category of diseases of the heart, only one diagnosis, angina pecto- zoo 150 1900 1910 192o 1930 1940 1950 1900 1970 1990 TIME TREND OF US. MORTALITY dae to diseases of the heart shows a steep Increase in the death rates from the 1920's to the 1950's and then an equally steep decrease that began in the 1960's. The rates have been scaled to fit the distribution of ages In the U.BA populatioa ac- cording to the cenaus of 1940. In this way the death rates for dffiereat years can be compared in spite of the changes In age structure t® the past 80 years. T6e vertical scale is loprkhmk. 54 ris, showed an increase coincident in time with the one for the broader cate- gory. Angina pectoris means pain in the chest, and it results from the inadequate flow of blood to the heart. The pain it- self is not considered now to be a likely cause of death, but in 1912, when James B. Herrick alerted the medical profes- sion to a lethal disease marked by occlu- sion of the coronary arteries, he called the condition angina pectoris. In 1930 a new rubric, diseases of the coronary arteries, was introduced. Ap- parently it replaced angina pectoris as the designation of choice for the lethal disease, which continued to rise rapidly through the 1930's and 1940's. In 1949 diseases of the coronary arteries disap- peared from the classification system. It was absorbed into a more modern term, arteriosclerotic heart disease, which in turn gave way (in 1968) to ischemic heart disease. The time trends in this century for other forms of heart disease show no other plausible candidate to account for the overall increase in dis- eases of the heart. Ischemic heart dis- ease can therefore be traced backward and downward in a logically consistent chain to the sudden change in mortality that marked the year 1920. . . .~ ~ _ . ~ i ~ , ~ The mortality rates for ischemic heart disease reached their peak in the mid- 1960's, and by 1970 a declining trend was well established. The interpretation of the data for these years is mildly com- plicated by the 1968 revision of the sys- tem for coding the causes of death. One result of the revision was the transfer of a substantial number of deaths from hy- pertensive diseases into ischemic heart disease. Nevertheless, the decreasing trend of ischemic heart disease has con- tinued through 1978, the most recent year for which information is available. Indeed, the rate of decrease appears to have accelerated since 1973. In the 1968 revision the category of ischemic heart disease was divided into two major sub- groups, acute myocardial infarction and chronic ischemic heart disease, with re- markable results: acute myocardial in- farction has fallen since 1968, where- as chronic ischemic heart disease has shown no consistent decline. P arenthetically, the decrease in the risk of death from acute myocardi- al infarction for women is sharply at odds with the popular supposition that the redefinition of women's roles in American society (in particular their ap- pearance in large number in executive offices around the country) will result in a redistribution in their pattern of illness. Increases in peptic ulcer and in myocardial infarction are projected, and sometimes cited. Since the mortali- ty from peptic ulcer is decreasing as steeply as the mortality from myocardi- al infarction, the thesis appears to be contradicted by the observations. The natural experiment involving the incidence of ischemic heart disease in the U.S. is signaled by marked variation in the geographic distribution of the dis- ease as well as the changes over time. Before 1950 a geographic pattern for the mortality from ischemic heart dis- ease was established in which the death rates were highest along the eastern and western seaboards (particularly to- ward the Northeast) and lowest on the Great Plains. A number of investiga- tors sought to determine whether these differences were biologically genuine or were instead the result of variations in customs of diagnosis or in the availabil- ity and quality of medical care. The gen- eral conclusion was that although the data were shaky, the basic pattern could not be explained away as artifact. The geographic pattern for 1950 per- sisted through 1960, but some impor- tant trends began to emerge. The states with the highest death rates in 1950 had the smallest increase in the decade that followed. The result was a relative shift -of ischemic heart disease toward the Southeast. An analysis of the data for California by Nemat O. Borhani and H. H. Hechter of the University of Cal- ifornia at Davis shows that between

. ~• •. • ... •• •... 41 000,0000 • . , :•..'.• ••,•!...• •• ~'-74 • 1,000 I T g 100 •, , • , ~, a 10 ~ • . ~.. . •• ~ „ • ~ • 85+ , • 85+ ~ . •. .• i ; ••• .,. •,• • ,••• 0 • • i 00* "o 90 0 0 00 i • ,.0 . ~J4~t ,0600 .0~ , ~ 0 .' I , ' ~ •,, • ' •••• -~ ` •., •• +. , . I ' + ..• 00 ~ ,... ..••.•. • ••••• ~ ~-74 - i y' 00 0 • 00•..,.0 ,00 0 0 • • • •• 000 00 9 0•0 • •• 0 0 .• .• • •, ~ „• • • • • •, , . • ~ ~ . ' ,. •.~ •, •• •• ' ••••• ,• 45-54 - 45-54 •• • • • 90 , 64604 10 0 0 •0 •, • •• •• ••• • 0• • • 35-44 . • ..... ,.• • •• .•0•*110 0 • , 35-" • „• ..• • ~ ,•o, 0,~ • 0-34 '• .• ~ , • ! ( . • •,., 25-34 •., •• "00 1910 1920 1930 U.B. MORTALITY BY AGE GROUP is plotted for diseases of the reart (1900-1953), amterLoscleruttc heart diseose (194"7) and is- cbemtic beart disease (1968-77). The last two dia;nosea both signify the obstruction of the arteries that nourish the tierrt. Tkrough 1950 1940 1950 1980 1970 19®0 the death rates Increased steadily for the old and decresced steadily for the young. Ttie change In the trend of the death rates that came i. 1920 affected the people whose ages were in between. In con- trast, the recent decline in the death rates tiae affected all age gonps. 55

1950 and 1969 the death rates for is- chemic heart disease in that state actual- ly decreased by about 20 percent for white males and 25 percent for white fe- males. The peak mortality in Califor- nia came in about 1955. In retrospect, therefore, one can see that California gave the earliest mani- festation of a decline in ischemic heart disease that extended to other parts of the U.S. in the 1960's and 1970's. By 1970 ischemic heart disease was heavi- ly concentrated east of the Mississippi. This was primarily owing to the geo- graphic distribution of acute myocar- dial infarction. Chronic ischemic heart disease was dissociated from acute myo- cardial infarction in geographic pattern as well as in the national time trend. A this point some of the conundrums that have appeared repeatedly in research on the epidemiology of ische- mic heart disease should be mentioned. In a number of population-based studies the factors most consistently predictive of ischemic heart disease have been blood pressure, the blood-serum con- centration of cholesterol and cigarette smoking, together with a strong selec- 400 300 200 100 tion for males and an increased risk with increasing age. In the U.S., however, the associations with cigarette smoking, se- rum cholesterol and inale gender fade among people over 65 years old. More- over, in countries with a very low mor- tality rate from ischemic heart disease a male preponderance is not evident at any age. Related to this point are the international comparisons of morbidity and mortality directed by Ancel Keys of the University of Minnesota, which failed to show an association of ische- mic heart disease with cigarette smoking or with serum cholesterol in those com- munities where the frequency of the dis- ease was low. These epidemiological observations are consistent with the hypothesis that occlusive disease of the coronary arter- ies is resolvable into two components that are partially differentiated in the current system of disease classification as acute myocardial infarction and chronic ischemic heart disease. Accord- ing to this hypothesis, the rise and fall of the mortality for ischemic heart disease and the marked differences in the inci- dence of the disease in different regions of the U.S. are best understood as being primarily due to trends in the incidence of acute myocardial infarction, a condi- tion strongly associated with both cig- arette smoking and serum cholesterol and having a special affinity for males. Chronic ischemic heart disease may have varied to a lesser degree, but its history cannot be traced clearly through the confusion of diagnoses applied to the deaths due to chronic heart disease. T he rise and fall of ischemic heart dis- ease affords a special opportunity to link changes in the factors that appear to determine the risk of the disease in in- dividuals to the sweeping changes in the frequency of the disease in the pop- ulation at large. For any single factor to account neatly for the overall trend it should meet certain well-defined con- ditions. - If its effect was immediate, it should have appeared in about 1920, increased through the 1930's and 1940's and then begun to disappear in the mid-1960's. An immediate link is likely only if the causative factor affects the most la- bile component of the occlusive process, namely clot formation. If the causative factor is presumed to stimulate the de- I . . . • . i . ~ . . y r I , T . ~ .. f ~ I ~ ~ i 1900 1910 1920 1930 U.S. MORTALITY EY GENDER is plotted for dtreases of the heart. The death rates for males are In color and those for females are ir 1940 1950 1960 1970 1980 black. The upward trend that bepa is 1920 was markedly Zreat- er for maleL This is evidence that the increase was not fortuitous. 56

velopment of atherosclerosis, then a la- tent period of at least 20 years must be subtracted from the critical years in the trend of mortality in order to derive the trend of the causative factor. At its inception the causative factor should have affected men and women at tbout the same time, but the effect on men should have been notably greater, and it should have been greatest on peo- ple in their mid-adult years. After 1968 the death rates for acute myocardial in- farction declined in equal proportion "for men and women and also for young- er and older adults; that is, the force of mortality has been constant for these age- and gender-specific categories. The implication is plain: if the decrease in mortality in one of these groups is due to the decrease in the intensity of some sin- gle causative factor, that factor should show an equal decrease for each of the other groups. . Finally, the causative factor should have begun to decrease in California a decade before it decreased in the north- eastern states, and a decade after that it should still be strong in the southeast- ern states. If a single factor fails to meet all these criteria, some combinations of factors might be considered, although the argument must thereby be weakened and made more complex. The evidence is strong that a diet high in fat, and particularly saturated fat, in- creases the concentration of serum cho- lesterol. In the controlled environment of the hospital the intentional alteration of serum cholesterol by manipulation of the diet has been fully documented, and epidemiological observations in many environments around the world have re- lated low-fat diets to low mean choles- terol levels with comfortable consisten- cy. (Low-fat diets in those communities, however, are usually found to be linked with strenuous physical activity.) Certainly dietary customs in the U.S. have changed in important ways. In recent years saturated fats have been replaced to a considerable degree by unsaturated fats, and data are availa- . ble to support the contention that se- rum-cholesterol values for large popu- lation groups decreased from the 1960's to the 1970's. And yet if recent dietary changes have caused the mortality rates to fall, surely the deprivations of the de- pression years should have left a similar ;nark on the curve. Moreover, the con- ventratian of serum cholesterol is not ssaociated with the risk of ischemic heart disease in older people, yet people over 65 have shown the same propor- tional decrease in the death rate as younger adults. There the matter rests for now. Some evidence supports the argument that the decline in the death rates for ischemic heart disease could be due, at least in part, to decreasing cholesterol, and this could be due, at least in patrt, to changes 100 10 i ~ i I ... .. "~ ...••., DISEASES • .• ,• ' tSC~IEMIC . , : • ' HEART ~ ~T , ~: DISEASE . • . .' ARlERIOSCLERO TIC • • • ~ •• HEART DISEASE • . .. . • DISEASES • OF THE . CORONARY ARTERIES . . i . i i .•~ '••• • ~ •' ' I ~ . ~ uWiNA PECTO RIS . , • . . . . ~ . . . . . . . 1900 1910 1920 Isoo 1940 t9eo 1970 1900 SSUCCESSION OF DIAGNOSTIC TEItM$ within "djseases of the heart" includes four whose death rates (black dats) appear to form a sequence. Eegiaaing in 1920 the attribution of deaths to the Brst tern, angina pectoeis (literally "pain ha the chest"), Increased notably. The newest terse In the seqaesce is ischemic heart disease. The death rates have not been adjusted for age. 57

. in fat ingestion. Diet, however, does not provide a satisfactory explanation for the rise in ischemic heart disease in 1920. The death rates for ischemic heart disease did not track dietary changes in the ensuing 40 years, and the recent de- cline in mortality is not fully consonant with the changes in diet or in serum cho- _ lesterol. Cigarette smoking spread through 4 American culture after World War I, co- inciding with the increase in ischemic heart disease, and during the past 15 ' years or so smoking has decreased. On ~ the assumption that a major portion of ~ the harmful effects of cigarette smoking ~ has little or no latency, so that the risk of ~ ischemic heart disease rises and falls al- 3 1 most immediately with changes in ciga- rette consumption, the correlation looks quite good. T he decline in cigarette smoking has been much more pronounced, how ever, in middle-aged men than in mid- dle-aged women, a difference that is not at all in accord with the equivalence in the decline in mortality for the genders. ' Moreover, the lack of association for older people between ischemic heart dis- ease and smoking as well as serum cho- lesterol raises the question of how a re- duction in smoking could effect a de- crease in risk for this age group. Little information is available about regional variations in cigarette smoking, but a concordance between such variation and ° the geographic pattern of the mortality = from ischemic heart disease also does not seem likely. Even though high blood pressure is known to be associated with an in- " creased risk of myocardial infarction, the death rates for hypertensive disease have decreased in log-linear fashion since 1950, whereas those for ischemic heart disease increased, reached a peak and then began to decline. Programs for the detection and treatment of hyperten- sion have become far more effective in the past decade, but the forces responsi- ble for the rise and fall of ischemic heart disease must be operating quite inde- pendently, and with much greater effect. In an interval of about 20 years Amer- icans changed from a nation of walkers 58 A GEOGRAPHIC PATfERNS in the death rates for iscbemic heart disease are demon- strated by maps of the U.S. for 1950 (top), 1960 (n+iddlr) and 1%9-71 (bottom). In each map the 10 states in which the death rates for white males SS through 64 years old were geatest are shown in dark color, the 10 states in which the death rates were smallest in white. States with intermediate death rates are shown in three intermedLte shades of color. Califor- aia's drop in death rates from 1950 to 1960 siatedated by a decade the trend that spread through the nation. By 1971 lschemic heart dise.se was concentrated in the eastern U.S.

lo a culture dependent on automobiles, and the beginning of that interval coin- cides well with the rise in ischemic heart disease. Concurrently the flow of cheap fuel reduced the caloric expenditure of people at work to some small fraction of its previous value. Almost no one now spends his life at hard labor as our grandparents understood it; we are tall- er, heavier and probably stronger, but we are not inured to the daylong ex- penditure of effort in walking and work- ing that was expected of most people in the 19th century. Yet if an increasingly sedentary way of life is invoked to ac- count for the increase in ischemic heart disease, the decrease is left unexplained. The national mania for recreational jog- ging has not spread to large numbers of upper-middle-aged men and women, both black and white, whose death rates are falling as rapidly as those of younger people. Although arguments are heard about ~'1 the effectiveness of specialized treatment facilities for ischemic heart disease, it is unlikely that these treat- ments have had a major effect on popu- lation mortality data, because the mor- tality rates for a large population are quite insensitive to such influences. It is just as unlikely that the benefits of a treatment would accrue nearly equally to males and females and whites and nonwhites across a broad age span, or that an effective treatment would have been introduced in California 10 years before it appeared in states such as New York and Massachusetts. Any residual doubt on this issue could be laid to rest if the morbidity rates for ischemic heart disease in a large population were known. Stated quite simply, if the inci- dence of the disease is declining in par- allel with the mortality, the treatment of the disease is of no consequence in explaining the decline. Unfortunately, but almost predictably, the evidence is equivocaL In summary, four major variables are known to be associated with the risk of ischemic heart disease in individuals. Among the four, hypertension does not fit the trend of the mortality from ische- mic heart disease at all; physical activity fits only the rising curve, serum choles- terol fits only the falling curve and only cigarette smoking fits both. In no case is the fit as precise as one would like. This raises doubt that any of the factors is a fully satisfactory explanation for the variation in mortality. Much more could be clarified by the use of existing data. 'The nomenclature for diseases has been relatively stable since 1949, a period that includes a sig- nificant part of the increase in ischemic heart disease and all of its decline. Sam- ples of death cxrtificates for this period should be recoded according to the clas- sification system adopted in 1968 in or- t der to delineate the apparent dissocia- tion between acute myocardial infarc- tion and chronic ischemic heart disease. In the same set of data, attention to oc- cupation and other indexes of social status would provide important infor- mation, and detailed analyses of the shifting geographic patterns would be particularly useful. We now have strong assurance that programs based on our present knowl- edge can reduce the risk of death from ischemic heart disease. Unfortunately these programs require people to stop doing certain things they like to do, such 00 a* as smoking cigarettes or eating whatever they want to, or to do things they do not want to do, such as taking antihyperten- sive drugs or exercising strenuously. If the individual characteristics and habits now known to influence the frequency of the disease do not contain sufficient information to fully explain the epide- miological patterns in time and in space, and the evidence presented here sug- gests that they do not, then we may di- rect our attention to more general envi- ronmental factors in a search for more effective and more acceptable means of prevention. GEOGRAPHIC lATI'ERNS FOR TWO COA4PONENTS oi kchemic heart disease are io..n tor 1l69-71. The upper map displays the death rates for chronic lschemic hesri disease. IMe lower map displays the death rates for acute myoardial infaretion: the sudden blockate .t a coro..ry artery. Again dark color a,arla the 10 states in which death rates were greatest f.r white males from SS throoh 64, and white marla the 10 states In which rates were smallest. Axote myop,tdiv l.isrcts account for southeastern concentration of ischemic heart diseasa 59