Tobacco Documents Online

Journal of Public Healtih Policy 4/4 IDec. 1983 Research and Demonstration Projects in Community Cardiovascular Disease Prevention INTRODUCTION 9 rZG-*?,e-Z~ HERE is something new under the sun in public health: e1 community research and' demonstration projects on car- T diovascular disease prevention sponsored by the National Institutes of Health. The several U.S. undertakings have evolved separately, from widely disparate ongins, but with common conceptual bases. Each uses different pop- ulations and educational strategies and has different program emphases, but all have generally similar objectives. The history and! development of the majpr U.S. programs and the Finnish project will be touched on, along with background, design, program and evaluation in the Minnesota Heart Health Program (MHHP), funded by a National Heart, Lung and Blood Institute (NHLBI) research grant to the Division of Epidemiology, School of Public Health at Minnesota (HL 25523)• Major common objectives of these research and demonstration projects are i) to develop and test population-wide educationall programs to reduce cardiovascular morbidity and mortality rates, and 2) to assess how much disease change is attributable to risk factor and health behavior change due to the program. Emphasis is given to designand measurement to strengthen inference about program effects against a background of change in societal behavior and risk factors recognized to be ongoing in most U.S. commu- nities. These project centers seek also to advance methods in health pro- gram and! evaluation and to serve as resources to the public health commu- nity for material, training and consultation. BACgGROUiND A biologic concept behind these efforts has developed from evidence that mass adult diseases in affluent societies are the result of ubiquitous and - 398

BLACKBURN • CAR'DIOVASCUiLAR DISEASE PREVENTION 399 ~ powerful environmcntat factors acting on very wide population suscepti- bility. The powerful enviroromental factors are predominantly culturally determined, involving socially learned behaviors. The widespread popula- tion susceptibility is thought to be a human evolutionary legacy. A disso- nance is postulated between modern lifestyle and the evolutionary legacy;' that major physiologic and metabolic adaptations favored survival in a hunter-gatherer existence, while civilization, so recent in the evolutionary sense, is a stress on~those adaptations (all the while, human lifespan has been prolonged by reduced infection and trauma, with improved sanitation, education and health care). The practical corollary of these concepts, and the biologic basis for a population-wide intervention strategy, is that an unfavorable environment encourages maximal exhibition of susceptible phenotypes and a heavy disease burden-while a favorable environment assures their minimal exhibition. The rationale for primary prevention derives from the mass nature of adult diseases in affluent societies, the insidious development of the underlying processes, particularly hypertension and atherosclerosis, the high immedi- ate out-of-hospital mortality of these conditions, the inadequacy of treat- ment for the advanced stages ofarterialidisease when manifestations appear, and' the long-term excessive risk after survival of an initial episode. The potential for the primary preventive approach in cardiovascular diseases is based on the huge differences in rates found among populations and the strong ecologic and individual associations between risk factors and disease rates, along with congruence of these population observations with clinical evidence and with plausible mechanisms of cause. Further, the safety and feasibility are established of modifying cardiovascular risk char- acteristics and changing risky behavior through medical' and educational strategies. Finally, the dynamic nature of changes in reported deaths from hypertension; stroke, and coronary heart disease (CHD) in many nations, at the rate of i to zoj'~ a year and greater, both upward and downward, indicate their preventability, though none of these trends is adequately ' explained, The rationale for a population or community-wide preventive strategy, in eontrast to a high*risk or medical approach, is based~ on all the above considerations, plus the d'emonstration that entire populations and whole communities are at excess relative risk. Focus solely on the portion of highest risk people among high risk cultures appears to be a useful medical part of a community-wide prevention approach. But it is insufficient and inefficient as a sole or major public health strategy, mainly because the bulk c

4 SERUM CHOLESTEROL RELATIVE AND ATTRIBUTABLE RISK WITH CHOLESTEROL LEVEL zsSK4,Eo ( 16 ( 14 11 / 5 ~ ~ ~ / . " 7 .I .- t i-- r 12 U 6 ~ 4 ¢ ~ 0 ATTRIBUTABLE OR EXCESS CHD 50 = DEATHS/1000/10 YEARS• a (EXCESS RISK X W POPULATION EXPOSED) U , z J % EXCESS j TC %POPULATION DEATHS w ¢ a < 220 45 15 O 220-310 50 75 ¢ W > 310 5 10 N w a •ROSE:FRAMINGHAM,26:1970• U WHO EXPERT COMMITTEE REPORT, 1982 190 220 250 280 310 340 SERUM CHOLESTEROL (MG/DL) Figure i shows in graphic and tabular form the relative risk and the risk attributable to serum cholesterol values, the curved line being the Framingham logistic fit of individual serum cholesterol data and subsequent CHD risk in dcaths per thousand per year. The histogram is the simple distribution of serum cholesterol values in Framingham, in percent of its population in each class. The numbers at the celumn heads are the excess deaths attributable to the cholesterol level, obtained by multiplying the increase in relative risk times the number of people in the class exposed. The sum- mary mary table illustrates that 75% of excess or attributable deaths occurs in the class 220-310 mg/dl. E 1 . ,

BLACKBURN • CARDIOVASCULAR DISEASE PRE`'ED7'TION, 401 of attributable (excess) cases comes from the large central part of the population distribution, as illustrated and explained in Figure 1. Furthermore, concentration of preventive effort on high risk adults, through the medical care system, tends to ignore the outpouring of youth into early adulthood, bearing the physical characteristics of excess risk and already having well-developed risky behaviors. Therefore, a rational public health strategy would be directed toward all ages and segments of the community, over a sustained period-with the ultimate objective to prevent elevated risk and risky behaviors in the first place. The rationale for a multiple risk factor approa& as contrasted to single risk campaigns is its theoretically greater efficiency in lowering population risk (though the actual staging of educational campaigns to involve one risk factor at a time may be the best educational strategy). The evidence from population data about risk factors is, in my view, compatible with eithcr of two alternatives: that all three of the standard risk factors must be elevated (i.e. blood pressure, blood low density lipoprotein (LDL) and cigarette smoking) for there to be a mass population burden of atheroscle- rosis, or, that the habitual diet, which most determines elevated popul'ation levels of blood' cholesterol (LDL), is the necessary factor for mass disease. These alternatives are illustrated by the "natural experiments" of Japan, the U'.S. and Finland (i). CHD rates are exceedingly low in Japan despite levels ofblood pressure and smoking comparable to the U.S. and Northern Europe. But in Japan mass hyperlipidemia is absent. Conversely, in high CHD incidence countries, all three risk factors, and their related behaviors, co-exist and are highly prevalent. Finally, multiple strategies in community-wide prevention are needed because several behaviors determine elevated risk in the population and different population segments require specific strategies (i.e., youth, adults, seniors, industrial populations, housewives, religious and! social organiza- tions, educated, blue collar, minorities, etc.). The several educatiom in- volvement strategies are assumed to be complementary, perhaps even synergistic, i.e. more efficient applied together than individually. HISTORY The history of research and demonstration programs in CVD prevention deserves a thorough review in itself. This should await an historical per- spective from a longer experience of the several programs underway. But the concept goes back at least to the late 6os in the World Health Organi- zation (Comprehensive Community Care) and a few pretests and commu- t fY .R-

0 402 JOURNAL OF PUBLIC HEALTH POLICY • DECEMBER 1983, I~ nity projects coordinated by the WHO Cardiovascular Section in Geneva. My evaluation of those programs is that they were mainly extensions of medical care models, welU suited'to the social medical organization of the countries where they were tested, and that they emphasized control rather than primary prevention. They contained rather few elements outside the social medical system. Nor did thcy utilize, in any systematic way, existinb community leaders, social networks, mass campaigns or extensive direct edilcarion for the general population. They are not, therefore, considered here to be integrated, community-based, primary prevention activities or population-wide strategies. The story of the development of the first two major community projects in primary CVD prevention, the North Karelia Project in Eastern Fin- land (2) and the Stanford Heart Disease Prevention Program, Three Com- munity Study (3), is better left to their own historians. The fortuitous conjugation of thoughtful people on the Stanford campus is noteworthy, however; one of thern, John Farquhar, joined minds with the developer of the concept of social learning, Albert Bandura, arid the pioneer communi- cations researcher, Nathan~ Maccoby. Together, they evolved the well known Stanford community intervention concepts and the Three Com- munity Study (3) ongoing today in the much expanded Five Cities Study (4). The North Karelia Project, on the other hands was truly an internal revo- lution in community thought and action-based on an unusual external' stimultls„an international epidemiological study going on in that area, the •Seven Countries Study (1). As far as is known, the Finnish project was a happenstance and not the brainchild or deliberate instigation~ of any indi- vid". Rather it grew out of a community whichl activated: itself sponta- neously in response to awareness of its unique status: having the highest heart=attack risk worldwide. That fact was demonstrated by the Finnish investigative team under Martti Karvonen which had in the i95os joined the Minnesota Schooll of Public Health team in Physiologieal'': Hygiene under Ancel Keys. Based on early observations of the remarkable hospital prevalence of coronary disease in peaceful, rural Finland'y by Karvonen, Paul Dudley White and Ancel Keys, a sy stematic study evolved under the aegis of the Minnesota group; an NIH-sponsored project was carried out by trained international', teams in Northern and Southern~Europe, the U.S. and Japan, dating from 1957. IImmediately following, the ten-year examination of inen~ in the East Finland area, at: the end of the 196os, and following the investigators' third public report to community leaders, the latter seemed suddenly to grasp

BLACKBURN • CARDIOVASCULAR DISEASE PR'EVENTiION 403 their unique condition. In the previous decade, at the end of each, quin- quennial examination, they had accepted the same information quietly and stoicly, even sardonically,' with toasts of the "here-today, gone-tomor- row" variety! But, following the ten-year examination, there was not the same light spirit, but rather a more thoughtful, even querulous one: "Why is it so bad with us?" and then, "Why doesn't someonc help us do some- thing about it?" Professor Karvonen, the leader of the Finnish investiga- tors, wasin a particularly apt position to "get people to help do something about it," with~ his University position in Helsinki, his presidency of the Finnish Medical Association, and his advisory role in the World Health Organization. So the community activated itself, and he and the scientific community responded by assembly of a full scale WHO Expert Group, on which I served, and which heard the ideas of the local people and their University consultants. Thus, in the fall of 1971, the North Karelia Study was born. The Min- nesota Laboratory of Physiological'Hygiene shortly thereafter came under my direction with Ancel Keys' retirement in the spring of 1972. We in Minnesota had the good fortune to be exposed to the thinking of the Stanford group and to be powerfully influenced by our own direct expe- riences: the observation of large cultural differences among the Seven Countries andlour inadvertent activation of the entire North Karelia com- munity. We set about making plans for a Minnesota model of community cardiovascular disease prevention as a research undertaking. All the tradi- tion and experience of the Laboratory of Physiological Hygiene were ap- propriate. We had for three decades been a truly cross-disciplinary institu- tion, focusing varied disciplines and skills on specific major public health issues, going from~field to laboratory and back to the field, using strategies we thought appropriate to the stage of knowledge, applying methods consonant with the questions asked, and having a tradition of applied re- search employing rigorous design and method. We were not ready to eschew science for public health program. Rather, we hoped to put more science into preventive efforts and to provide the evidentiall base for public health policy and program in this field. WHAT' ARE RESEARCH AND DEMONSTRATION PROJECTS IN PREVENTION? What they are not is formal hypothesis-testing, in the experimental! sense, about the effects of risk-factor lowering: they are research demonstrations. We consider them public health trials. We contendl that, with careful at~ I O W ~ . a 1w-

F NHLBI Focus 0 THE NATIONAL HEART, LUNG AND BLOOD INSTITUTE RESEARCH SPECTRUM "THE LEVY ARROW" Idea ti Idea Idea G " nera on e Communication Utilization and Idea Development Diffusion Into Practice Basic and Clinical Research Applied Research and Development Demonstration and Education Programs - Knowledge Knowledge Transfer Knowledge Validation Acquisition ~ FIGURE 2 Hea Ith Practice sssVUEo

C BLACKBURN • CARDIOVABCULAR' DISEASE PREVENTION 405 tention to design, inference can be reasonably drawn about effectiveness of a given educational strategy and; program. A body of literature dealing with such community experiments has rarely been considered in medical, preventive or public health undertakings (5). It seems to us that failure to use these logical methods limits medical science and the public health, especially when the perfect experiment in~ health behavior of populations is infeasible. In the absenee of a substantial number ofrandolrlized experimental units, what are the characteristics of d'esign and evaluation that strengthen causal' inference about program~ efforts? (Andl what are the concepts and! methods which have led several cardiovascular disease investigators to give such large parts of their professional careers to such ambitious undertakings?) Certainly the rand'omized clinical trial is one of the more powerful tools in all science. It would not be totally out of the question: to have a sizablee number ofrandomizedlunits for future community education experiments. But within the NIH resear& review and funding system, this amount of central planning, collaborative proposals and orderly operation of a multi- center effort did not seem to be in the cards, either for NIH administrators or for the investigators. Thus, each devcloped its own unique model and submitted investigator-initiated proposals for research-at Stanfor& (3,4), Minnesota (6), Rhode Island (7)s andl Pennsyl1vania (8), three of which were eventually NIH-funded. THE' INSTITUTIONAL NICHE Figure Z has become popularly known as "The Levy Arrow" because it integrates his thinking and appeared in Forward Plans for NHLBI during his tenure. The arrow shows demonstration~ and'~ education programs as a natural part of the NIH research contitiuum, placed between the discovery of new knowledge and applied research on~one hand and health action on the other. The diagram does not insist that this step is always an essential one in the transposition of knowledge to practice, but that it may be a logical and desirable step under certain conditions. Dr. Levy developed this broad view of the continuum of NIH researches in consultation with those in epidemiology and public health fields, and in response to the demands of Congress for "technology transfer." He became convinced of the useful, probably even necessary, role of the National Institutes of Health in the transposition~ of knowledge and, more particularly, in research on the transposition of knowledge. He came to defend this view against strong elements in the scientif c community opposed to such use of research funds, .,-

. C C 406 JOURNAL OF PUBLIC HEALTH POLICY • DECEMBER 1983 or to the NIH serving as a base for such activities. Dr. Levy and his sup- porters not only felt that the research and demonstration projects in pre- vention were a logical part of the NIH research continuum, but that some of the problems of basic science in general, and of NIH inparticular, lay in a negative public image of "valuc-frce science" as well as the failure of scientists to take responsibility for guiding the outcomes and applications of new knowledge. One probable result had been a dwindling of the base of scientific investigation and support from its rapid growth in the sos and its heights in the 6os. Thus, applied and translation activities were entered into the NHLBI plan in a modest way, predominantly in descriptive epi- demiological research and! trials, but also in program offices for Health Promotion and for National High Blood Pressure Education. All this was done at a time when the leadership of NIH did not support such applied! activities in public health, rarely testified on public health matters, and, in fact, considered the recent progress of medical science entirely within the increased understanding of DNA, heralding the "baroque beauty of biol- ogy" and marveling at the proliferation of hundreds of newly identified inborn errors of metabolism, all without mention of any interaction of intrinsic determinants with the powerful socioeconomic, cultural, behav- ioral, and environmental factors in' mass disease! Thus, the development of the institutional! base for research and demon- stration in prevention can be attributed to the courage and foresight of NHLBI where leadership surpassed that of the other institutes in~ applied, translation and prevention activities. One can only speculate whether a modest provision, say 5%,, of the National Cancer Institute's budget to such activities over the last decade would have put it in a much stronger position now to d'efend its important bench research and its unending search for ever more active anti-cancer drugs. PEER REVIEW Competent peer review in this field''I of research~ and demonstration has been a problem. At NIH it has improved significantly. The Division of Research Grants jealously guards its independence in determining peer re- view. The three major funded U.S. research and demonstration studies would be ungracious indeed! to criticize a: system that recognized eventu- ally the merit of their research propositions. But several important issues in the process arose and still need consideration by the institutes, by the scientific community whic6 serves them as reviewers, and by the public .ft-

BLACKBURN • CARDIOVASCULAR DISEASE PREVENTION 407 K health community now so active in prevention and health promotion legislation (9,10). First, there appears to be a "natural lag time" between innovation and the development' of scientific peer understanding. Without gambling on innovation, there would be nothing, or much less, new. However, by definition, "what is established is best." The "establishment" fuels inertia and'! delays change. On the other hand, if every innovative, untested idea were funded, the world! would soon go broke. But if basic science and its traditional research establishment determined everything (i.e., controlled all review and funding allocations); there would surely be no innovation at all in applications research. It seems that there must be an institutional'! "will'uigness" to set aside, by formal planning, some proportion of re- sources for well-designed research in applications and'. technology transfer. Presumably this is now in place at NHLBI. The Levy Arrow has the support of the new director, Claude Lenfant, who intends to maintain the broad research program of NHLBI, and the N'ational Cancer Institute finallyhas an active office for prevention activities, the Division of Cancer Cause and Prevention. . Second, peer review in: this field need's to be informed andl confident enoug4 to recognize good work and proposals and not to exclude such activity out of ignorance or bias against applied' researches. Herein lies another problem in~ the system. Because of the magnitude and complexity of community research and~ demonstration projects, the Division of Re- search~Grants still tends to treat them as traditional "program projects" for which "there must be at least one expert" to review each, component of the endeavor (i.e., cardiology, nutrition, physiology,,epidemiology, biom- etry, sociology, communications, psychology, education, administration, etc., etc. !!). Thus, instcad of an investigator having the broadest view of, say, nutrition and its public health applications, or a physician having experience in prevention and public health eff'orts outside the medicaU care setting, the early reviewing experts were, more often than not, specialists from laboratory and' clinical disciplines interested in theory and method, but whose views and! experiences may not have extended much beyond their sub-subspecialty. Siniilarly, for those from the social: sciences, which have so many traditions, "schools," and versions of psychology and soci- ology, the scope, directions„cotnpleYity and cost of such community-wide research strategies may have been so confusing aibd intimidating that they would exercise quietly their anonymous weapon: a low priority score. Third, the issue of actual numbers of reviewers on any large NIH grant 4 lw-