3.2 Public health policy in developed countries*
Oxford Textbook of Public Health
Public health policy in developed countries*
Assessing the effects of policy
Examples of policies to improve health
Some issues emerging from examples considered
Social institutions and health: psychosocial versus material emphases
Public health problems and public health investments
The search for equality
Making progress safe
Politics is a strong and slow boring of hard boards. It takes both passion and perspective. (Weber 1921)
The scope and purpose of public health policy is implicit in widely used definitions of public health. The Acheson Report in England defined it (following Winslow) as:
The science and art of preventing disease, prolonging life and promoting health through organised efforts of society. (Secretary of State for Social Services 1988)
Public health policies are thus the policies that guide these ‘organised efforts’ to protect and improve health—and the real interest in policy is the prospect that it might make a difference to the public’s health. Although policy is mainly the province of governments (national, regional, and local), civic organizations such as professional bodies and major health charities along with supranational bodies, such as the World Health Organization (WHO) and the World Bank, are also involved. This chapter is limited to ‘developed countries’, excluding countries in transition from socialism. These can be regarded as equivalent to the ‘high-income’ category used by the World Bank: the countries of Western Europe, Canada, the United States, Australia, New Zealand, Israel, Singapore, and Japan (excluding Taiwan and various small states for which data are less readily available).
Assessing the effects of policy
It is paradoxical that the greatest interest in public health policy now exists in developed countries where the benefits of public health activity may seem least apparent. In other times and places, the value of this endeavour could seem clear. Thus in 1890 John Simon could look back on half a century of sanitary reform in England and confidently attribute declining mortality from infectious causes to the organized efforts to clean up the cities and towns (Simon 1890). Or, if one were to take ‘lower-middle’ income countries in today’s world, three- to fourfold differences in child mortality rates are associated with substantial differences in the coverage achieved by public health programmes. Countries that are ‘good performers’ in reducing child mortality (such as El Salvador, Jordan, and Sri Lanka) have higher immunization coverage and larger proportions with safe water supplies—suggesting that public health programmes are making important contributions to declines in child mortality (McMichael and Powles 1999).
Analogous attempts to demonstrate the value of public health endeavour in lowering mortality levels in developed countries soon run into difficulties. In examining changes in life expectancy over the last two decades, there is some, unsurprising, evidence of ‘catch-up’ with countries starting at lower levels tending to show greater gains (Fig. 1). Outliers, relative to the trend, are as ‘good performers’ Japan, France, Singapore, Canada, and Australia, and as ‘poor performers’ Denmark, Norway, the United States, The Netherlands, and Portugal. It is not at all plausible that these ‘poor performers’ have lagged behind and lost rank just because they underinvested in public health endeavours. (Life expectancy at birth in these countries is not significantly related to income per person.)
Fig. 1 Change in life expectancy at birth for both sexes combined between 1975/1980 and 1995 by starting level, high-income countries. (Data from World Resources Institute 1998.)
There are three major impediments to relating overall mortality levels in rich countries to their public health endeavours. Firstly, there are no readily available measures of the ‘independent variable’—the amount of ‘organised effort’ to prevent disease and injury that has actually been taking place. Secondly, variation in adult mortality levels is strongly influenced by the timing and magnitude of two major overlapping epidemics: (a) diseases caused by tobacco smoking, and (b) vascular diseases. Both of these have proved relatively insensitive, in the short term, to the effects of identifiable public health endeavours. Thirdly, variations in mortality levels in middle to late life represent the lagged effects of changes in disease determinants over preceding decades and these temporal relationships are not easy to specify or quantify. Illumination is thus unlikely to come from analyses employing short time frames and highly aggregated disease groupings. A sharper focus is needed on different types of policy responses to different types of public health problems, if useful lessons are to be drawn. A fourfold classification for this heuristic purpose is set out in Table 1. Note that this is a classification for operational programmes directed towards major components of the avoidable burden of disease. It does not adequately cover important maintenance functions, nor the ‘infrastructure’ needed to support operational programmes. Nor does it address social inequalities in health. (This last topic will be touched on later.) Furthermore, the examples will not be discussed comprehensively but rather used to illustrate points that may be of wider significance in understanding the relation of public health policies to health levels in developed countries.
Table 1 Classification of public health problems by the predominant type of response needed
Examples of policies to improve health
Administrative means: fluoridation
Fluoridation of water supplies provides a classic illustration of public health improvement by administrative means; its limited implementation may yield lessons in public health politics and economics.
In 1929, a dentist in Colorado, in the United States, observed that mottled tooth enamel (which he suspected was associated with the water supply) was associated with fewer dental caries. The factor in the water supply was soon identified as fluoride. During the 1930s, inverse associations between fluoride concentrations in the water supplies of 21 cities and a newly developed quantitative index of dental caries (decayed, missing, and filled teeth) were reported. Caries prevalence declined as concentrations of natural fluoride in the water supply approached 1 ppm, with little additional benefit at higher concentrations. In 1945 community intervention trials of the effects of adjusting fluoride levels to 1 to 1.2 ppm began in four pairs of cities, with one pair in Canada. After 13 to 15 years caries was reduced by 50 to 70 per cent in the experimental cities compared with the controls. Official recommendations for the fluoridation of community water supplies were formalized in the United States in 1962 and the proportion of the population receiving fluoridated water from community supplies rose from 40 per cent in the mid-1960s to around 56 per cent in 1992. Over this same period, the mean decayed, missing, and filled teeth among all 12-year-olds fell dramatically from around 4 to just over 1. With time, the observed effects of water fluoridation have lessened as those without such supplies get fluoride from other sources, including toothpaste and drinks and foods manufactured in fluoridated areas (Division of Oral Health 1999).
It has been estimated that for large cities in the United States, the (undiscounted) costs of fluoridated water per individual over a lifetime are of the order of $10 (American dollars used throughout this chapter). By contrast, the lifetime costs of restoring a carious tooth could exceed $1000. Fluoridation is thus one of the most cost-effective preventive measures available (US Department of Health and Human Services and Batelle 1995).
Despite this attractiveness, the expansion of fluoridation slowed in the United States in the 1980s and 38 per cent of the American population served by community water supplies was still not benefiting from fluoridation in 1992 (Division of Oral Health 1999). In the United Kingdom, only 10 per cent of the population receives fluoridated water and no fluoridation schemes have been implemented since 1985 (British Dental Association 1999). Failure to fluoridate needs explanation.
In both the United States and the United Kingdom fluoridation enjoys the support of around 70 to 75 per cent of respondents to national opinion surveys (American Dental Association 1999; British Dental Association 1996). However, implementation requires local decisions and these have frequently been blocked by opponents. In the United States, local referenda have been lost because opponents have been more committed and better organized than proponents. Legal challenges in the lower courts have also obstructed progress, though the highest courts have upheld none. In the United Kingdom, privatization of the water companies created an additional obstacle with several water companies refusing to implement fluoridation when requested by the local health authorities (British Dental Association 1996).
The vocal minority opposed to fluoridation rest their claims on health risks that have repeatedly failed to be substantiated (National Research Council 1993). Their continued currency attests to the widespread, and perhaps increasing, distrust of technical manipulation of the human environment, also evident in the United Kingdom in the opposition to genetically modified foods. Popular sentiments tend to generalize, and well-founded observations on the progressive degradation of the living environment by industrial expansion may easily join together with ill-founded suspicions of particular technologies.
Fluoridation was not mentioned in the most important public health policy document produced by the British Conservative governments between 1979 and 1997 (Secretary of State for Health 1992) and was ‘overlooked’ in the incoming Labour government’s discussion paper which claimed to be ‘setting out a third way’ in public health policy ‘between the old extremes of individual victim blaming on the one hand and nanny state social engineering on the other’ (Secretary of State for Health 1998). It appeared that the political aversion to ‘engineering’, as a means of health improvement, also encompassed civil engineering. The omission of explicit support for fluoridation was widely criticized and the final policy document promised ‘to introduce a legal obligation on [privatized] water companies to fluoridate where there is strong local support for doing so’—assuming a satisfactory outcome from a newly commissioned review of the scientific evidence (Secretary of State for Health 1998).
Inaction on fluoridation in the United Kingdom between the mid-1980s and the late 1990s appears to have been partly a consequence of the ascendancy of neoliberal political ideas. Even the notionally social-democratic Blair Labour government (elected with an overwhelming majority in 1997) appeared, during its initial period of office, to be more afraid of accusations from its opponents that it favoured a ‘nanny state’ than of criticism from its own supporters that it was not using state powers to beneficial effect.
Fluoridation thus introduces several themes pertinent to the consideration of public health policy in rich countries. One is the power of research using quantitative methods, including experiments on whole communities, to expand the repertoire of effective means for controlling disease and injury. Another is the possibility of massive disjunctions between the cost-effectiveness of a preventive measure and the political feasibility of its implementation. Perceptions of risks and benefits held by vocal minorities may depart substantially from those of experts, and governments may be more sensitive to their reputations in the eyes of the press and other powerful bodies than they are to public opinion. Some observers see controversy and conflicts about risks as being typical of ‘late modernity’ (Giddens 1998).
Enhanced coverage with clinical procedures: control of high blood pressure
Raised blood pressure is a major contributor to risk of ischaemic heart disease and stroke, ranked respectively first and third as leading causes of disability-adjusted life years lost in rich countries (Murray and Lopez 1996). Meta-analyses of prospective studies have shown that a prolonged difference of about 5 mmHg in usual diastolic blood pressure is associated with a 21 per cent difference in risk of ischaemic heart disease and a 34 per cent difference in risk of stroke (MacMahon et al. 1990). All of the excess stroke risk appears to be reversible if blood pressure is lowered to target levels by medication and about half of the excess risk of heart attack may be reversible (Collins et al. 1990).
It is typically doctors who are expected to find cases of high blood pressure, to advise on drug and other therapies, and to help maintain control. Thus although all these processes also depend on the active co-operation of the public, high blood pressure provides the most important example of a persisting major public health problem that is being mainly addressed by professionally controlled measures applied to individuals.
The National High Blood Pressure Education Program adopted in the United States in 1972 is perhaps the best-known public health policy addressing high blood pressure (Joint National Committee 1997). Such policies have aimed to make case-finding more complete and control more effective. Progress has, rather confusingly, been monitored by using as a denominator individuals with blood pressures above 140/90 on a single occasion plus those reporting antihypertensive medication, and then calculating the percentages ‘aware’ that they have high (usual) blood pressure (which some of them will not have), are on treatment, or are controlled (Table 2). Despite their problems, these data show a substantial improvement in case-finding and control between the late 1970s and the late 1980s with little subsequent gain.
Table 2 Trends in awareness, treatment, and control of high blood pressure in people aged 18 to 74 years in the United States, 1976 to 1994
The Framingham Study cohorts have provided the opportunity to track changes over a longer time span, though in a population that is likely to be more health conscious than average. The proportion of those aged 45 to 74 who reported antihypertensive medication increased from 2 per cent in the 1950s to 25 per cent in the 1980s in males and from 6 to 28 per cent in females. Those with blood pressures above 160/100 measured on a single occasion (and irrespective of treatment status) fell from 19 to 9 per cent in the case of males and from 28 to 8 per cent in females. Greater proportionate declines occurred in progressively higher blood pressure strata and the prevalence of left ventricular hypertrophy fell markedly. These findings are consistent with other data indicating substantial secular declines, especially in severe hypertension (Mosterd et al. 1999).
In assessing the National High Blood Pressure Education Program as a public health programme, two questions need to be addressed. To what extent has the improvement in case-finding and management for high blood pressure been attributable to the ‘organised effort’ of the National High Blood Pressure Education Program? How much health benefit is attributable to the more effective clinical management of blood pressures?
Studies designed to answer the first question appear to have been very limited. In rural Kentucky a community high blood pressure control programme was run in two counties between 1979 and 1984 with a third county serving as control. In the intervention counties the percentage of ‘hypertensives’ whose blood pressure was controlled to below 140/90 rose from 37 to 53 per cent, with no change in the control county. Cardiovascular mortality fell in the intervention counties, while remaining constant in the control (Kotchen et al. 1986).
Some American observers believed, in the early 1980s, that ‘the documented improvements in hypertension control since the beginning of the National High Blood Pressure Education Program must be considered a major contribution … to the decline in cardiovascular mortality rates’ (Lenfant and Roccella 1984). Between 1972 and 1994 in the United States, age-adjusted death certification rates for stroke fell by nearly 59 per cent and for ischaemic heart disease by 53 per cent. The fall in stroke was steepest in the early period and appears to have ceased in the early 1990s (Joint National Committee 1997).
Risk of stroke provides the more sensitive test of the benefits of enhanced control of high blood pressures because risk of stroke rises more steeply (and exponentially) with increasing blood pressure; furthermore, it is more completely reversed by treatment. For people aged 60 to 74, data from the American national health surveys show a substantial shift downwards in blood pressure distributions from the early 1960s to the most recent survey period around 1990. Because nearly all treated people stay above the median, reductions in the median provide a useful measure of shifts in the central tendency due to causes other than treatment; reductions in the mean, by contrast, will be due to both reductions in the central tendency and inward shrinkage of the top tail of the distribution. Median systolic pressures at ages 60 to 74 fell over this period by about 16 mmHg, from 148 to 132 mmHg. The shift in the upper tail of the distribution was more marked, with 90th centiles falling by about 30 mmHg—from around 191 to 160 mmHg. Some of this proportionally greater shift in the top tail could be attributable to the shrinkage of the distribution as the central tendency falls; for example, at ages 18 to 29, where the effects of medication are likely to be small, the 90th centile fell by 14 mmHg when the median fell by 8 mmHg (all estimates from smoothed distributions in Fig. 1 of Burt et al. (1995)). Nevertheless much of the fall in the prevalence of people at risk because of their high blood pressures can be assumed to have been due to enhanced clinical control of blood pressure.
Rose coined the term ‘prevention paradox’ to describe how, when risk is related monotonically to a quantitative attribute such as blood pressure, the interventions which offer most to the individuals at high risk contribute less to reducing the population burden of the disease than do small downward shifts in the whole distribution (Rose 1985). Strachan and Rose (1991) reworked these analyses taking account of the misclassification of risk status when blood pressure is only measured on a single occasion. Because distributions of true (usual) blood pressures are not readily available they modelled them using data from the Whitehall Study with three alternative levels of intraindividual variation. Their models suggested that, assuming a reliability coefficient of 0.5, over 50 per cent of the population risk of fatal stroke attributable to true (usual) blood pressures higher than those in the lowest decile, was to be found in those whose true pressures were in the top decile. Yet, even in these apparently promising circumstances, a ‘high-risk’ case-finding strategy that correctly identified all in the true top decile, and that achieved an average reduction of 7.5 mmHg diastolic in all those offered treatment, would reduce stroke mortality only by about the same amount as would result from a 3-mmHg reduction in diastolic blood pressures across the whole distribution (Strachan and Rose 1991). Thus, although classification on the basis of usual blood pressures enhances the relative effectiveness of the ‘high-risk’ strategy in relation to stroke, it still remains modest when compared with downward shifts in the whole distribution of blood pressures.
No analyses of this kind appear to have been reported using data for the United States. However, earlier analyses of the contribution of hypertension treatment to the decline in stroke mortality between 1970 and 1980 placed it in the range of 6 to 25 per cent (Bonita and Beaglehole 1989).
In conclusion, the ‘organised efforts’ of the National High Blood Pressure Education Program will account for part of the improved case-finding and management for people with usual blood pressures above treatment thresholds. This improved management will account for part of the decline in the prevalence of people above treatment thresholds. The decline in the prevalence of people above treatment thresholds will account for part of the decline in stroke mortality attributable to raised blood pressures. Furthermore, given that MONICA (Monitoring of Trends and Determinants of Cardiovascular Diseases) results show that there are major determinants of stroke other than blood pressure (Stegmayr et al. 1997), it is also likely that all the causes of downward movements in blood pressure combined will have been only one part of the environmental changes contributing to declining stroke mortality.
Despite this cumulative diminution of the contribution of the National High Blood Pressure Education Program (in accordance with Rose’s ‘prevention paradox’), that contribution is still likely to have been very worthwhile because even small reductions in the heavy burdens imposed by heart attack and stroke will add up to a large benefit in absolute terms. Furthermore, the gains attributable to the National High Blood Pressure Education Program are notable for having been achieved within a pluralistic and organizationally diverse system of medical care.
This example also illustrates the main appeal of preventive strategies involving clinical procedures—their mode of action can be readily appreciated by practising doctors and their effectiveness at an individual level can be determined by randomized controlled trials—as well as the quantitative limitations of these strategies where risk is a continuously graded function of the determinant to be modified.
Behaviour change: HIV and sudden infant death
The epidemic of HIV infection in Europe followed that of the United States. The time course of the epidemic through the population of homosexual and bisexual males in England and Wales has been reconstructed by statistical modelling (de Angelis et al. 1998). The incidence of HIV infection appears to have peaked in 1983 and then to have fallen sharply (Fig. 2). These estimates are broadly consistent with the time trends shown by laboratory reports of hepatitis B infection in homosexual males, which peaked in 1984 but then showed a somewhat slower decline. Against the time course of the new infection rate can be set the timing of the formal control measures. Intensive ‘social marketing’ campaigns to promote changes to safer sexual practices were not launched by the British government until 1986 (Acheson 1993). It is thus likely that most of the change in sexual practices responsible for the sudden decline in HIV incidence after 1983 had occurred before the formal programme began. How is this to be explained? Almost certainly new knowledge about the dire consequences of HIV and how it was spread passed through both the general news media and through communication channels used especially by homosexual and bisexual communities. Because the epidemic in England substantially lagged behind that in the United States (perhaps by 3 years) there was an opportunity to learn from the United States where suspected modes of transmission had been identified as early as 1982, well before the identification of the virus in 1984 (US Department of Health and Human Services and Batelle 1995). From a societal point of view, this example shows the possibility of ‘public health success without programmes’. In circumstances such as those surrounding the early HIV epidemic, the ability of formal public health programmes to contribute to health improvement may be limited by the need to await the building of a supporting political consensus. In the United States, a national household drop of an eight-page brochure, Understanding Aids, from the Surgeon General was not conducted until 1988.
Fig. 2 HIV incidence in homosexual and bisexual males, England and Wales. Estimates by back-projection for 1979–1990 (with 95 per cent confidence interval) and timing of main public health campaign. (Source: D. De Angelis, personal communication.)
The decline in sudden infant death syndrome in England illustrates the same point. The relationship between sleeping position and risk of sudden infant death syndrome was discussed in the letters pages of the Lancet during 1988 (Beal 1988), and this and other theories were widely discussed in magazines commonly read by mothers with young infants. Death rates from sudden infant death syndrome fell by more than a third in the 3 years before the British government’s formal public health programme (‘Back to sleep’) began in December 1991 (OPCS 1988, 1995; Hiley and Morley 1994), again suggesting that mass behaviour change had occurred in response to new information flowing through the general mass media and, in an uncoordinated way, via health professionals. It is important to note in this case, however, that the rate of decline did accelerate sharply after the formal programme, with the rate halving in the following 12 months.
Thus, in highly literate and health-conscious populations, much, if not most, of the health benefit from new knowledge may flow more or less automatically from its dissemination through channels other than formal public health programmes; in these circumstances, it is the advance of public health science, perhaps even more than the strength of public health programmes, that sets the pace for health improvement (Dwyer and Ponsonby 1996). This is not to say that the incremental gain from formal programmes is typically negligible; it may still be very worthwhile relative to their typically modest resource requirements. But the main point to emerge from these examples is that the ‘organised efforts’ that have contributed most to reducing the burden of these diseases have been the research efforts. Thus, in developed countries, investment in the development of public health science is the most fundamental component of public health policy. In the cases of HIV and sudden infant death syndrome, the benefits flowing from scientific advance were realized largely without the aid of practising professionals. Therefore medicine and public health should not be understood just as domains of professional practice; they are, more fundamentally, cultural resources appropriated by all members of society—lay as well as professional. The mistake of confusing ‘medicine’ and ‘public health’ with the professional practice of these disciplines is commonly made, as in the title of a recent review of the benefits deriving from new knowledge: ‘Medicine [used here to mean professional medicine] matters after all’ (Bunker 1995).
Behaviour change: road traffic injuries
Traffic injuries rank fifth in their contribution to the burden of disease and injury in developed countries (Murray and Lopez 1996). Because of the short time lags between control measures and their expected effects, traffic injuries also provide a sensitive field in which to explore the relationship between policies, programmes, and effects.
Death rates from traffic crashes per unit registered vehicles follow a very general, and pronounced, downward trend as the number of motor vehicles increases in relation to population (Smeed’s law: deaths/vehicle = 0.0003 (vehicles/population)–0.66) (Smeed 1972). In the mid-1960s, two-thirds of 70 populations analysed by Smeed had rates within 40 per cent of his prediction. This implies that societies generally learn how to use motor vehicles more safely as both familiarity with them and the resources available for safety measures increase. Because the overall tendency is general, it is unlikely to depend on the specifics of policies variably adopted. However, it is also the case that, around this general trend, some societies have performed better than others.
Figure 3 shows the decline in Victoria, Australia, in deaths per 10 000 vehicles as the number of vehicles increased in relation to population. Over 80 per cent of the reduction in fatalities per 10 000 vehicles that occurred between 1920 and 1995 happened before 1970. During these five decades, Victoria generally had rates in excess of Smeed’s prediction. Then, in a little more than two decades from 1970, it changed from being a relatively poor performer in this domain to being one of the best (Fig. 4).
Fig. 3 The decline in traffic fatalities per 10 000 vehicles with increasing motorization, State of Victoria, Australia, 1920 to 1995. (Data from Smeed 1972; Hawthorne 1991.)
Fig. 4 Decline in traffic fatalities per 10 000 vehicles, 1960 to 1995, State of Victoria, Australia. (Data from VicRoads (1993); Smeed 1972; Department of the Environment, Transport and the Regions 1998.)
During the 1960s there had been a growing political consensus in Victoria that the loss of so many lives on the roads was unacceptable. In December 1970 it became the first jurisdiction in the world to make the wearing of seatbelts compulsory. A string of legislative measures followed, including random testing of the breath alcohol concentrations of drivers in 1977. After the decline in fatality rates faltered in the late 1980s, a very strong ‘social marketing’ campaign was launched in combination with intensive policing (Powles and Gifford 1993). The number of speed camera checks per year rose to eight per licensed driver and the proportion of vehicles recorded as speeding fell from 20 to 3 per cent. In 1994 1.6 million random breath tests were performed, a number equal to about half of the driving age population (Hendrie and Ryan 1995). Fatality rates fell sharply and have stayed down. During the early 1990s the Traffic Accident Commission, which carries all compulsory traffic injury insurance and makes ‘no-fault’ compensation payments to victims, spent several dollars per person on traffic accident prevention programmes, mainly of a ‘social marketing’ character, with intensive use of paid television advertising. Ten per cent of the spend was allocated to evaluation, from which the Traffic Accident Commission concluded that its benefits-to-costs ratio, from reduced injury claims, was at least 3 (Cameron and Newstead 1996). The ratio was larger still when estimates of the social costs of traffic injuries were used rather than average compensation payments. Thus effects that seem small relative to background trends may still be very worthwhile relative to the costs incurred.
Four points can be made as follows. Most of the very large secular declines in traffic injury deaths per unit vehicle (or distance travelled) observed around the world are likely to have occurred with a substantial degree of independence from the specific policies and programmes adopted in different political jurisdictions. Against this broad background trend, a second-order, but nevertheless important, degree of variation seems attributable to the intensity and nature of the control measures taken. In the relatively compact political environment of an Australian state, it was possible to build support for the escalation of control measures as less forceful measures proved inadequate to achieve widely desired goals—notwithstanding a political culture that valued personal independence. The comparative trend line for Britain (in Fig. 4), ending with rates among the lowest in Europe, shows that Victoria does not differ so much in the level it has attained as in its rate of improvement during two and a half decades of intensive political attention.
As an addendum, one may also note the powerful social benefits of having single substantial ‘pots of gold’ for dealing with leading sources of disease or injury. By linking the size of the ‘pot’ to the size of the problem—in this case by the level of compensation payments for traffic injuries—a resource is created that bears some commensurability with the public health challenge faced. The custodians of this fund can then gain social approbation both by reducing a recognized ‘evil’ and by reducing the financial levy needed to compensate for it. (Antismoking programmes funded by hypothecated tobacco taxes, as also exist in Victoria and in California, exploit an analogous link.)
Behaviour change: smoking
It is ironic that medicine provided the ‘cultural bridge’ that enabled tobacco use to be rapidly transferred from the exotic rituals of the Amerindian cultures to the everyday life of seventeenth century Europe. By explaining tobacco’s properties within the contemporary humoral theories of well being, doctors such as the Sevillian Nicolas Monades provided what was to be the main (medical) justification for tobacco use until into the nineteenth century, when ‘recreational’ justifications came to the fore (Goodman 1993). With the advent of manufactured cigarettes in the late nineteenth century, tobacco use was made more convenient and more deadly. In the twentieth century, increased purchasing power resulting from economic development has been almost universally accompanied by widespread adoption of what is now known to be essentially an addiction to nicotine (Tobacco Advisory Group 2000). These epidemics of nicotine addiction can, on the experience of ‘early adopters’ such as England, be expected to last at least a century (Lopez et al. 1994). As a public health problem, cigarette smoking is thus distinguished not only by the great quantity of disease attributable to it—recently accounting, in the United Kingdom, for about a third of male and a quarter of female deaths between 35 and 69 years of age (Peto et al. 1994)—but also by the protracted time-scale over which it evolves. For example, it will be over half a century before the full health effects of onsets of nicotine addiction in today’s adolescents will be manifest. Tobacco smoking is, in addition, a form of addiction that is both licit and extremely profitable.
The nature and magnitude of the health effects of cigarette smoking were mostly revealed by epidemiological studies conducted between the late 1940s and the mid-1960s. A question of interest is how this new knowledge (since much strengthened) has influenced the course of the smoking ‘epidemic’ and the epidemics of disease that have followed in its wake.
From around 1950 to the mid-1960s, new knowledge of the health effects of smoking flowed from scientists to the public mainly through the general news media. The Royal College of Physicians report in the United Kingdom in 1962 (Royal College of Physicians 1962) and the Surgeon General’s report in the United States in 1964 (US Public Health Service 1964) were ‘organised efforts’ that nevertheless depended for their effects on such news coverage. A study of this process in the United States showed that the intensity of print media discussion of the risks of smoking was closely mirrored by adult smoking cessation rates through the 1950s and 1960s (J. Pierce, personal communication, 1999). The same group have also shown a close match between the intensity and targeting of commercial efforts and smoking uptake rates in adolescents (Pierce et al. 1994, 1998).
The intensification of ‘organised efforts’ to discourage tobacco smoking dates mainly from the 1970s. The main measures have included price increases (by specific taxes), requirements for warning labels on tobacco products and on advertisements, restrictions on smoking in public places, health education in schools, mass education and persuasion, enhanced advice (and assistance with cessation) from health professionals, and bans on advertising and other forms of promotion.
Attempts to assess the contribution of these measures to national trends in smoking prevalence (and, with appropriate lags, to national trends in attributable mortality) must take account of the fact that the onset of the epidemics of cigarette smoking, before its health effects were understood, took place at different times in different countries. In Europe, British males and females and Finnish males (but not females) were ‘early adopters’ of cigarette smoking (Lopez 1996). A general pattern of ‘first in, first out’ of the smoking epidemic might have been expected, to some extent independently of the relative strength of national counter-measures. Lung cancer mortality in early middle-age (ages 35–54 years) peaked in 1955 to 1959 in British males and in 1960 to 1964 in Finnish males. Falls since these peaks have exceeded 50 per cent and have recently been somewhat more rapid in Finland. It also happens that Finland (which banned tobacco advertising in 1978) (Harkin et al. 1997) and the United Kingdom have been among the leaders in efforts to reduce smoking. Finnish females, whose delayed smoking epidemic came to maturity during a period of tobacco control activity, have experienced a lung cancer mortality peak (at about 7 per 100 000 for those between 45 to 54 years old in 1990) less than one-third as great as that experienced by British females (about 27 per 100 000 for those between 45 to 54 years old in 1975) (Peto et al. 1994). The size of the smoking epidemic in British females was largely determined in the precontrol period (Lopez 1996). This pattern—of smoking epidemics having lower amplitudes when maturing in an environment of tobacco control activity—suggests that control measures are effective.
Based on systematic assessments of the effects of a number of tobacco control measures, a 1999 World Bank report concludes that tobacco control measures are highly cost-effective, with price increases by specific taxes giving the greatest yield (Table 3). The Bank is now a strong advocate of comprehensive tobacco control policies (Jha and Chaloupka 1999).
Table 3 Estimates of the cost-effectiveness of tobacco control measures in high-income countries (World Bank 1999)
Against the generally favourable trends in adult smoking prevalences in high-income countries stand adverse trends since the late 1980s in smoking uptake by adolescents in a number of countries—the United States, Canada, Australia, the United Kingdom, and Austria (Hill et al. 1995; Gilpin and Pearce 1997; Harkin et al. 1997; Jarvis 1997; Spurgeon 1999). Responding appropriately to these reversals poses a number of challenges: the limited effectiveness of school-based programmes, especially those that are knowledge-based (US Department of Health and Human Services 1994), and the long time lag before substantial benefits will be experienced (as noted above). In the light of these, the World Bank document stresses the need to maintain an emphasis on smoking cessation in adults if benefits are to be seen in the next half century. Conclusions such as these may be supported by calculations of the discounted present value of future health states, which show that gains in the mid to distant future count for little today. An alternative appreciation of this situation is considered further below in the section on social capital.
In summary, cigarette smoking remains the leading public health problem in developed countries. It is without rival in the disease burden it generates. It also illustrates well the central importance of calendar time in the assessment of public health problems. Whilst peak smoking prevalences appear to have been lower in the higher educated strata in ‘late-uptake’ countries—where knowledge of health effects has had more opportunity to influence behaviour—high smoking prevalences among professional groups, such as Spanish doctors (Harkin et al. 1997), shows the limited effects of knowledge outside the context of comprehensive control policies and normative change. Reductions in attributable mortality within the next half century will need to come from encouraging and supporting cessation in current smokers. But if the course of the epidemic of nicotine addiction is to be curtailed, intergenerational transmission must also be minimized. The World Bank analyses show that resources allocated to tobacco control are nowhere commensurate with the magnitude of potential gains to health still to be won.
The history of efforts to reduce the damage to health caused by tobacco also illustrates how the development of quantitative methods has supported appropriate policy responses. It is striking that high-level policy debates in the United Kingdom during the 1950s and 1960s revolved around a largely illusory search for ‘proof of causation’ rather than estimation of how much was at stake (Pollock 1999). With the subsequent consolidation of epidemiological findings and their transmission to political and wider publics (Peto et al. 1994), quantitative assessments have become much more central to policy deliberations—reflecting again the preoccupation with ‘risk’ characteristic of ‘late modern’ societies.
Unsolved problems: physical inactivity and obesity
Unless Prudence be a constant attendant on Opulence … ’tis better living on a slender fortune. Richard Mead (1673–1754)
The material basis of modernization lies in the replacement of muscle, wind, and water power by energy mobilized through steam, electricity, and liquid hydrocarbons. Of the main adverse consequences for health, two—air pollution and transport injuries—have been largely brought under control. The third—the physiological consequences of declining energy expenditure—remains unsolved.
As recently as the mid-1950s many occupations still required substantial muscular exertion. Coal miners studied in Fife, Scotland, in 1952 had to walk over a mile to the underground coalface where they would hew and load 6 to 9 tons of coal onto a moving belt each working day. Compared with the colliery clerks, they expended some 3.6 more megajoules per working day (15.3 versus 11.7 MJ/day) (Garry et al. 1955). The virtual disappearance of such strenuous occupations has contributed to the decline in daily energy expenditure in developed countries. So too has the replacement of walking and cycling by mechanized transport. Even in as short a space as 7 years (1985 to 1992), there were substantial measurable declines in the distances walked and cycled by English children and corresponding increases in distances travelled by car (DiGuiseppi et al. 1997). The other side of the coin is the increase in time spent in ‘activities’ requiring little more than basal energy expenditure. School-aged children in the United States spent a mean of over 20 h/week watching television in the late 1980s (AC Nielson Company 1990, cited in Robinson et al. 1993).
Data, of known validity, on time trends in energy expenditure is generally unavailable for developed populations. Because of the technical difficulties involved, the measurement of total energy turnover in representative free-living individuals is a major challenge for contemporary public health surveillance. The ‘doubly-labelled water’ technique provides a gold standard but is too expensive for large-scale use. Individually calibrated heart rate monitoring is the next best but so far only one study has reported findings from a broadly representative study population (Wareham et al. 1997).
In the absence of data on trends in energy expenditure over time, data on recorded energy consumption may be used as a proxy, bearing in mind that such records tend to underestimate true intake and that an assumption of a roughly constant under-reporting bias over time is required. Data for English adults show substantial declines since the first 7-day weighed dietary intakes of the 1930s (Widdowson 1936; Widdowson and McCance 1936; Bingham et al. 1981; Prentice and Jebb 1995). Data abstracted from a large series of dietary studies in the United States show falls of around 17 per cent in recorded energy consumption of American adults (without adjustment for increasing body weight) between the 1940s and the early 1980s (Stephen and Wald 1990).
The Physical Activity Level is the ratio of total energy expenditure to basal metabolic rate (James and Schofield 1990). It is an important determinant of public health via two types of effect. Firstly, activity is directly protective of health (independently of its effects on body composition and of its contribution to aerobic fitness) (US Department of Health and Human Services 1996; Wareham et al. 2000). Secondly, as activity levels decline, the prevalence of obesity increases (Prentice and Jebb 1995).
The most dramatic evidence of the scale of public health problems associated with declining energy expenditure is the rise in obesity in most rich countries. The prevalence of obesity, defined as body mass index (weight in kilograms over height in meters squared) of 30 or more, doubled—from 10 to 20 per cent—in American men between 1960 and the early 1990s, with a particularly rapid increase over the latter part of this interval. In American females the absolute increase was similar but the proportional increase less, with a recent prevalence of around 25 per cent. In English men prevalence jumped from 6 to 15 per cent and in English women from 8 to 16.5 per cent between 1980 and 1995. Elsewhere trends have been less marked: in Dutch men prevalence has risen only modestly to 8.5 per cent since the mid-1980s and there has been no clear increase in Dutch women (WHO 1998). Whilst a high proportion of fat in the diet may also be contributing to obesity in rich countries (Jebb 1997), the recent secular rise in obesity in the United States has occurred concurrently with a decline in the proportion of dietary energy coming from fat (Stephen and Wald 1990).
The adverse effects of obesity are reasonably well quantified because body mass index is easily and reliably measured. In rich populations where adult deaths from infection are uncommon and a high proportion of deaths are from vascular causes, death rates among the obese are roughly double those for people of desirable weight for height (body mass index 20 to 25). A recent American review of physical activity and health concluded that ‘health benefits appear to be proportional to the amount of activity’ and noted that a quarter of American adults are ‘not active at all’ (US Department of Health and Human Services 1996).
In summary, recent declines in adult mortality in rich countries have, in most cases, been occurring in spite of adverse trends in one important health determinant. Declining physical activity merits recognition as a major unsolved public health problem.
The difficulty faced in solving this problem may be compared with that of changing the composition of the diet in order to favour health. Although there are hedonistic attractions in unhealthy dietary elements such as chocolate and ice cream, attractive alternatives that favour health are also available (for example, Mediterranean diets). However, exertion is not as attractive as indolence. During our evolutionary past, there was unlikely to have been a survival advantage in exertion in the absence of hunger or other immediate need. In this light, the origins of our problems with obesity and inactivity are profoundly social: they are a consequence not so much of individual misbehaviour as of our collective transformation of the way we provision society and the resulting marked reduction in the need for muscular exertion. However, it is still true that, within the material culture of late industrialism, some become obese while others do not and this leaves plenty of room for fatness to be attributed to moral failure. The individuals who find it hardest to avoid obesity are those who are genetically susceptible (Stunkard et al. 1986). Although no one can exercise moral responsibility in choosing their genes, those who become obese are heavily stigmatized. Some indication of the extensive efforts that citizens of rich countries are obliged to make to avoid becoming obese is that 64 per cent of American men and 78 per cent of American women reported in 1996 that they were making conscious efforts to lose or control weight (Serdula et al. 1999). (The desire to be thinner will have arisen from a mixture of concerns with both personal appearance and with health.) Given the apparent difficulties in achieving these objectives, it is not surprising that those who are most socially disadvantaged have the least success. The problems created by low energy turnover are thus also intimately connected with social inequalities in disease burdens.
Possible solutions to physical inactivity and obesity
Investment in new knowledge is a clear priority. Strategic importance, and the indications that solutions will not be easy, both indicate the need to establish physical activity and energy balance as high priorities in public health research. Given the rapid advances in identifying genetic susceptibility, preventive strategies will be needed at all levels—universal (for the whole population), selective (for the susceptible), and targeted (for those already affected) (WHO 1998). Assuming further research confirms the fundamental importance of low physical activity levels, the most feasible and attractive ways of increasing such activity will need to be found (Blair et al. 1996). The implications for institutional change are discussed further below.
Unsolved problems: sustainability
Averting harm to health from the disruption of the ecological systems on which human well being depends is unlike other public health challenges. The need for action cannot be inferred from empirical observations of previous harm from this source, but rather is to be inferred from highly uncertain models of what may happen in the future. Those who prefer to base decisions on ‘science’ rather than ‘speculation’ might be inclined to defer judgement until there has been time for the relevant models to be more thoroughly challenged, and the contributory evidence better marshalled. The argument against delay is that the interacting momenta of population increase and economic development are likely to result in ‘overshoot and collapse’ unless corrective action begins now. To a projected approximate doubling of the current world population one might add, for illustrative purposes, the fivefold increase in global income that would result from levelling world incomes (currently averaging about $5000/person/year) up to current rich country levels (around $25 000/person/year). Without any reduction in energy and resource use per unit of economic product this would translate to a 2 × 5 = 10-fold increase in the global ‘material economy’. Even allowing for the likelihood of more efficient use of materials and energy, the point remains that it is not the ecological sustainability of the current global ‘material economy’ that is in question, but that of a several-fold multiple of it.
If attempts to extend the current pattern of energy and resource-intensive industrialism to the whole of an increasing human population are likely to come seriously unstuck, then the sooner the transition to more sustainable material culture is begun the better. A case for scaling down the quantum of ecological disruption per unit of economic product can be based more specifically on:
evidence that productive activities are already adversely effecting major global meteorological systems, for example ‘global warming’ and stratospheric ozone depletion;
evidence that the absolute scale of rich country material usage is unlikely to be generalizable;
time trends in measurable effects on ecosystems.
In order to limit the carbon dioxide build-up to a doubling of its preindustrial concentration (that is, from 275 to 550 ppm.)—a level which climatologists think would be tolerable—with a population of 10 to 11 billion by 2100, carbon dioxide emissions per person would need to be reduced to the levels of the 1920s (Wigley et al. 1996), a reduction of approximately two-thirds from today’s level. The 1997 Kyoto Protocol was for an average 7 per cent cut, and restricted to developed countries.
Use of materials and absorption of wastes
Citizens of Germany, Japan, The Netherlands, and the United States use between 45 and 85 metric tons of materials per year. A majority of this is ‘hidden’ from commerce, for example mine tailings and soil erosion, and much of it occurs off shore: ‘More than 70 per cent of the materials that flow through the Dutch economy….never touch Dutch soil’ (World Resources Institute 1998). The kind of resource-intensive production that is commonplace in developed countries probably cannot be replicated in a large number of other countries without causing serious environmental harm. A target of halving the rate of global materials used in the coming decades while leaving room for economic development in low- and middle-income countries has been estimated to require a 90 per cent reduction, in rich countries, in the materials used per unit of economic product. Although this ‘factor 10’ reduction has been adopted as an objective by the Organization for Economic Co-operation and Development (Adriaanse et al. 1997), the magnitude of the transformations in productive technologies and consumption habits entailed do not appear to have penetrated far into either public or business opinion.
A related guide to the sustainability has been calculated as the amount of ‘average’ Earth’s surface needed to ‘produce the resources consumed and to assimilate the wastes generated by that population on a continuous basis’. Rees and others have estimated that most high-income countries have an ecological footprint several times larger than their national territories and that the total world population exceeds global carrying capacity by up to one-third (Rees 1999).
Effects on ecosystems
The ‘living planet index’ is one of the first systematic attempts to quantify the effects of human activity on natural ecosystems. It gives equal weight to three contributing indices: forest ecosystems (area of natural forest cover), freshwater ecosystems, and marine ecosystems (trends in the populations of 70 and 87 indicator species respectively). Set to 100 in 1970, it was estimated to have fallen to 68 by 1995—a clearly unsustainable rate of decline (Loh et al. 1998).
Ecological sustainability is an important issue for public health professionals: firstly, because it concerns the biological basis of human well being, whether or not the specific adverse health effects can be predicted with any confidence at this time; secondly, because ‘organised efforts’ will be needed to optimize health outcomes within ‘ecologically-constrained’ material economies; and thirdly, because the rethinking about economic life that is flowing from a concern with sustainability provides important leads to new ways of thinking about the connections between economic life and health. These are discussed further below.
Some issues emerging from examples considered
At the opening of this chapter seven public health topics were selected for further discussion. They were chosen to represent a variety of challenges to public health policy development, in the hope that a consideration of them would highlight important issues in this field. No claim to comprehensiveness is advanced. Four interim conclusions will be recapitulated before taking up a broader theme.
Governments may be more concerned to protect their reputations in the eyes of the press (and other powerful institutions) than to implement measures with high public support and dramatically favourable cost–benefit ratios (for example, fluoridation as an administrative measure to protect health).
Enhanced coverage with preventive measures applied to individuals appeals to doctors but may, in many circumstances, offer only modest gains in health (for example, the control of hypertension, illustrating the ‘prevention paradox’).
Formal programmes to promote change to healthier ways of life may have small (but still worthwhile) effects compared with the informal processes promoting such changes but both formal and informal processes depend critically on new knowledge (US Department of Health and Human Services and Batelle 1995). Investment in new knowledge is therefore the most fundamental component of public health policy (for example, changes from sexual behaviours associated with HIV transmission, changes from infant care practices associated with sudden infant death, changes from high-risk driving practices, and cessation of cigarette smoking).
Combinations of regulatory measures (including taxation) and persuasion are likely to be more effective in changing behaviour than the latter alone, but these are only likely to be politically feasible where there is widespread public appreciation that stronger measures are needed if valued health gains are to be secured (for example, traffic injury reduction and smoking reduction).
The broader theme, to be taken up now, concerns the institutional underpinning of health protection within a given social order. Its importance was suggested in the above discussion of cigarette smoking and in the discussion of physical activity and ecological sustainability. Current discussion of the role of ‘social capital’ in the processes of economic and social development is taken as a starting point (Dasgupta and Serageldin 2000).
The inability of quantitative predictors to account for the dramatic successes of the East Asian economies, or for the equally dramatic failures of economic transition in Russia, has focused attention on the institutional sources of these phenomena. In bodies such as the World Bank, attention is switching from a critique of the state as a displacer of markets to a desire to better understand why large-scale organizations function well in some circumstances but not in others, and how the development and effectiveness of state organizations can influence patterns of economic and social development (World Bank 1997, 1999; Dasgupta and Serageldin 2000). This interest has been sharpened by the generally poor performance of state institutions in sub-Saharan Africa and by the dire consequences of the collapse of state institutions in several ‘global trouble spots’.
Economic collapse in Russia has been associated with perhaps the most severe deterioration in public health yet experienced in the industrialized world. Its institutional origins are therefore of interest to our theme, even if they remain poorly understood. Richard Rose has characterized contemporary Russia as an ‘antimodern’ society ‘characterised by organisational failure and the corruption of formal organisations’ (Rose 2000). Modern societies, by contrast, are to be distinguished by ‘the predominance, in both the market and the state sectors, of social capital in the form of large, impersonal bureaucratic organisations operating according to the rule of law [citing Weber], such as IBM, commercial airlines, social security agencies, and universities. Even though informal networks can supplement or at times substitute for formal bureaucratic organisations, in modern society they are of much less importance than in a traditional or premodern society’. Soviet Russia was simultaneously ‘overorganised’ and ‘underbureaucratised…in that the rule of law did not apply and the system encouraged people to create informal networks as protection against the state and to circumvent or subvert its commands’. This interpretation follows Max Weber in seeing bureaucratic rationalization as essential to the effective functioning of modern societies and is in strong contrast to the sociologically naive view that ‘bureaucracy’ is an unattractive characteristic of state organizations that can be dispensed with by privatization. (The attractiveness of bureaucracy is really a side issue. Weber himself was pessimistic: ‘Not summer’s bloom lies ahead of us, but rather a polar night of icy darkness and hardness, no matter which group may triumph externally now’ (Weber 1991).)
The conclusions drawn by Putnam et al. (1993) from their detailed and prolonged investigation of the establishment of a regional tier of government in Italy have been particularly influential in the social capital literature. They found that the regional governments established in the early 1970s worked well in the north and badly in the south, despite their identical structures and equivalent legal and financial resources. The regional characteristics most closely associated with effective government were not those indicative of economic development but rather those indicative of a strong ‘civic community’: the empirical measures used were voting behaviour (including turnout, not preferences), newspaper readership, and density of sports and cultural associations. In the ‘civic’ regions, ‘the community values solidarity, civic engagement, cooperation, and honesty. Government works’. The authors traced the origins of these different institutional inheritances back to the emergence of republican city governments in the late Middle Ages. In this interpretation, inherited stocks of social capital are important determinants of the good government and economic well being of today’s citizens.
Against the background of these recent debates, there is clearly no need to be defensive or apologetic about the need to move beyond the quantitative evidence to discuss the institutional requirements for health protection (Breslow 1996). However, it should be acknowledged that public health strategies now need to be framed within a more strongly liberal (European sense) political culture in which respect for government is currently much less than it was in the early years after the Second World War. (Giddens (1998) cites a drop in the United States from 76 per cent in 1964 to 25 per cent in 1994 in the proportion of opinion poll respondents answering ‘all’ or ‘most of the time’ when asked ‘How much of the time do you trust the government in Washington to do the right thing?’.) A related development is the shift in power away from national governments to supranational bodies above and regional governments below (World Bank 1999).
The three examples used to justify this detour to the social capital literature were tobacco smoking, inactivity, and ecological sustainability.
It was noted above that quantitative analyses of the discounted present value of future health gains from increasing smoking cessation versus reducing smoking uptake would push policy emphases strongly towards the former. However, this matter can be approached differently by thinking of a society’s institutional defences against tobacco smoking as part of its stock of ‘health capital’ and as a valuable asset to be accumulated and transmitted to future generations. To minimize the cumulative toll of tobacco smoking (the area under the epidemic curve) each generation will need to transmit strengthened institutional defences against cigarette smoking to the next, based in a sound popular understanding of its health risks, and including a strong disapproval of tobacco smoking. Such a process might be less likely in a society relying heavily on smoking cessation because most smokers who quit do so in middle-age. The intergenerational transmission of nicotine addiction is sensitive to smoking prevalences in young parents, and this will reflect both the general strength of tobacco control activities and, more specifically, the strength of measures directed to adolescents and young adults (Distefan et al. 1998; Farkas et al. 1999). Nor is it necessary to be too pessimistic about the responsiveness of adolescents to tobacco control measures: smoking initiation rates among American males aged 15 to 20 years halved between the 1950s and the 1980s and fell substantially in females after the mid-1970s (Gilpin et al. 1994). (Among American black adolescents, smoking had almost gone out of fashion by the late 1980s (McIntosh 1995). In respect of school-based programmes, those concentrating on teaching relevant social skills, rather than knowledge and norms, have generally been effective (US Department of Health and Human Services 1994).)
The problem of declining physical activity levels is embedded in the everyday realities of late industrialism. ‘Labour-saving’ investments have increased productivity and profits; they are unlikely to be abandoned. Scope will still need to be sought for inserting more activity back into the working day. Patterns of commuting to work reflect the patterns of investment in residential settlements and in transport infrastructure. Sprawling cities in North America and Australasia have dependence on the car ‘built in’. Such massive investments cannot suddenly be undone. Moves towards walking and cycling will require a widespread public recognition that increasing daily activity is an inescapable requirement if optimum health levels are to be attained in rich countries. Without such recognition, the radical changes required will not be politically feasible. The health problems generated by declining physical activity levels in the late twentieth century thus bear this similarity to the increased transmissibility of infection recognized as a consequence of rapid urbanization in the second quarter of the nineteenth century. Effective solutions are likely to require an adaptive reconfiguration of urban life and government. Walking and cycling will need to be made more attractive and this is unlikely to happen without substantial investments in infrastructure (McCarthy 1999). The trend towards driving children to school is unlikely to be reversed unless parents can be more confident that their children are safe on the streets. Waking time spent at very low levels of physical activity, for example watching television, appears to be especially predictive of weight gain, at least in adolescents (Kimm et al. 1999). If children are to be diverted to outdoor play they will need attractive and secure environments.
Just as new competencies, born of new policies, were needed by local governments in order to protect nineteenth century urban dwellers from infection transmitted by water, food, and urban crowding (Szreter 1988), so is it likely that another renewal of local government will be required to foster activity patterns that are optimal for health. Via their connections with energy and resource use, there is a natural bridge between public health issues related to physical activity levels and those related to sustainability.
[Society is a partnership] not only between those who are living, but between those who are living, those who are dead and those who are to be born.
Edmund Burke, Reflections on the revolution in France, 1790 (cited in Giddens 1998)
To operate the idea of sustainability within economic analyses, attention needs to be diverted from measures of ‘flow’ (income) to measures of ‘stock’ (capital), including natural capital and human resources. Conventional national income (‘flow’) accounts are biased for this purpose in that they treat depletions of natural capital as income (as when a forest is cut down to make furniture), and are insensitive since they provide little indication of future legacies. The Environment Department of the World Bank has noted that if ‘sustainable development is about leaving future generations more capital per capita than we have had, then the rate of genuine saving becomes a good measure of whether our aggregate activities are on a sustainable path’ (Serageldin 1996). Genuine saving is evidenced by increases in a proposed broad measure of wealth combining the estimated values of natural and human resources along with those of produced assets (capital as traditionally considered). Human resources include the ‘human capital’ embodied in individuals (for example, health and education levels) and ‘social capital’ embodied in institutions, customs, and knowledge (Environment Department World Bank 1998).
Such a shift of emphasis from economic flows to stocks, especially of human and natural resources, helps to clarify the scope for enhancing health at any given level of income—an important objective if we are to maximize well being while minimizing flows of materials and energy. It is also in tune with the consistent empirical finding that human and social resources play a more important role in determining mortality levels than income. For rich countries, health evidently depends less on the consumption opportunities provided by income than on personal and social capacities to protect and enhance health.
Social institutions and health: psychosocial versus material emphases
In direct contrast to German philosophy, which descends from heaven to earth, here we ascend from earth to heaven.
Marx and Engels, The German ideology, 1846 (1959)
Much of the literature on ‘social determinants of health’ has an emphasis on ‘psychosocial’ phenomena—stress, the psychosocial environment at work, relative income, and social networks and support (see for example many of the contributions in Marmot and Wilkinson 1999)—and the role of such phenomena in generating inequalities in health. A shortcoming of this work is its inadequate recognition of the importance of major health determinants that are simultaneously social and material, that is, of phenomena that might alternatively be described as components of material culture. Mediterranean food cultures and the Russian drinking culture can be cited as examples of phenomena which are clearly anchored to the institutional inheritances of certain cultural regions (and therefore qualify as social rather than natural phenomena) and which are exerting powerful effects on recent mortality trends.
Being born into a Mediterranean culture might plausibly have a greater influence on one’s risk of premature death from coronary disease than differential access to modern medical care or discretionary individual behaviours such as exercising or even smoking (Willett et al. 1995). It is striking that Albania, by far the poorest country in Europe but bordering the Mediterranean, has adult mortality levels substantially below those of richer ex-socialist states further east and north (Gjonca and Bobak 1997).
The catastrophic increase in adult mortality that befell Russia between the mid-1980s and around 1994—in contrast, for example, to Georgia or Armenia, which have suffered even greater falls in income (World Resources Institute 1998)—appears to be partly the result of a dramatic increase in binge drinking (Leon et al. 1997). This has clearly had important short-term determinants, including the political unsustainability of Gorbachev’s anti-alcohol campaign and the deregulation of sales during economic liberalization, but the outcome is only explicable if local drinking customs are taken into account (White 1996).
This dependence of an individual’s health prospects on their cultural inheritances is akin to the findings of Putnam et al. (1993), in the political domain, that the chances of being well governed and prosperous may depend substantially on place of birth. The effect of these inheritances is not static, but is continually projected forward by evolving material possibilities arising from economic development and by dynamic social processes. The rapid evolution of eating and drinking habits, for example, powerfully influences health trends. Mediterranean countries such as France, Italy, and Spain have enhanced their advantage in mortality levels from vascular causes as they have become more affluent, despite their rising fat consumption (Powles and Sanz 1999).
Public health institutions are inevitably concerned with changing consumption norms. The institutions that normatively regulate consumption are fundamental to the regulation of social and political life itself—’a society’s notion of political order is a result of its evaluation of desire and identification of need (that which is socially necessary)’ (Berry 1994). Since the eighteenth century, the dominant liberal idea has been that desires should be respected as the ‘authentic voice of individual preference’. But not all: most jurisdictions make it illegal to use heroin and most place heavy (‘luxury’) taxes on tobacco. Contention about the acceptability of certain desires and the legitimacy of public action to suppress their fulfilment is close to the surface in many debates on public health policy.
Public health problems and public health investments
The opening discussion—questioning whether one can see in health trends, the effects of public health programmes—presumes that the programmes are always the cause and the health trends the effects. This is, of course, much too simple. Causes could, and probably often have, worked in the opposite direction, where it is the nature and magnitude of health problems experienced that has determined the strength of the public health response. Finland and Australia provide cases in point.
In the late 1960s mortality from vascular disease in middle-aged males in the Finnish province of North Karelia was far above levels in other developed countries and the risk of dying before the age of 65 approached 50 per cent. These risks were perceived, by the local people, as unacceptable and so they petitioned the national government to mount a preventive programme. From this the North Karelia Project was born and it, in turn, stimulated investment in public health institutions (Vartiainen et al. 1994). Despite its modest population, Finland now has over 700 staff in its National Institute of Public Health and it ranks at the top in its rate of publication in leading international epidemiological journals (Fig. 5). The fact that this substantial investment has not been sufficient to push Finland to the fore of ‘good performers’ in Fig. 1 (it lies just below the trend line) may mainly be due to the intrinsic difficulties Finns face in lowering their vascular mortality. Without such investment, Finland may have done substantially worse. Evaluative studies of the North Karelia Project itself suggest it was effective. Lung cancer mortality fell sooner there and has fallen further than elsewhere in Finland consistent with early results for changes in smoking prevalence (Puska et al. 1993). Conversely, the very poor performance of Denmark (Fig. 1) contrasts with its apparently heavy investment in public health science (Fig. 5), suggesting that science does not always find effective application.
Fig. 5 Publications in the American Journal of Epidemiology plus the International Journal of Epidemiology classified by country of author’s address, per million population, average for 1995 to 1998. (Author’s calculations using Medline records.)
Australia, like Finland, faced adverse mortality trends in the first two postwar decades with male life expectancy at birth falling during the 1960s. Death rates rose from coronary heart disease, car crashes, and suicide. A country which had been notable for its favourable mortality levels at the beginning of the century, which had experienced a long postwar economic boom, and which thought of its way of life as being especially favourable to health had to come to terms with a serious loss of rank in international health comparisons. Strong institutional responses evolved in relation to traffic injuries (discussed above), heart disease, and tobacco control (Powles and Gifford 1993). Through the 1980s there was a rapid expansion of masters level programmes in public health disciplines, with enrolment levels exceeding those in most (if not all other) high-income countries. These institutional developments have plausibly contributed to Australia’s recent ranking as a relatively good performer in reducing overall mortality (Fig. 1).
The strong development of public health institutions oriented towards chronic disease control in countries such as Finland and Australia may be contrasted with experience in countries such as France, Italy, and Spain where the evolving nature of public health challenges has been different. Vascular mortality in these countries did not persist at high levels, but tended to fall, often rapidly; this brought down all-cause mortality rates and made the case for reinvigorating public health institutions to prevent less pressing chronic disease. In these countries traffic injuries, HIV, tobacco control, and the reduction of harm from alcohol use are among the most salient challenges.
The search for equality
Recent favourable trends in overall adult mortality have been accompanied by growing inequalities in states such as the United Kingdom, because mortality declines have been much greater in more favoured strata. In England and Wales, between 1977–1981 and 1987–1991, male life expectancy at 15 years of age increased from 58.8 to 60.5 years for people in professional and managerial occupations but only from 55.1 to 55.8 years for people in semiskilled and unskilled jobs; for females the respective trends were from 64.1 to 65.8 years and from 62 to 62.4 years (cited in Shaw et al. 1999). The causes of, and appropriate remedies for, these inequalities in health have been a major preoccupation in public health policy discussions in the United Kingdom (Black et al. 1980; Acheson 1998).
Comparisons of mortality differentials between males in manual and non-manual occupations across 11 European countries found them to be broadly similar (though somewhat greater in France and Finland) with ‘no evidence that mortality differences are smaller in countries with more egalitarian socio-economic and other policies’ (Kunst et al. 1998a). The causes of death contributing most to these differences did, however, vary markedly between countries: ‘mortality from ischaemic heart disease was strongly related to occupational class in England and Wales, Ireland, Finland, Sweden, Norway, and Denmark, but not in France, Switzerland, and Mediterranean countries. In the latter countries, cancers other than lung cancer and gastrointestinal diseases made a large contribution to class differences in total mortality. Inequalities in lung cancer, cerebrovascular disease, and external causes of death also varied greatly between countries’ (Kunst et al. 1998b).
Black et al. (1999) have articulated a ‘materialist’ interpretation of the cause of inequalities in the United Kingdom. This gives primacy to ‘material deprivation’ (both absolute and relative) and draws attention to the marked increase in income inequalities there between 1979 and 1995–1996; for example, over this period the number of people living in households with less than half the national average income increased from 4.5 to 12.2 million. However, the finding of Kunst et al. (1988a) that relative mortality inequalities are not less in countries with more equal income distributions does not support this interpretation. Furthermore, constrained consumption opportunities are not everywhere associated with high mortality levels: impoverished Cuba has lower adult male mortality than the United States, and Sri Lanka has adult male mortality levels comparable to some European countries (WHO 1999). Within Europe, Cretan villagers observed in the 1960s and 1970s in the Seven Countries Study had favourable mortality levels, despite their extremely frugal circumstances (Keys 1980). Thus, within some material cultures (all of which seem to have warm climates) it has become possible to attain low mortality on low incomes. The health effects of limited consumption opportunities therefore appear to depend strongly on the context in which consumption choices are made. Materialist explanations, if they are to be persuasive, need either to acknowledge their limited sphere of applicability (‘northern commodity-intensive cultures’?) or, more informatively, to incorporate explicit reference to the kinds of differences between life in a Cretan village in the 1960s and life on housing estates in a British industrial city in the 1990s that are likely to be most important for health: for example, dietary traditions (related also to local food-producing possibilities), norms governing alcohol and tobacco use (and purchasing power for cigarettes), and obligatory daily energy expenditure. (The greater absolute poverty of the Cretans probably protected them from tobacco-attributable disease by limiting their tobacco consumption, especially prior to the Second World War.)
Given that the relative importance of causes contributing to mortality inequalities varies by country, responses should also differ. In France, where inequalities in males appear to be the greatest within Western Europe, chronic diseases related to the volume of alcohol consumed make a major contribution; policies to reduce consumption will therefore be important. In Finland, injuries related to drunkenness are more salient, indicating the need both for ‘harm-reduction’ policies (such as control of drunk driving), and for programmes to encourage a change away from the ‘peak drinking’ pattern. Measures to counter smoking are of primary importance in countries where a mature smoking epidemic is combined with a high background risk of vascular disease (roughly the ‘northern’ countries). (In countries at earlier stages in their smoking epidemics, programmes to encourage quitting may have the effect of increasing mortality inequalities. This does not mean that they should not be implemented.)
Jarvis and colleagues have shown for the United Kingdom that the current social gradient in smoking prevalence has been mainly created by greater rates of smoking cessation in the upper social strata: ‘What we need to explain above all is not so much why poor people start smoking, but why they do not give it up’ (Jarvis and Wardle 1999). Plasma cotinine levels among smokers show that nicotine dependence increases systematically with deprivation and that poor smokers obtain more nicotine per cigarette smoked. Using the indirect method of Peto et al. (1992) (in which lung cancer mortality is used as a measure of tobacco exposure to estimate the proportion of other deaths attributable to smoking), it is estimated that, in the United Kingdom, smoking-attributable deaths contribute about two-thirds of the excess mortality in the less favoured groups. The most obvious short-term policy response is to provide even greater assistance for quitting—including free or subsidized supplies of nicotine replacement therapy. In the long term, all measures that contribute to making tobacco use uncommon will have helped to reduce a major actual or potential cause of health inequality.
Making progress safe
Material progress both favours and harms health. It has been one of the main responsibilities of public health institutions to help resolve this ambivalence by countering the manifest and potential harms to health arising from material progress. This has enabled the net effect of affluence on health to approximate more closely towards its beneficial effect. In nineteenth-century Britain, industrialism did not impress as ‘progress’ until ways had been found to control the increase of fatal infection in the new industrial towns (Szreter 1997). In the twentieth century, the increased consumption opportunities generated by economic development has permitted a global epidemic of nicotine addiction which, especially when combined, in susceptible food cultures, with ‘early dietary affluence’, resulted in epidemic waves of tobacco-caused cancer and tobacco-amplified vascular disease. These epidemics were sometimes large enough, at least in males, to nullify substantially the beneficial effects of economic development on traditional infective killers of adults such as tuberculosis and pneumonia. Today, in many developed countries, these two related epidemics are in retreat. However, challenges and unsolved problems are ever renewed. As noted above, the uptake of tobacco smoking by young people has ceased declining in many developed countries and no plausible solution to the rising prevalence of obesity is in sight. The sustainability of industrialism in its current form, once it is generalized to the whole human population, is improbable. Although we cannot predict the exact ways in which the cumulative disruption of major ecological processes will rebound on our health, the likelihood of serious harm from this source is now substantial. Public health endeavour will continue to be an important determinant of what we are able to mean by ‘progress’ and of whether we shall be able to make it safe.
*The author is grateful to Nick Day for the examples relating to HIV transmission and the decline in sudden infant death syndrome, and to Daniela de Angelis for the model data on HIV in England.
AC Nielson Company (1990). Nielson report on television. Nielson Media Research, New York.
Acheson, E.D. (1993). Behold a pale horse: a view from Whitehall. PHLS Microbiology Digest, 10, 133–40.
Acheson, E.D. (1998). Independent inquiry into inequalities in health. HMSO, London.
Adriaanse, A., Bringezu, S., Hammond, A., et al. (1997). Resource flows: the material basis of industrial economies. World Resources Institute, Washington, DC; Wuppertal Institute; Netherlands Ministry of Housing, Spatial Planning, and Environment; National Institute for Environmental Studies, Japan.
American Dental Association (1999). Fluoridation facts (revised). American Dental Association, Chicago, IL.
Beal, S. (1988). Sleeping position and SIDS. Lancet, ii, 512.
Berry, C.J. (1994). The idea of luxury: a conceptual and historical investigation. Cambridge University Press.
Bingham, S.A., McNeil, N.I., and Cummings, J.H. (1981). The diet of individuals: a study of a randomly chosen cross section of British adults. British Journal of Nutrition, 45, 23–35.
Black, D., Morris, J.N., Smith, C., and Townsend, P. (1980). Inequalities in health: report of a research working group. Department of Health and Social Security, London.
Black, D., Morris, J.N., Smith, C., and Townsend, P. (1999). Better benefits for health: plan to implement the central recommendation of the Acheson Report. British Medical Journal, 318, 724–7.
Blair, S.N., Booth, M., Gyarfas, I., et al. (1996). Development of public policy and physical activity initiatives internationally. Sports Medicine, 21, 157–63.
Bonita, R. and Beaglehole, R. (1989). Increased treatment of hypertension does not explain the decline in stroke mortality in the United States, 1970–1980. Hypertension, 13, 169–73.
Breslow, L. (1996). Social ecological strategies for promoting healthy lifestyles. American Journal of Health Promotion, 10, 253–7.
British Dental Association (1996). Oral health, tooth decay and the need for water fluoridation (Parliamentary Briefing). British Dental Association, London.
Bunker, J.P. (1995). Medicine matters after all. Journal of the Royal College of Physicians of London, 29, 105–12.
Burt, V.L., Culter, J.A., Higgins, M., et al. (1995). Trends in the prevalence, awareness, treatment, and control of hypertension in the adult United States population. Data from the health examination surveys, 1960 to 1991 Hypertension, 26, 60–9. (Erratum, Hypertension, 27, 1192 (1996).)
Cameron, M. and Newstead, S. (1996). Mass media publicity supporting police enforcement and its economic value. Monash University Accident Research Centre, Melbourne.
Centers for Disease Control (1999). Fluoridation of drinking water to prevent dental caries. Morbidity and Mortality Weekly Report, 48, 933–40.
Collins, R., Peto, R., MacMahon, S., Hebert, P., et al. (1990). Blood pressure, stroke, and coronary heart disease. Part 2: Short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet, 335, 827–38.
Dasgupta, P. and Serageldin, I. (2000). Social capital: a multifaceted perspective. World Bank, Washington, DC.
De Angelis, D., Gilks, W.R., and Day, N.E. (1998). Bayesian projection of the acquired immune deficiency syndrome epidemic. Applied Statistics, 47, 449–98.
Department of the Environment, Transport and the Regions (1998). Road accidents Great Britain 1997: the casualty report. HMSO, London.
DiGuiseppi, C., Roberts, I., and Li, L. (1997). Influence of changing travel patterns on child death rates from injury: trend analysis. British Medical Journal, 314, 710–13.(Erratum. British Medical Journal, 314, 1385.)
Distefan, J.M., Gilpin, E.A., Choi, W.S., and Pierce, J.P. (1998). Parental influences predict adolescent smoking in the United States, 1989–1993. Journal of Adolescent Health, 22, 466–74.
Dwyer, T. and Ponsonby, A.L. (1996). The decline of SIDS—a success story for epidemiology. Epidemiology, 7, 323–5.
Environment Department, World Bank (1998). Expanding the measure of wealth: indicators of environmentally sustainable development. World Bank, Washington, DC.
Farkas, A.J., Distefan, J.M., Choi, W.S., Gilpin, E.A., and Pierce, J.P. (1999). Does parental smoking cessation discourage adolescent smoking? Preventive Medicine, 28, 213–18.
Garry, R.C., Passmore, R., Warnock, G.M., and Durnin, J.V.G.A. (1955). Studies on expenditure of energy and consumption of food by miners and clerks, Fife, Scotland, 1952, pp. 1–69. HMSO, London.
Giddens, A. (1998). The third way. The renewal of social democracy. Polity Press, Cambridge.
Gilpin, E.A. and Pierce, J.P. (1997). Trends in adolescent smoking initiation in the United States: is tobacco marketing an influence? Tobacco Control, 6, 122–7.
Gilpin, E.A., Lee, L., Evans, N., and Pierce, J.P. (1994). Smoking initiation rates in adults and minors: United States, 1944–1988. American Journal of Epidemiology, 140, 535–43.
Gjonca, A. and Bobak, M. (1997). Albanian paradox, another example of protective effect of Mediterranean lifestyle? Lancet, 350, 1815–17.
Goodman, J. (1993). Tobacco in history: the cultures of dependence. Routledge, London.
Harkin, A.M., Anderson, P., and Goos, C. (1997). Smoking, drinking and drug taking in the European Region. WHO Regional Office for Europe, Copenhagen.
Hawthorne, G. (1991). Pre-driver education. An evaluation of a traffic safety education program for senior students in Victorian post-primary schools. PhD Thesis, Monash University, Australia.
Hendrie, D. and Ryan, G.A. (1995). Review of road safety practices in Australia and recommendations for Western Australia. Road Accident Prevention Research Unit, Department of Public Health, University of Western Australia, Perth.
Hiley, C.M.H. and Morley, C.J. (1994). Evaluation of government’s campaign to reduce risk of cot death. British Medical Journal, 309, 703–4.
Hill, D., White, V., and Segan, C. (1995). Prevalence of cigarette smoking among Australian secondary school students in 1993. Australian Journal of Public Health, 19, 445–9.
James, W.P.T. and Schofield, C. (1990). Human energy requirements. Oxford University Press.
Jarvis, L. (1997). Smoking among secondary school children in 1996: England. HMSO, London.
Jarvis, M.J. and Wardle, J. (1999). Social patterning of individual health behaviours: the case of cigarette smoking. In Social determinants of health (ed. M.G. Marmot and R.G. Wilkinson), pp. 240–55. Oxford University Press.
Jebb, S.A. (1997). Aetiology of obesity. British Medical Bulletin, 53, 264–85.
Jha, P. and Chaloupka, F.J. (1999). Curbing the epidemic: governments and the economics of tobacco control. World Bank, Washington, DC.
Joint National Committee (1997). The sixth report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure. Archives of Internal Medicine, 157, 2413–46. (Erratum. Archives of Internal Medicine, 158, 573 (1998).)
Keys, A. (1980). Seven countries: a multivariate analysis of death and coronary heart disease. Harvard University Press, Cambridge, MA.
Kimm, S.Y.S., Glynn, N.W., and Obarzanek, E. (1999). Obesity and its relationship to physical activity and inactivity during adolescence. Abstracts of the 39th Annual Conference on Cardiovascular Disease Epidemiology and Prevention, p. 10.
Kotchen, J.M., McKean, H.E., Jackson-Thayer, S., Moore, R.W., Straus, R., and Kotchen, T.A. (1986). Impact of a rural high blood pressure control program on hypertension control and cardiovascular disease mortality. Journal of the American Medical Association, 255, 2177–82.
Kunst, A.E., Groenhof, F., and Mackenbach, J.P. (1998a). Mortality by occupational class among men 30–64 years in 11 European countries. EU Working Group on Socioeconomic Inequalities in Health. Social Science and Medicine, 46, 1459–76.
Kunst, A.E., Groenhof, F., Mackenbach, J.P., and EU Working Group on Socioeconomic Inequalities in Health (1998b). Occupational class and cause specific mortality in middle aged men in 11 European countries: comparison of population based studies. British Medical Journal, 316, 1636–42.
Lenfant, C. and Roccella, E.J. (1984). Trends in hypertension control in the United States. Chest, 86, 459–62.
Leon, D.A., Chenet, L., Shkolnikov, V.M., et al. (1997). Huge variation in Russian mortality rates 1984–94: artefact, alcohol or what? Lancet, 350, 383–8.
Loh, J., Randers, J., MacGillivray, A., Kapos, V., Jenkins, M., Groombridge, B., and Cox, N. (1998). Living planet report, 1998. WWF International, Gland, Switzerland; New Economics Foundation, London; World Conservation Monitoring Centre, Cambridge.
Lopez, A. (1996). The lung cancer epidemic in developed countries. In Adult mortality in developed countries (ed. A. Lopez), pp. 111–34. Oxford University Press.
Lopez, A.D., Collishaw, N.E., and Piha, T. (1994). A descriptive model of the cigarette epidemic in developed countries. Tobacco Control, 3, 242–7.
McCarthy, M. (1999). Transport and health. In Social determinants of health (ed. M. Marmot and R.G. Wilkinson), pp. 132–54. Oxford University Press.
McIntosh, H. (1995). Black teens not smoking in great numbers. Journal of the National Cancer Institute, 87, 564–5.
MacMahon, S., Peto, R., Cutler, J., et al. (1990). Blood pressure, stroke, and coronary heart disease. Part 1: Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias . Lancet, 335, 765–74.
McMichael, A.J. and Powles, J.W. (1999). Human numbers, environment, sustainability and health. British Medical Journal, 319, 977–80.
Marmot, M. and Wilkinson, R.G. (ed.) (1999). Social determinants of health. Oxford University Press.
Marx, K. and Engels, F. (1959). The German ideology (excerpts) (written 1846, trans. 1939). In Marx and Engels: basic writings on politics and philosophy (ed. L.S. Feuer), pp. 246–61. Doubleday, New York.
Mead, R. (1775). The medical works of Richard Mead, M.D. Alexander Donaldson and Charles Elliot, Edinburgh. (Reprinted by AMS Press, New York, 1978.)
Mosterd, A., D’Agostino, R.B., Silbershatz, H., et al. (1999). Trends in the prevalence of hypertension, antihypertensive therapy, and left ventricular hypertrophy from 1950 to 1989. New England Journal of Medicine, 340, 1221–7.
Murray, C.J.L. and Lopez, A.D. (1996). The global burden of disease: a comprehensive assessment of mortality and disability from diseases, injuries and risk factors in 1990 and projected to 2020. Harvard School of Public Health on behalf of WHO and the World Bank. Harvard University Press, Cambridge, MA.
National Research Council (1993). National Academy of Sciences Committee on Toxicology. Health effects of ingested flouride. National Academy Press, Washington, DC.
OPCS (Office of Population, Censuses and Statistics) (1988, 1995). OPCS Monitor, Series DH3 Sudden Infant Deaths. OPCS, London.
Peto, R., Lopez, A.D., Boreham, J., Thun, M., and Heath, C.J. (1992). Mortality from tobacco in developed countries: indirect estimation from national vital statistics. Lancet, 339, 1268–78.
Peto, R., Lopez, A.D., Boreham, J., Thun, M.J., and Heath, C., Jr (1994). Mortality from smoking in developed countries, 1950–2000: indirect estimates from national vital statistics. Oxford University Press.
Pierce, J.P., Lee, L., and Gilpin, E.A. (1994). Smoking initiation by adolescent girls, 1944 through 1988. An association with targeted advertising. Journal of the American Medical Association, 271, 608–11.
Pierce, J.P., Choi, W.S., Gilpin, E.A., Farkas, A.J., and Berry, C.C. (1998). Tobacco industry promotion of cigarettes and adolescent smoking. Journal of the American Medical Association, 279, 511–15. (Erratum. Journal of the American Medical Association, 280, 422 (1998).)
Pollock, D. (1999). Denial and delay: the political history of smoking and health, 1951–1964. Action on Smoking and Health, London.
Powles, J.W. and Gifford, S. (1993). Health of nations: Lessons from Victoria, Australia. British Medical Journal, 306, 125–7.
Powles, J.W. and Sanz, M.A. (1999). ‘Arcadian bias’ in discussions of the Mediterranean advantage. Proceedings of the 15th International Scientific Meeting of the International Epidemiological Association, p. 37. IEA, Florence.
Prentice, A.M. and Jebb, S.A. (1995). Obesity in Britain: gluttony or sloth? British Medical Journal, 311, 437–39, 1568–9.
Puska, P., Korhonen, H.J., Torppa, J., et al. (1993). Does community-wide prevention of cardiovascular diseases influence cancer mortality? European Journal of Cancer Prevention, 2, 457–60.
Putnam, R.D., Leonardi, R., and Nanetti, R.Y. (1993). Making democracy work: civic traditions in modern Italy. Princeton University Press.
Rees, W.E. (1999). Consuming the earth: the biophysics of sustainability. Ecological Economics, 29, 23–7.
Robinson, T.N., Hammer, L.D., Killen, J.D., et al. (1993). Does television viewing increase obesity and reduce physical activity? Cross-sectional and longitudinal analyses among adolescent girls. Paediatrics, 91, 272–80.
Rose, G. (1985). Sick individuals and sick populations. International Journal of Epidemiology, 14, 32–8.
Rose, R. (2000). Getting things done in an anti-modern society: social capital networks in Russia. In Social capital: a multifaceted perspective (ed. P. Dasgupta and I. Serageldin), pp. 147–71. World Bank, Washington, DC.
Royal College of Physicians (1962). Smoking and health. Pitman Medical, London.
Secretary of State for Health (1992). The health of the nation. HMSO, London.
Secretary of State for Health (1998). Our healthier nation: a contract for health. A consultation paper. HMSO, London.
Secretary of State for Health (1999). Saving lives: our healthier nation. HMSO, London.
Secretary of State for Social Services (1988). Public health in England: the report of the Committee of Inquiry into the future development of the Public Health Function (D. Acheson, Chairman). HMSO, London.
Serageldin, I. (1996). Sustainability and the wealth of nations. First steps in an ongoing journey. Environmentally Sustainable Development Studies and Monographs Series No. 5. World Bank, Washington, DC.
Serdula, M.K., Mokdad, A.H., Williamson, D.F., Galuska, D.A., Mendlein, J.M., and Heath, G.W. (1999). Prevalence of attempting weight loss and strategies for controlling weight. Journal of the American Medical Association, 282, 1353–8.
Shaw, M., Dorling, D., and Davey Smith, G. (1999). Poverty, social exclusion and minorities. In Social determinants of health (ed. M.G. Marmot and R.G. Wilkinson), pp. 211–39. Oxford University Press.
Simon, J. (1890). English sanitary institutions, reviewed in their course of development, and in some of their political and social relations. Cassell, London.
Smeed, R.J. (1972). The usefulness of formulae in traffic engineering and road safety. Accident Analysis and Prevention, 4, 303–12.
Spurgeon, D. (1999). Studies reveal increased smoking among students in Canada. British Medical Journal, 319, 1391A.
Stegmayr, B., Asplund, K., Kuulasmaa, K., Rajakangas, A.M., Thorvaldsen, P., and Tuomilehto, J. (1997). Stroke incidence and mortality correlated to stroke risk factors in the WHO MONICA Project. An ecological study of 18 populations. Stroke, 28, 1367–74.
Stephen, A.M. and Wald, N.J. (1990). Trends in individual consumption of dietary fat in the United States, 1920–1984. American Journal of Clinical Nutrition, 52, 457–69.
Strachan, D. and Rose, G. (1991). Strategies of prevention revisited: effects of imprecise measurement of risk factors on the evaluation of ‘high-risk’ and ‘population-based’ approaches to prevention of cardiovascular disease. Journal of Clinical Epidemiology, 44, 1187–96.
Stunkard, A.J., Sorensen, T.I., Hanis, C., et al. (1986). An adoption study of human obesity. New England Journal of Medicine, 314, 193–8.
Szreter, S. (1988). The importance of social intervention in Britain’s mortality decline c. 1850–1914: a re-interpretation of the role of public health. Journal of the Society for the Social History of Medicine, 1, 1–37.
Szreter, S. (1997). Economic growth, disruption, deprivation, disease, and death: on the importance of the politics of public health for development. Population and Development Review, 23, 693–728.
Tobacco Advisory Group (2000). Royal College of Physicians. Nicotine addiction in Britain. Royal College of Physicians, London.
US Department of Health and Human Services (1994). Preventing tobacco use among young people: A report of the Surgeon General. Serial No. 017-001-00491-0. US Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, Atlanta, GA.
US Department of Health and Human Services (1996). Physical activity and health: a report of the Surgeon General. US Department of Health and Human Services, Centers for Disease Control and Prevention, Atlanta, GA.
US Department of Health and Human Services and Batelle (1995). For a healthy nation: returns on investment in public health. US Government Printing Office, Washington, DC.
US Public Health Service (1964). Smoking and health: report of the advisory committee to the Surgeon General of the Public Health Service. US Department of Health, Education and Welfare, Washington, DC.
Vartiainen, E., Puska, P., Jousilahti, P., Korhonen, H.J., Tuomilehto, J., and Nissinen, A. (1994). Twenty-year trends in coronary risk factors in North Karelia and in other areas of Finland. International Journal of Epidemiology, 23, 495–504.
VicRoads (1993). Road Traffic Accidents Involving Serious Casualties. VicRoads, Melbourne.
Wareham, N.J., Hennings, S.J., Prentice, A.M., and Day, N.E. (1997). Feasibility of heart-rate monitoring to estimate total level and pattern of energy expenditure in a population-based epidemiological study: the Ely Young Cohort Feasibility Study 1994–5. British Journal of Nutrition, 78, 889–900.
Wareham, N.J., Wong, M.-Y., and Day, N.E. (2000). Glucose intolerance and physical inactivity: the relative importance of low habitual energy expenditure and cardio-respiratory fitness. American Journal of Epidemiology, 152, 132–9.
Weber, M. (1991). Politics as a vocation (originally published as Politik als Beruf, 1921, trans. by the editors). In From Max Weber: essays in sociology (ed. H.H. Gerth and M. Wright). Routledge, London.
White, S. (1996). Russia goes dry: alcohol, state and society. Cambridge University Press.
Widdowson, E.M. (1936). A study of English diets by the individual method, part I. Men. Journal of Hygiene, 36, 269–79.
Widdowson, E.M. and McCance, R.A. (1936). A study of English diets by the individual method, part II. Women. Journal of Hygiene, 36, 293–99.
Wigley, T.M.L., Richels, R., and Edmonds, J.A. (1996). Economic and environmental choices in the stabilization of atmospheric CO2 concentrations. Nature, 379, 240–3.
Willett, W.C., Sacks, F., Trichopoulou, A., et al. (1995). Mediterranean diet pyramid: a cultural model for healthy eating. American Journal of Clinical Nutrition, 61, 1402S–6S.
WHO (World Health Organization) (1998). Obesity: preventing and managing the global epidemic. Report of a WHO Consultation on obesity, Geneva 2–5 June 1997. WHO, Geneva.
WHO (World Health Organization) (1999). World Health Report 1999: Making a difference. WHO, Geneva.
World Bank (1997). The state in a changing world: world development report 1997. Oxford University Press for the World Bank, New York.
World Bank (1999). Entering the 21st century: World Development Report 1999/2000. Oxford University Press for the World Bank, New York.
World Resources Institute (1998). United Nations Environment Programme, United Nations Development Programme, and World Bank. World resources 1998–99. Oxford University Press, New York.