Disease prevention is an essential component of public health, which has been defined as ‘‘the science and art of preventing disease, promoting health, and prolonging life through organized efforts of society’’ (Committee of Inquiry into the Future Development of the Public Health Function, 1988: 1). Its history stretches far back into classical Greek mythology, which already distinguished between treatment and prevention, with the two daughters of Asklepios, Panacea and Hygeia. Panacea represented treatment and Hygeia embodied living wisely and preserving health.
In the more recent past, the major gains in health were brought about through actions that could be described as preventive. The decline in many infectious diseases in industrialized countries since the nineteenth century can be attributed, to a considerable extent, to the actions of public health reformers, such as Chadwick in England, Virchow in Germany, and Pasteur in France. The leading causes of disease have changed now but prevention has continued to be important. As Geoffrey Rose (1992) noted, ‘‘the scale and pattern of disease reflect the way that people live and their social, economic and environmental circumstances, and all of these can change quickly. This implies that most diseases are, in theory, preventable’’ (Rose, 1992: 1).
The threats to health are extremely diverse; however, the scope for prevention can be illustrated by a concept that has its roots in the germ theory that evolved in the nineteenth century. This theory sees the emergence of disease as a function of four linked factors: the agent, the host, the vector, and the environment. The agent is defined as the factor whose presence is necessary for the occurrence of disease, for example, a microorganism such as Mycobacterium tuberculosis causing tuberculosis, a chemical such as the components of tobacco and tobacco smoke, or the presence or absence of a dietary factor such as salt. The host is defined as the individual who uses the factor or is exposed to it, for instance, the respiratory tract as a way of entry of M. tuberculosis or the smoker using tobacco. Often, the transmission of an agent to the host is facilitated by a vector, for example, the malaria-infected mosquito or the tobacco manufacturer who produces and distributes the agent (i.e., tobacco). The environment traditionally refers to the physical, chemical, and biological components that impact on health, with cultural, historical, economic, and political factors increasingly being recognized as important in the web of disease causation. In the tobacco example, an important element includes the powerful effects of pro-tobacco forces on political decision making that may result in reduced support for tobacco control activities.
While this model of disease causation greatly simplifies what for many diseases is a highly complex interrelationship, it proves useful to classify approaches in prevention as is shown here.
Classification of Disease Prevention
Traditionally, prevention has been classified into three types or levels, corresponding to different phases in the development of disease: primary, secondary, and tertiary prevention. A fourth level, primordial prevention, was the most recently defined (Table 1).
Table 1. Levels of prevention
Adapted from Beaglehole R, Bonita R and Kjellstrom T (1993) Basic Epidemiology, Geneva: WHO.
Primary prevention comprises measures that prevent the occurrence of a disease by individual and collective efforts such as improved nutrition, immunization, and eliminating environmental risks. In epidemiological terms, primary prevention seeks to lower the incidence of disease and other departures from good health by controlling causes and risk factors. This is the task of public health.
The decline of infectious diseases in many industrialized countries since the turn of the nineteenth century was linked to public health improvements such as the provision of clean drinking water, safe sewage disposal, and improved food safety. In terms of the model of disease causation introduced earlier, these measures aimed at reducing or eliminating the causative agent from the environment and so reduce exposure.
In contrast, preventive interventions directed at the host seek to increase resistance to the agent, mainly through immunization, or to reduce risk of exposure. Systematic immunization has led to further reduction or elimination of many infectious diseases during the twentieth century; the success of strategic campaigns in the United States and Europe since the 1950s stimulated the concept of disease eradication. Smallpox was the first (and so far only) disease to be eradicated globally in 1980, after more than 10 years of campaigning that involved systematic immunization of entire populations in endemic countries. The Global Polio Eradication Initiative was launched in 1988 by the 41st World Health Assembly. However, despite considerable progress, eradication has proved difficult to achieve and by the end of 2006 polio was still endemic in four countries (Global Polio Eradication Initiative, 2007).
Other preventive measures targeting the host include chemopropylaxis to prevent infection or its progression to disease (e.g., by using insect repellent to avoid mosquito bites and thus transmission of malaria or yellow fever), and behavioral changes, to reduce or eliminate risk of exposure to the agent. Thus, using condoms will markedly reduce transmission of HIV or other sexually transmitted diseases (STDs).
Another entry point for primary preventive action to control infectious disease is directed at the vector as the main route of transmission for many infectious agents. Means to control the vector directly include: chemical measures such as spraying of insecticides; environmental actions such as elimination of breeding sites of mosquito larvae; and biological measures, for example, the introduction of predators and parasites to control the vector population. A range of measures is directed at the infectious agent itself so as to remove it from the environment or to inactivate it through interventions such as cleaning, cooling, disinfection, and sterilization. In practice, the design of suitable programs will seek to provide the most effective mix of different control strategies that target the host, the agent, and routes of transmission for infectious disease control.
The model of disease causation is also useful to illustrate strategies of primary prevention of NCDs, injuries, and risk factors. Tobacco use is an example of an important risk factor for lung and many other cancers and cardiovascular disease, and interventions mostly target the host – that is, the smoker – and the environment and, more recently, also the vector, that is, the tobacco manufacturers (Asma et al., 2002). Strategies to influence the host usually comprise educational programs to raise awareness and information levels about the dangers of smoking, or means to help smokers quit such as smoking cessation programs. Interventions directed at the environment involve mass media campaigns, attempts to promote policy change, and regulatory efforts to reduce exposure to secondhand smoke through smoking bans in public places and/or at the workplace. Meanwhile, tobacco manufacturers have undertaken a variety of measures to change the actual agent, for instance, by producing low-tar cigarettes, although mostly for presentational purposes rather than any desire to reduce health risks. Regulatory efforts strive to control both the agent and the activities of the vector (the tobacco industry). Recently, increasingly effective action has been taken – especially in the United States – targeting the vector through litigation against the tobacco industry.
Another useful example of primary preventive action is to reduce the incidence of artherosclerosis. Similar to the prevention of infectious disease or tobacco use, interventions will need to address the different levels in a continuum that extends from the individual to the community, national, and possibly international level. Interventions directed at the host will consider educational strategies and/or dietary counseling for those at increased risk because of elevated cholesterol levels. Measures targeting the environment may include mass media interventions and measures to improve the supply of healthy food. The agent, that is, dietary fat composition, has been addressed by the food industry, with some sectors promoting an increasing number of foodstuffs with lower fat contents, especially dairy and meat products.
In the case of injuries – here using the example of a car crash – primary prevention strategies will again act at different levels. Actions directed at the host, namely the driver of the car, comprise measures such as seat-belt use and educational strategies to enhance knowledge about safe driving. Interventions aimed at the environment would include improvements in transport infrastructure and enforcement of safety controls. The actual agent – the collision, or more precisely, mechanical energy – may be influenced by lower speed limits to reduce energy associated with the collision. Finally, a variety of measures are directed at the vector, here the car, through vehicle design such as the introduction of air bags.
In general, primary prevention may appear very similar to primordial prevention (discussed at the end of this section), or health promotion. The main difference is that primary preventive efforts aim at reducing specific causative ‘agents,’ whereas primordial prevention takes a broader perspective, directed at establishing health-promoting living conditions.
Secondary prevention is directed at diseases in which there are measurable risk factors or an abnormal condition that predate the emergence of disease. It aims to intervene before the disease arises either by reducing the risk factors or treating the underlying abnormality. This is the task of preventive medicine.
Secondary prevention can be applied only when the natural history of a given disease includes an early period when it is easily identified and treated, thereby allowing for an interruption of the progression to a more serious stage. Appropriate secondary prevention strategies require safe and accurate methods of disease detection, preferably at a preclinical stage, and effective methods of intervention.
Important examples include screening for the early detection of cervical cancer and breast cancer. While the efficacy, specifically of breast screening programs based on mammography is still being debated, cumulating evidence suggests that cytological screening for cervical cancer is associated with a decline in associated mortality. With advances in medical technology, screening tests for different cancer sites have become available and are being discussed for implementation at the population level such as screening for colorectal cancer.
Other forms of screening measures include programs for the early detection of congenital disorders in newborn children such as phenylketonuria (PKU) or galactosemia. PKU can be identified by measuring the level of phenylalanine in neonatal blood and thus identified children, when treated appropriately (i.e., with special diet) will develop normally.
Further examples of secondary prevention include measures to prevent the spread of infectious diseases by tracing, and subsequently treating, contacts; antiretroviral prophylaxis of perinatal transmission of HIV-infected pregnant women and their babies; identification of hypertension to lower the risk of stroke; prophylactic treatment of asthma; provision of antenatal care to reduce complications in pregnancy; and early treatment and monitoring of diabetes. Measures of secondary prevention may also include interventions to promote behavior to seek health care early.
Tertiary prevention seeks to reduce the impact of established disease by eliminating or reducing disability, minimizing suffering, and maximizing potential years of quality life. In epidemiological terms, tertiary prevention aims to reduce the number and/or impact of complications. This is the task of therapy and rehabilitation.
Strategies for tertiary prevention include screening of patients with diabetes for diabetic retinopathy to prevent progression to blindness through prompt treatment; prevention of opportunistic infections in HIV patients; provision of prostheses and medical devices to enable persons to take part in social life; follow-up of chronically ill patients to ensure adherence to medication regimen, monitor changes, and assist them in maintaining independence in daily life; rehabilitation of stroke patients to restore functions, such as through physiotherapy, and prevent recurrence or further complications through appropriate medication.
Tertiary prevention is often difficult to separate from treatment. Several authors refer to the treatment of, for example, symptomatic cancer or surgical interventions for acute conditions such as appendicitis as tertiary prevention because these interventions seek to prevent recurrence of disease and, ultimately, reduce case fatality.
Primordial prevention aims to inhibit the emergence and establishment of environmental, economic, social, and cultural determinants of lifestyles that are known to increase the risk of disease. Recognizing that individuals face many constraints on their freedom of action, it is achieved primarily through intersectoral action that enables people to make healthy choices.
The term was originally introduced in a 1982 report by the World Health Organization (WHO) on coronary heart disease, in recognition of accumulating knowledge about the epidemiology of cardiovascular diseases (CVD) and the potential role of social and cultural factors in the development of CVD and NCDs in general. The importance of primordial prevention has assumed greater urgency because of the increasing pace and scale of societal change and processes of globalization such as the expansion of international travel and trade, and global environmental changes.
Developing countries are increasingly affected by NCDs. This is all the more critical as many of these countries have not yet fully controlled communicable diseases, and their health systems are ill prepared to provide the costly care required for diseases such as CVD. Yet, urbanization and industrialization increase the exposure to unhealthy lifestyles, elevating levels of risk factors for NCD. A comprehensive long-term strategy to control NCD should thus aim to prevent the emergence of these risk factors in the first place.
The taxonomy of prevention provided here is based on Beaglehole et al. (1993) and Last (2001). A clear distinction between different levels is however often difficult to determine as this may vary according to the specific aims of the preventive measures. For example, dietary recommendations on saturated fat intake may be viewed as primary prevention of artherosclerosis. They can also be seen as secondary prevention if targeted at reducing the risk of coronary heart disease in patients with subclinical artherosclerosis. Also, the taxonomy may vary according to the discipline that defines the different levels, for instance, public health or clinical medicine. The distinction between primordial prevention (health promotion) and primary prevention is often not clear-cut mainly because contents and strategies frequently overlap. However, as Last (1995) has pointed out, the differences in opinions as to the precise boundaries between the various levels are semantic rather than substantive.
Disease Prevention Strategies
Epidemiological research into the determinants of disease has uncovered a wide range of factors suggested to play a causal role in the etiology of disease, such as tobacco use and lung cancer, or high blood pressure and stroke. Consequently, individuals showing these factors are considered to be at risk for a particular disease. This is usually expressed as a relative risk, that is, the risk of occurrence of a disease among exposed individuals relative to that among the unexposed, and it might explain why some individuals become ill while others stay healthy. Yet it provides only a partial answer to the question of why a certain disease is more common in one population than another. This is because the determinants of individual differences of characteristics within a population are not necessarily the same as the determinants of differences between populations.
Geoffrey Rose (1985) identified two basic approaches to prevention, the high-risk strategy, which seeks to identify individuals at high risk and to offer them some individual protection, and the population strategy, which seeks to control the determinants of incidence in a population as a whole.
The High-Risk Strategy of Prevention
The high-risk strategy of prevention aims at individuals at high risk for disease. It follows a medical approach and is considered appropriate to the individual, for instance, as for an individual with hypertension it would seem appropriate to advise salt restriction or commence pharmacological treatment. Persons identified as ‘at risk’ are likely to adhere to interventions to reduce the risk for the development of more serious disease (subject motivation). Physician motivation is likely to be high also, as the intervention seems to be justified on the grounds of proven physiological deviation from the norm (i.e., high blood pressure). The benefits of the intervention are likely to outweigh the risks or adverse effects. For example, in the case of the hypertensive patient the risk of possible side effects of pharmacological treatment may be regarded as low compared to the risk of developing a more serious disease such as stroke. The high-risk approach is usually also considered to make cost-effective use of limited resources, as it concentrates on cases where the need, and thus the potential to benefit, is likely to be greatest.
However, relying on screening measures to identify individuals at risk is problematic and often involves high costs. Also, the high-risk approach does not seek to alter the underlying causes of the disease as it identifies and protects those individuals who are susceptible only. This approach is thus palliative, temporary and local and not radical as Rose (1985) pointed out. The potential to benefit is usually quite limited for the population in particular, mainly because the ability to predict future disease is weak: most individuals at risk will remain well for some time while others considered at low risk may experience unexpected illness. High-risk individuals might contribute little to the overall burden of disease in a population while high numbers of individuals at low risk may produce more cases of disease. Furthermore, the high-risk approach normally does not take into account the circumstances of the individual at risk and is thus behaviorally inappropriate: many people may find it very difficult to individually alter lifestyle habits that are largely socially determined. Finally, although the high-risk approach may be cost-effective in the use of resources, the widespread introduction of screening programs to identify high-risk individuals pose substantial practical problems as they relate to feasibility and costs.
The Population Strategy of Prevention
In contrast, the population strategy for prevention seeks to control the determinants of disease, to lower the mean level of risk factors and thereby to reduce average risk in the population. Classic examples of this approach include environmental control measures such as clean water supply and sanitation mentioned earlier. It is based on the ‘‘recognition that the occurrence of common diseases and exposures reflects the behavior and circumstances of society as a whole’’ (Rose, 1992: 95). Its main advantages over the high-risk strategy are that it is radical, as it aims to remove the underlying causes that make a disease common, for instance, by addressing (societal) barriers that prevent people from adopting healthier behaviors. It also holds large potential for the population: It was estimated, for instance, that a reduction of 2% of mean (diastolic) blood pressure could potentially avert almost one million deaths from CVD per year in the eastern Asian region by 2020 (Rodgers et al., 2000). This equates to approximately 15% of all stroke and 6% of coronary heart disease deaths in that region. The population strategy is also considered to be behaviorally appropriate as it strives toward a general change in behavioral norms, thereby facilitating individual efforts to adopt healthier behaviors: smokers are more likely to quit smoking if public attitudes favor nonsmoking. Also, efforts to increase physical activity in a population may be more successful in a social climate that encourages this behavior through, for example, providing easy and affordable access to exercise facilities.
However, the population approach offers little benefit to each individual within a given population because their absolute risks for disease are relatively low (i.e., the ‘prevention paradox’; see Rose, 1981). The use of seat belts illustrates this paradox: To prevent one death resulting from a car crash many hundreds of people must wear seat belts. Most individuals will thus be wearing seat belts while driving without being involved in a car crash throughout their entire (driving) life. Also, to reduce the death rate from coronary heart disease many people will have to lower their saturated fat intake by, for example, using soft margarine instead of butter. For the single individual this change will make only a small difference to his or her health prospects, at least in the foreseeable future. Thus, in contrast to the high-risk strategy, motivation of the individual may be low, as few people may be willing to alter their behavior without prospects of a direct and visible health benefit. Also, the population approach offers only poor motivation for physicians. And not only is concern for future health a poor motivator for action by individuals and physicians, it usually ranks even lower in the priorities of governments. Yet, without their support the implementation of this strategy will be limited (Rose, 1992).
Finally, the benefit-to-risk ratio of the population approach may be lower than expected. This is in some way the reversal of the argument against the high-risk approach that a large number of individuals at small risk may contribute more to the total burden of disease in a population than a small number at high risk. If a preventive measure exposes many people to a small risk then the harm it does may outweigh the potential benefits. This possibility arises when the association between exposure and risk is U- or J-shaped, as in the case of alcohol consumption and coronary heart disease. Removal or reduction of the exposure, here alcohol consumption, may remove a protective factor against heart disease. The likelihood of this actually happening is, on the basis of current research on the relationship between alcohol and heart disease, not certain. However, the analogy implies that any population-wide strategy that seeks to shift the distribution of a particular risk variable must ensure that such measure is safe. The uncertainty will be low in strategies that aim to remove an abnormal factor and restore ‘biological normality’, such as giving up smoking, avoiding obesity, or reducing intake of saturated fat, or environmental control mechanisms such as reducing exposure to sulfur dioxide or providing clean drinking water (Rose, 1981). Such strategies will presumably be relatively safe. In contrast, approaches to prevention that do not intend to address the underlying causative agent but impose some ‘new,’ putative protective intervention such as immunization, long-term medication, or excessive exercise will need to ensure sufficient benefit and safety of the measure to be adopted.
From a public health perspective, the population approach to prevention has the intuitive appeal of appropriately improving population health, and Rose noted that ‘‘the primary determinants of disease are mainly economic and social, and therefore its remedies must also be economic and social’’ (Rose, 1992: 129). In reality, however, in the field of NCDs, the high-risk approach has dominated preventive efforts in the developed world during the past 20 or so years (Beaglehole, 2001). Only recently, with the recognition of a rapid rise of NCD, especially in low- and middle-income countries, has it regained considerable attention (WHO, 2000).
Intervention Strategies for Prevention: Screening
Strategies at different levels of prevention comprise a wide range of measures, aimed at the host and/or the causative agent, each having an important role within the spectrum of disease prevention. This section examines screening.
Screening can be defined as ‘‘the presumptive identification of unrecognized disease or defect by the application of tests, examinations or other procedures which can be applied rapidly’’ (U.S. Commission of Chronic Illness, 1951, as quoted in Wilson and Jungner, 1968: 11). It is concerned with individuals who have not sought medical attention for that specific disease or defect and it is targeted at apparently well people. The initiative for screening usually originates from an investigator or health-care provider and thus implicitly promises that those who are screened will benefit, although screening does normally not provide a definitive diagnosis and needs to be confirmed by special diagnostic procedures. Mostly concerned with chronic illness, screening aims to detect disease not yet under medical care (Wilson and Jungner, 1968). It distinguishes between those with a normal test result and thus a low probability of disease who do not receive further special attention and those with abnormal test results. The ultimate objective is to reduce morbidity or mortality from the condition among those screened by early and appropriate treatment of causes discovered.
Screening may identify risk factors, genetic predisposition, and precursors, or early evidence of disease. It therefore does not necessarily involve high technology equipment or procedures such as radiography or laboratory tests. Simple screening tests include, for instance, blood pressure measurements or specific screening questionnaires such as AUDIT (Alcohol Use Disorders Identification Test), which was developed by the WHO to identify persons with hazardous and harmful drinking behaviors. Boundaries between screening for selected risk factors and for disease are, however, not clear-cut. For example, hypertension can be interpreted as a risk factor for cardiovascular disease and as an early manifestation of CVD.
The availability of a test that allows a disease to be detected in its preclinical stage does not necessarily imply that population-wide screening should be undertaken. This should be governed by principle guidelines or criteria, relating to the characteristics of the disease, the screening test itself and program policy. Table 2 shows some of the main criteria identified by Wilson and Jungner (1968) and others.
Table 2. Criteria for instituting a screening program
Adapted from Wilson JMG and Jungner G (1968) Principles and Practice of Screening for Disease. Public Health Papers, vol. 24, Geneva, Switzerland: World Health Organization.
Thus, the disease to be targeted should be an important (public) health problem and clearly defined. This may include common and, at the societal level, resource-intensive conditions such as hypertension. It may also include diseases that are relatively rare but may have serious (individual) consequences if not diagnosed early such as phenylketonuria (PKU.) To be suitable for early detection and treatment the disease must have a preclinical stage with a reasonable duration before the manifestation of symptoms, and early treatment must provide some advantage over later treatment (e.g., improve survival or quality of life). This requires that recognized and effective treatments as well as facilities for service delivery are available. Thus, the benefit of early detection of diseases where no effective treatment exists may be questionable, as, for instance, in the case of Alzheimer’s disease. Also, there is little benefit in screening programs uncovering large numbers of individuals with a particular condition when there is no integrated system for further follow-up investigation and treatment of newly identified patients.
The screening test itself should be acceptable and safe to both the individual to be screened and the person who administers the test. It should be easy to apply at relatively low cost, and provide valid and reliable results as measured by its sensitivity, that is, the ability to detect all those with the disease in the screened population (true positive), and specificity, that is, the ability to correctly identify those who do not have the disease (true negative). Ideally, a test should have both high sensitivity and high specificity. This, however, requires a clear definition as to what constitutes a disease, which may not be easy to determine as disease is normally a continuous process and definition of the cut-off point between normal and abnormal is often arbitrary. Higher sensitivity will inevitably lower specificity, that is, the number of screened persons with a false-positive result will rise. This may be acceptable for diseases with serious consequences (e.g., PKU in newborn children), but will be questionable when increasing sensitivity identifies relatively more cases of ‘abnormality’ that would not have presented clinically otherwise or, if they had, would have been cured anyway (e.g., a benign cancer). The decision as to whether to introduce a specific screening program will thus need to carefully balance issues of sensitivity and specificity, depending on the individual and societal implications of identifying false negatives and false positives. It will also have to take into account the associated costs, namely whether the benefits of screening outweigh the consequences of not doing so. It may be more effective to include only those at risk for the disease rather than the whole population. One common criterion is age as an important predictor of risk. For example, routine screening for breast cancer in industrialized countries is usually targeted at women aged 50 and over with screening of younger women considered not to be cost-effective.
Ultimately, the value of a screening program has to be judged against its impact on population health. This will largely depend on how the actual screening program is organized. For instance, the program can be proactive (or organized) and a defined population (e.g., community, factory workers) is offered screening by invitation, or it can be restricted to patients who consult a health practitioner usually for reasons other than screening (opportunistic). These different approaches are likely to have implications for the control of the disease at the population level, mainly because of differences in cost-effectiveness of the program and in uptake, which, in opportunistic screening programs, largely depends on the willingness of individuals to volunteer for screening.
Diseases and Injuries Prevention
To illustrate the challenges related to disease prevention, this section discusses specific health problems that have been the focus of prevention strategies, using examples of NCDs and injuries and drawing on both successes and failures of different preventive approaches.
Prevention through Community Programs: Coronary Heart Disease
CVD is a major contributor to the global burden of disease, with coronary heart disease and stroke alone accounting for an estimated one-fifth of total mortality and morbidity worldwide (Lopez et al., 2006). The proximal causes of CVD are well known, with epidemiological research having identified smoking, elevated serum cholesterol, and high blood pressure as among the major risk factors. These factors relate closely to individual behaviors and lifestyles, suggesting that curtailing the risk of heart disease and stroke requires modifying individual lifestyles. At the same time, controlling blood pressure and cholesterol with pharmacological treatment also reduces CVD risk. Applying this knowledge to the community setting would thus appear to be the most promising strategy to prevent CVD.
Following this line of reasoning, several community-based intervention programs were set up in the 1970s and 1980s. They were based on the premise that targeting the entire community to modify risk factor levels and lifestyles would have a higher public health impact than targeting individuals at high risk only. Programs thus introduced included the North Karelia Project in Finland (1972), the Stanford Three-Community Study (1972) and the Stanford Five-City Project (1978), the Minnesota Heart Health Program (1980) and the Pawtucket Heart Health Program (1980) in the United States, to name only the most influential programs. This section describes the North Karelia Project as an early attempt to organize CVD interventions at the population level.
The North Karelia Project originated from a local population’s petition for national aid to reduce the high burden of CVD mortality in North Karelia, in eastern Finland, and was launched in 1972 (Puska et al., 1985). Its main objective was to reduce cardiovascular morbidity and mortality, with special emphasis on middle-aged men who had particularly high rates. Based on the knowledge about the natural history of CVD, the program sought to reduce cigarette smoking, serum cholesterol levels, and blood pressure in the community (primary prevention) and to promote early detection, treatment, and rehabilitation of heart disease patients (secondary prevention). This was to be achieved through mobilizing community resources with full involvement and participation of the local population, and providing for continuous follow-up with feedback to the community using a wide range of prevention activities as shown in Table 3.
Table 3. Intervention strategies in the North Karelia Project
Adapted from Puska P, Nissinen A, Tuomilehto J, et al. (1985) The community-based strategy to prevent coronary heart disease: Conclusions from the ten years from the North Karella Project. Annual Reviews in Public Health 6: 147–193.
The program was initially designed for a limited period only (1972–77) but was gradually expanded to include broader objectives of integrated prevention of major NCDs and health promotion. Eventually it influenced the entire country and international programs such as the Country-wide Integrated Non-Communicable Disease Interventions (CINDI) program by the WHO.
The impact of the North Karelia Project on the populations’ health was assessed using repeated cross-sectional health surveys. These showed that in the first 5 years risk factor levels among men aged 30–59 fell more rapidly in North Karelia than in the reference region, the Finnish county Kuopio, with further progress observed thereafter. Mortality from coronary heart disease among middle-aged men also fell, and this decline was significantly greater than in other areas of Finland; national coronary heart disease rates did not decline before the end of the 1970s. Much of the observed fall in coronary heart disease mortality was attributed to risk factor changes and evidence suggests that changes in diet may have been the most important determinant.
Other community intervention programs such as the Stanford Five-City Project have produced mixed findings; however, the North Karelia Project has illustrated the potential of the community approach to contribute to favorable changes that are ultimately required for sustained long-term improvement in population health.
Prevention by Confronting Vested Interests: Tobacco Control in California
Tobacco use is the single most important preventable risk factor for cardiovascular disease and most cancers. It was estimated that, with current smoking patterns, global deaths attributable to tobacco use are likely to rise from 5.4 million in 2005 to over 8 million in 2030 (Mathers and Loncar, 2006). The epidemic is increasingly affecting low- and middle-income countries where most of the world’s smokers live. Accumulating scientific evidence has demonstrated that exposure to tobacco smoke also has harmful effects on the health of those who do not smoke. Thus, preventing the uptake of tobacco use among young people, reducing its use in young people and adults, and eliminating nonsmokers exposure to secondhand smoke are key public health objectives. However, it must be recognized that there are strong forces promoting exposures that are hazardous to health.
Approaches to reducing tobacco use include educational (public health information), regulatory (advertising and promotion, product regulation, access, smuggling, clean indoor air regulation), economic (taxation), and health services (promote smoking cessation) strategies. Available evidence suggests that a combination of strategies will be most effective in reducing tobacco use. This section looks at the California Tobacco Control Program in the United States as an example of a comprehensive program to curtail smoking.
In 1988, the California Tobacco Tax and Health Protection Act (Proposition 99) increased the tobacco excise tax by 25 cents per pack of cigarettes. It also earmarked 5 cents per pack for an educational campaign (the Health Education Account) which led to the California Tobacco Control Program (CTCP) in 1989. Drawing in part on community intervention programs such as the Stanford Five-Cities Project mentioned previously it employed multiple interventions that targeted individual, social, and environmental factors through community and school programs, and a statewide media and public relations campaign. Examples include smoking restrictions or bans at the workplace and in public places as well as antismoking television advertisements. From 1993 tobacco control efforts within the CTCP increasingly focused on three priority areas: (1) to reduce exposure to secondhand smoke, (2) to reduce youth access to tobacco, and (3) countering pro-tobacco influences in the community. The ultimate goal was to promote social norms that would make tobacco use and secondhand smoke unacceptable in the community.
Analyses of the impact of the CTCP on tobacco consumption suggest that it accelerated the decline in cigarette smoking in California. Thus, between 1989 and 1993, adult smoking prevalence fell from 23.3 to 18% in California compared with 26.2 to 23.3% nationwide. Subsequently, the rate of decline slowed down in both California and the rest of the United States, with little change in smoking prevalence until 1996, which has been attributed to cutbacks in the CTPC in 1992. Recently, the CTCP was also associated with a significant decline in mortality from heart disease in California, with changes in age-adjusted death rates shown to parallel the changes in per capita cigarette consumption (Fichtenberg and Glantz, 2000).
The overall success of the CTCP in curbing smoking in California was generally attributed to the multifaceted approach of the program. Action on secondhand smoke is considered an important component not only because it prevents exposure to passive smoke. Workplace and public place smoking bans help current smokers quit and those who continue smoking to cut consumption levels; they also make smoking socially less acceptable. However, several commentators have emphasized that delegitimizing tobacco use and the tobacco industry have to play a central role if a campaign is to be successful.
Failure of Prevention? Injuries in Eastern Europe
Injuries are the leading cause of death and disability among young people in many parts of the world, with an estimated 5.2 million dying of injuries worldwide in 2001. About 11% of the global burden of disease has been attributed to this cause, largely concentrated in low- and middle-income countries (Lopez et al., 2006). Yet, although injuries are largely preventable and many effective strategies are available, they have received relatively little attention from the public health community because they have traditionally been viewed as ‘accidents,’ or random events. This section draws on the experience of Central and Eastern Europe to illustrate the factors that have contributed to the neglect of injuries as an important public health problem in that region.
Injuries have long been a much greater problem in Eastern Europe than in the West and the mortality gap has widened in the early 1990s as death rates continued to fall in most Western countries while Eastern Europe had to confront the challenges of the political transition (Figure 1). There were many reasons for the rise in injury mortality in that region, related in part to a growth in traffic and associated increase in deaths from car accidents and the opening of the borders of the former Soviet Union that included increasing substance abuse, especially alcohol consumption. The increasing social and economic uncertainties arising from the economic downturn in many former Communist countries created environments that placed especially children at an increased risk. This is reflected in rising injury mortality among young people, particularly in the countries of the former Soviet Union that now form the Commonwealth of Independent States (Figure 1). And although rates have been declining recently in many parts, the gap with the West remains wide.
Figure 1. Mortality from injury and poisoning in the European region. Source: World Health Organizational Regional Office for Europe (2007). aNew EU members: Bulgaria, Cyprus, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Romania, Slovak Republic, Slovenia. bEU before 2004: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Luxembourg, Ireland, Italy, Netherlands, Portugal, Spain, Sweden, United Kingdom.
However, the public health response to injuries in this region has been relatively weak (McKee et al., 2000) in part because the scale and scope of the burden attributable to injuries was not well known and thus not perceived as an actual problem. Those responsible for public health were marginalized and often lacked the appropriate skills required to identify public health problems and develop strategies to address them, although this is now changing. Unlike in Western countries, nongovernmental organizations that have often taken the lead on addressing issues of injuries, especially childhood injuries, have either not existed or been slow to develop. Also, a lack of a tradition in intersectoral working has hindered the development of multidisciplinary approaches that bring together a wide range of actors from the public and private sectors to develop and promote effective action. Finally, effective action requires effective implementation (e.g., through legislation). While this may have been possible during the Communist period, many countries faced substantial challenges during the transition, for example, the need to create new democratic structures, often from scratch. Thus, in many parts of the Eastern European region health promotion ranked relatively low on the political agenda and elicited little response.
This example illustrates some of the many obstacles public health may face in developing and implementing effective strategies to prevent disease. However, the case studies on the community approach to prevent coronary heart disease in North Karelia and the California Tobacco Control Program indicate that effective action is possible when there is a functioning public health infrastructure and matching political will.
Ethical Aspects and Evaluation of Disease Prevention
The following section describes ethical aspects of disease prevention and the evidence base of preventive interventions, highlighting the main problems public health professionals are likely to encounter when designing and implementing preventive programs.
Ethical Aspects of Disease Prevention
The purpose of prevention is to reduce the burden of morbidity and mortality in society so it is often considered a priori as ‘a good thing.’ This may be questionable, however, especially as far as measures of mass intervention are concerned. One example is fluoridation of drinking water. Fluoridation is associated with dental caries and because this measure may help to reduce the incidence of dental caries in a population it may be considered to be an important measure to promote public health. However, it could also be interpreted as compulsory medication of the population, violating human rights. The introduction of any preventive program therefore needs to carefully consider whether the intended health gain overrides its ethical costs.
Similar issues arise for immunization campaigns although the expected health gain may be interpreted differently than that achieved from measures such as drinking water fluoridation. Depending on the infectious agent, transmission routes, and the degree to which immunization protects against infection, mass immunization with high uptake will normally protect a given population because most of the population is resistant to infection through immunization (herd immunity). However, if a critical level of immunization is not achieved because of low uptake this may weaken herd immunity and increase the risk of disease outbreaks. Thus, if the public wishes to benefit from prevention through immunization then it must also agree to a degree of compulsion to ensure high uptake to prevent infectious diseases from spreading.
Another example is screening. As described earlier, screening is undertaken in apparently healthy people and a key ethical issue here is labeling or stigmatization. Someone who felt healthy may now be faced with the possibility of not being healthy at all but labeled as, for example, being hypertensive or having developed cancer. This may not only induce anxiety or loss of self-confidence; it may also have a social impact especially when test results are not kept confidential, potentially leading to isolation or loss of insurance (e.g., testing HIV positive). Potential harm also arises when the test gives an inaccurate result. Thus, those who tested false positive may receive treatment although it is not necessary or even harmful. In contrast, those who tested false negative will not only be given false reassurance, it may also lead to treatment being omitted when it is required.
There is also the risk of testing itself. For example, prenatal screening for Down’s syndrome is usually undertaken through amniocentesis, posing a risk to both mother and fetus as it may lead to complications or spontaneous abortion. Applying the test thus needs to be justified by proportionate reason, one being the age of the mother. Yet, while amniocentesis may detect Down’s syndrome it cannot tell whether the condition will be mild or severe. This is an important aspect of prenatal screening as diagnosis of Down’s syndrome may lead to recommendation for termination of pregnancy, which is especially problematic having the possibility of test inaccuracies in mind. Thus, information on the consequences of testing, diagnostic procedures, the disease to be detected, and followup treatment in case of a positive test and related aspects should form an integral part of any screening program.
It is also important to note that introducing a screening program means diverting resources away from other equally or more important health problems, which raises the question of equitable distribution of resources to obtain maximum health benefit. This is particularly problematic in low- and middle-income countries in which the implementation of resource-intensive, high-technology screening programs could reduce the resources available for more urgent health problems and may thus result in lowering the overall public’s health. Therefore, screening programs have to meet high medical and ethical standards to prove their efficacy and, ultimately, their benefit to the participating (ostensibly) healthy population.
Evaluation of Disease Prevention
Prevention effectiveness measures the impact on health of prevention policies, programs, and practices, taking account of efficacy, effectiveness, and cost. However, perspectives on effectiveness are likely to vary between different actors in the field of prevention. Policy makers and public health professionals responsible for priority setting and resource allocation will mostly be concerned with the relationship between investment in a program and its achievement in terms of health outcomes. The population who is to benefit from the intervention may judge effectiveness largely in terms of personal gains such as improvement in quality of life or opportunities for participation. Those involved in the provision of the intervention are likely to rate effectiveness mainly with respect to the practicalities of organization and implementation. From a scientific perspective, effectiveness may involve issues of methodological stringency and level of achievement of predefined outcomes (International Union for Health Promotion and Education [IUHPE], 2000).
The different perspectives highlight the complexity of the concept of effectiveness, with implications for developing an intervention and particularly its evaluation. Consequently, there is a wide range of approaches and methods to assess the effectiveness of preventive interventions. The main measure of effectiveness is the actual outcome of the preventive strategy. Different forms of interventions may produce different outcomes and defining outcomes associated with prevention activity thus forms a central part of program design and evaluation. The research into health promotion has provided a framework for defining outcomes in health promotion (IUHPE, 2000), which, to some extent, is also transferable to preventive intervention. It describes several outcome levels that reflect the different dimensions of health-promoting or prevention activities. Thus, health and social indicators such as morbidity, disability, mortality, and/or quality of life represent the ultimate outcome or endpoint of any health and medical intervention. Intermediate health outcomes relate to the actual determinants of the health and social outcomes that are modifiable and include risk factor levels, provision of and access to (preventive) health services, and environmental factors such as the physical and socioeconomic environment. A third level relates to personal, social, and structural factors that can be modified to change the determinants of health, also interpreted as intervention impact measures. They comprise measures of health literacy such as health-related knowledge, attitudes, and behaviors; social action such as social norms and public opinion; and public policy such as legislation, regulation, or resource allocation.
The gold standard for evaluating the effectiveness of an intervention is the randomized controlled trial (RCT). Yet, while well-conducted RCTs provide the most rigorous design for assessing the effect of an intervention, their application especially in the field of primary prevention and health promotion is often not feasible, because of the very nature of many programs acting at different levels. Available evidence from programs such as the California Tobacco Control Program described previously suggest that the comprehensive approach that comprises different strategies that target different components in the model of causation will be the most effective. However, such interventions are less amenable to control and evaluation by standard experimental designs such as RCTs compared to single components within the program. For example, several studies have demonstrated that providing information about risk factors alone is rarely sufficient to encourage people to change their behaviors or lifestyle. Yet, embedded in a broader strategy that also comprises environmental changes through regulatory efforts and provision of services to actively support efforts to change behavior, information measures are far more likely to make some impact.
This does not mean that complex programs cannot be evaluated or be based on sound evidence. It highlights, however, that assessing the effectiveness of an individual program over a longer period is a complex task. It is modulated by a range of factors, such as the size, comprehensiveness, and duration of the intervention, as well as a wide range of factors not directly related to the intervention, including secular changes in society. This has been illustrated by a meta-analysis of the effectiveness of community heart health programs (Sellers et al., 1997). Based on the observation of considerable variation in the effectiveness of such programs, raising questions about their usefulness, the analysis showed that some of the variation is attributable to characteristics of their evaluation design and implementation. Thus, studies with longer follow-up and higher response rates tended to produce higher effects. However, characteristics of the intervention itself also contributed to the variation, with programs that included environmental change as a substantial component producing larger effects than those that did not. In conclusion, the evidence produced by community interventions to date will ‘‘help us to refine our understanding of which interventions work how well with which populations under which conditions’’ (Sellers et al., 1997).
Issues outside the Health-Care Sector That Influence Disease Prevention
It is increasingly being recognized that decisions made outside the health-care sector can have major consequences for population health. One example is the collapse of the Communist regimes across Central and Eastern Europe around 1990, which had enormous social and economic effects; the former Soviet republics particularly experienced a deep economic recession. The speed by which the chain of events leading to the collapse of the Soviet Union unfolded was hardly predictable, nor was its impact on population health foreseeable in order to implement appropriate preventive measures in time. Thus, many countries experienced an immediate health crisis, mainly affecting adult men. In countries such as the former German Democratic Republic, Poland, and the Czech Republic, a temporary decline in life expectancy at birth among men was followed by a rapid and substantial increase. Other countries such as Russia and the Baltic states, however, experienced a spectacular worsening in health, with life expectancy at birth among Russian men falling by more than 6 years between 1990 and 1994. Since then life expectancy has been rising again although in Russia and Ukraine this improvement was arrested in 1998 and has since reversed.
The phenomenon of a marked rise in death rates of injuries and violence especially in Russia and the Baltic states around 1990 was noted earlier. Another point of concern is the reemergence of infectious diseases, in particular tuberculosis and HIV/AIDS, especially in the former Soviet republics. These developments not only pose a major threat to the population’s health within the region, but neighboring countries will also be at increased risk.
This last point illustrates what is usually linked to processes of globalization, that is, the more rapid spread of communicable diseases across national borders because of improving means of transport and communication, especially the substantial increase in worldwide travel (McKee et al., 2001). Also, the worldwide increase in antibiotic resistance has been associated with processes of globalization, largely because of over- and misuse of antibiotic drugs worldwide, which increases the threat of resulting resistance spreading through greater human mobility. A related issue is the current level of international migration, which introduces an additional complexity to the design of programs for infectious disease control.
Other aspects of globalization such as global trade also pose potential risks to population health. While the growth in global trade in general has brought many benefits to many people over time, its recent liberalization and deregulation is likely to not benefit countries or population groups equally, thereby increasing health inequalities within and between countries and potentially promoting political instability. Finally, an increasing threat to the world’s population health arises from global environmental changes such as global climate change, stratospheric ozone depletion, or land degradation.
These issues pose substantial challenges to public health professionals to develop sustainable strategies for disease prevention. From this brief description, however, it becomes clear that changes in health policy alone will not be sufficient to improve the population’s health. Instead, more health gains in terms of prevention will be achieved by influencing public policies in domains such as trade, food industry, pharmaceutical production, agriculture, urban development, and taxation policies.
- Asma S, Yang G, Samet J, et al. (2002) Tobacco. In: Detels R, McEwen J, Beaglehole R and Tanaka H (eds.) 4th edn., Oxford Textbook of Public Health 4th edn. vol. 3, pp. 1482–1502. Oxford, UK: Oxford University Press.
- Beaglehole R, Bonita R, and Kjellstrom T (1993) Basic Epidemiology. Geneva, Switzerland: World Health Organization.
- Committee of Inquiry into the Future Development of the Public Health Function (1988) Public health in England. London: HMSO.
- Fichtenberg CM and Glantz SA (2000) Association of the California Tobacco Control Program with declines in cigarette consumption and mortality from heart disease. New England Journal of Medicine 343: 1772–1777.
- Global Polio Eradication Initiative (2007) http://polioeradication.org/
- International Union for Health Promotion Education (IUPHE) (2000) The evidence of health promotion effectiveness. Brussels, Belgium: European Commission.
- Last JM (1995, 2001) A Dictionary of Epidemiology. Oxford and New York: Oxford University Press.
- Lopez AD, Mathers CD, Ezzati M, Jamison DT and Murray CJL (eds.) (2006) Global Burden of Disease and Risk Factors. Washington, DC: World Bank.
- Mathers CD and Loncar D (2006) Projections of global mortality and burden of disease from 2002 to 2030. Public Library of Science Medicine 3: 2011–2030.
- McKee M, Zwi A, Koupilova I, Sethi D, and Leon D (2000) Healthpolicy- making in central and eastern Europe: Lessons from the inaction on injuries? Health Policy and Planning 15: 263–269.
- McKee M, Garner P and Stott R (eds.) (2001) International Co-operation in Health. Oxford, UK: Oxford University Press.
- Puska P, Nissinen A, Tuomilehto J, et al. (1985) The community-based strategy to prevent coronary heart disease: Conclusions from the ten years from the North Karelia Project. Annual Reviews in Public Health 6: 147–193.
- Rodgers A, Lawes C, and MacMahon S (2000) Reducing the global burden of blood-pressure-related cardiovascular disease. Journal of Hypertension 18(supplement 1): S3–S6.
- Rose G (1981) Strategy of prevention: lessons from cardiovascular disease. British Medical Journal 282: 1847–1851.
- Rose G (1985) Sick individuals and sick populations. International Journal of Epidemiology 14: 32–38.
- Rose G (1992) The Strategy of Preventive Medicine. Oxford, UK: Oxford University Press.
- Sellers DH, Crawford SL, Bullock K, and McKinley JB (1997) Understanding the variability in the effectiveness of community heart health programs: A meta-analysis. Social Science and Medicine 44: 1325–1339.
- Wilson JMG and Jungner G (1968) Principles and Practice of Screening for Disease. Public Health Papers, vol. 24. Geneva, Switzerland: World Health Organization.
- World Health Organization (WHO) (2000) Global strategy for the prevention and control of noncommunicable diseases. Report by the Director-General to the 53rd World Health Assembly. Geneva, Switzerland: WHO.
- World Health Organization Regional Office for Europe (2007) European health for all database. Copenhagen, Denmark: World Health Organization Regional Office for Europe.