A Framework for Public Health Action: the Health Impact Pyramid

COMMENTARIES

7. Kaelber DC, Jha AK, Johnston D, 11. Varon J, Marik PE. Clinical infor- 15. Centers for Disease Control and 19. Pathela P, Harvey K, Blank S, et al. et al. A research agenda for personal mation systems and the electronic med- Prevention. Sexually transmitted diseases The utility of male urethral gram stain for health records (PHRs). J Am Med Inform ical record in the intensive care unit. treatment guidelines, 2006. MMWR informing treatment decisions on the day Assoc. 2008;15(6):729–736. Curr Opin Crit Care. 2002;8(6):616– Recomm Rep. 2006;55(RR-11):1–94. of clinic visit. Paper presented at: 17th 8. Chan KS, Weiner JP. Electronic 624. 16. Centers for Disease Control and Meeting of the International Society for health record-based quality indicators for 12. Shapiro JS, Kannry J, Lipton M, et al. Prevention. Update to CDC’s sexually Sexually Transmitted Disease Research/ ambulatory care: findings from a review Approaches to patient health information transmitted diseases treatment guidelines, 10th International Union against Sexually of the literature. Available at: http:// exchange and their impact on emergency 2006: fluoroquinolones no longer rec- Transmitted Infections World Congress; healthit.ahrq.gov/portal/server.pt/gateway/ medicine. Ann Emerg Med. 2006;48(4): ommended for treatment of gonococcal July 29–August 1, 2007; Seattle, WA. PTARGS_0_3882_217665_0_0_18/ 426–432. infections. MMWR Morb Mortal Wkly 20. Handel S, Schillinger JA, Borrelli J, e-indicator-lit-review.pdf. Accessed January 5, 13. Menke JA, Broner CW, Campbell Rep. 2007;56(14):332–336. et al. STD testing at emergency contra- 2010. DY, et al. Computerized clinical docu- 17. Kuperman GJ, Bobb A, Payne TH, ception visits to local STD clinics. Paper 9. Hillis SD, Owens LM, Marchbanks mentation system in the pediatric inten- et al. Medication-related clinical decision presented at: 2008 National STD Pre- PA, et al. Recurrent chlamydial infections sive care unit. BMC Med Inform Decis support in computerized provider order vention Conference; March 10–13, increase the risks of hospitalization for Mak. 2001;1:3. Available at: http:// entry systems: a review. J Am Med Inform 2008; Chicago, IL. ectopic pregnancy and pelvic inflamma- www.biomedcentral.com/1472-6947/ Assoc. 2007;14(1):29–40. 21. Borrelli J, Paneth-Pollak R, Wright S, tory disease. Am J Obstet Gynecol. 1997; 1/3. Accessed January 5, 2010. 18. Hunt DL, Haynes RB, Hanna SE, et al. The impact of introducing ‘‘express 176(1):103–107. 14. Evans KD, Benham SW, Garrard et al. Effects of computer-based clinical visits’’ for asymptomatic persons seeking 10. Brunham RC, Maclean IW, Binns B, CS. A comparison of handwritten and decision support systems on physician STD services in a busy urban STD clinic et al. Chlamydia trachomatis: its role in computer-assisted prescriptions in an performance and patient outcomes: system, 2005–2006. Paper presented at: tubal infertility. J Infect Dis. 1985;152(6): intensive care unit. Crit Care. 1998;2(2): a systematic review. JAMA. 1998;280(15): 2008 National STD Prevention Confer- 1275–1282. 73–78. 1339–1346. ence; March 10–13, 2008; Chicago, IL. A Framework for Public Health Action: The Health Impact Pyramid

A 5-tier pyramid best de- Thomas R. Frieden, MD, MPH scribes the impact of different types of public health interventions and provides a LIFE EXPECTANCY IN DEVEL- and ascending levels with de- of services used by states to allo- framework to improve health. At the base of this pyramid, oped countries has increased creasing impact that represent cate resources for mothers and 6 indicating interventions with from less than 50 years in 1900 primary, secondary, and tertiary children. Infrastructure building 1 6 the greatest potential impact, to nearly 80 years today. The care. Other frameworks more (e.g., monitoring, training, systems are efforts to address socio- greatest improvement occurred in specific to public health have been of care, and information systems) economic determinants of the first half of the 20th century, proposed. Grizzell’s 6-tier inter- is at the bottom of the pyramid, health.Inascendingorder when life expectancy in the United vention pyramid emphasizes pol- followed by population-based ser- are interventions that change States and many parts of Europe icy change, environmental en- vices (e.g., newborn screening, the context to make individ- increased by an average of 20 hancement, and community and immunization, and lead screening) uals’ default decisions healthy, 2 7 years, largely because of univer- neighborhood collaboration. and enabling services (e.g., trans- clinical interventions that re- sal availability of clean water and Hamilton and Bhatti’s 3-dimen- portation, translation, case man- quire limited contact but con- rapid declines in infectious dis- sional population health and agement, and coordination with fer long-term protection, 3 ongoing direct clinical care, ease, as well as broad economic health promotion cube incorpo- Medicaid), with direct health care and health education and growth, rising living standards, rates 9 health determinants (e.g., services at the top. 4 counseling. and improved nutritional status. healthy child development, biol- All of these models, however, Interventions focusing on Smaller gains in the latter half of ogy and genetics, physical envi- focus most of their attention on lower levels of the pyramid the 20th century resulted primar- ronments, working conditions, and various aspects of clinical health tend to be more effective ily from advances in treatment of social support networks) and evi- services and their delivery and, to because they reach broader cardiovascular disease and control dence-based actions to address a lesser extent, health system in- segments of society and re- of its risk factors (i.e., smoking, them (e.g., reorienting health frastructure. Although these are of quire less individual effort. high blood pressure, and high services, creating supportive envi- critical importance, public health Implementing interventions cholesterol).5 ronments, enacting healthy public involves far more than health care. at each of the levels can The traditional depiction of the policy, and strengthening com- achieve the maximum pos- The fundamental composition, 8 sible sustained public health potential impact of health care munity action). The maternal and organization, and operation of benefit. (Am J Public Health. interventions is a four-tier pyra- child health pyramid of health society form the underpinnings of 2010;100:590–595. doi:10. mid, with the bottom level repre- services, developed by the US the determinants of health, yet 2105/AJPH.2009.185652) senting population-wide interven- Health Resources and Services they are often overlooked in the tions that have the greatest impact Administration, consists of 4 levels development frameworks to

590 | Commentaries | Peer Reviewed | Frieden American Journal of Public Health | April 2010, Vol 100, No. 4 COMMENTARIES

(e.g., poverty reduction, improved Still, more than 900 million peo- education), often referred to as ple worldwide have no access social determinants of health, that to clean drinking water and about help form the basic foundation of 2.5 billion have no access to ade- a society.11,12 Socioeconomic status quate sanitation.21 As the World is a strong determinant of health, Health Organization’s Commis- both within and across countries.13 sion on Social Determinants Although the exact mechanisms of Health reported, ‘‘Social injus- by which socioeconomic status tice is killing people on a grand exerts its effects are not always scale.’’11(p26) apparent, poverty, low educational attainment, relative deprivation, Changing the Context to and lack of access to sanitation Encourage Healthy Decisions increase exposure to environmen- The second tier of the pyramid tal hazards.14 Educational status is represents interventions that also tightly correlated with car- change the environmental context diovascular risk factors, including to make healthy options the de- smoking.15,16 fault choice, regardless of educa- Although poverty increases ill tion, income, service provision, or FIGURE 1—The health impact pyramid. health within a society, economic other societal factors. The defining development can also increase ill- characteristic of this tier of inter- ness and death from noncommu- vention is that individuals would nicable disease. As living stan- have to expend significant effort describe health system structures. does not see such interventions as dards and life expectancy improve, not to benefit from them. For As a result, existing frameworks falling within the government’s risk for cardiovascular disease example, fluoridated water—which accurately describe neither the appropriate sphere of action. and some cancers increases.17 is difficult to avoid when it is the constituent elements nor the role Interventions at the top tiers are Much of this increase results from public supply—not only improves of public health. designed to help individuals rather modifiable risk factors related to individual health by reducing than entire populations, but they overconsumption of tobacco, un- tooth decay,22 but also provides A FIVE-TIER PYRAMID could theoretically have a large healthy food, and alcohol, with economic benefits by reducing population impact if universally aconcurrentdecreaseinphysical health spending and productivity An alternative conceptual and effectively applied. In practice, activity. Greater wealth can also losses. In countries without either framework for public health action however, even the best programs lead to more roads and an increase adequate natural or added fluori- is a 5-tier health impact pyramid at the pyramid’s higher levels in motor vehicle use, which can dation, health authorities are (Figure 1). In this pyramid, efforts achieve limited public health im- result in increased outdoor air limited to counseling inter- to address socioeconomic deter- pact, largely because of their de- pollution and more injury and ventions, such as encouraging minants are at the base, followed pendence on long-term individual death from traffic crashes. toothbrushing. by public health interventions that behavior change.9 As Rose writes, A third of the world’s urban Other contextual changes that change the context for health (e.g., population lives in slums.18 Sub- create healthier defaults include clean water, safe roads), protective Personal life-style is socially con- stantial health improvements in clean water, air, and food; im- ditioned... . Individuals are un- interventions with long-term ben- likely to eat very differently from high-poverty areas will require provements in road and vehicle efits (e.g., immunizations), direct the rest of their families and improved economic opportunities design; elimination of lead and clinical care, and, at the top, social circle... . It makes little and infrastructure, including reli- asbestos exposures; and iodiza- sense to expect individuals to 22 counseling and education. In gen- behave differently than their able electric power, sanitation, tion of salt. The potential soci- eral, public action and interven- peers; it is more appropriate to transport, and other basic ser- etal impact of decreasing cardio- tions represented by the base of seek a general change in behav- vices.19 Clean water and improved vascular risk factors by changing ioural norms and in the circum- the pyramid require less individ- stances which facilitate their sanitation introduced in the from saturated to unsaturated ual effort and have the greatest adoption.10(p135) United States in the late 19th and cooking oils was demonstrated in population impact. However, be- early 20th centuries may have Mauritius23; eliminating artificial cause these actions may address Socioeconomic Factors been primarily responsible for re- trans fatinfoodisanotherwayto social and economic structures of The bottom tier of the health ducing mortality rates by about prevent cardiovascular disease.24 society, they can be more contro- impact pyramid represents half and child mortality rates by Strategies to create healthier en- versial, particularly if the public changes in socioeconomic factors nearly two thirds in major cities.20 vironmental contexts also include

April 2010, Vol 100, No. 4 | American Journal of Public Health Frieden | Peer Reviewed | Commentaries | 591 COMMENTARIES

designing communities to pro- protective interventions that do usually asymptomatic, such as 1980.45 Two thirds of these in- mote increased physical activity; not require ongoing clinical care; hypertension, hyperlipidemia, dividuals were counseled by enacting policies that encourage these generally have less impact and diabetes. At least a third of a health care provider to lose public transit, bicycling, and walk- than interventions represented by patients do not take medications weight,46 yet daily calorie and fat ing instead of driving; designing the bottom 2 tiers because they as advised, and nonadherence intake continues to rise. buildings to promote stair use; necessitate reaching people as cannot be predicted from socio- Counseling, either within or passing smoke-free laws; and taxing individuals rather than collec- economic or demographic char- outside the clinical context, is tobacco, alcohol, and unhealthy tively. Historic examples include acteristics.41, 42 generally less effective than other foods such as soda and other sugar- immunization, which prevents 2.5 Rigorous accountability, incen- interventions; successfully inducing sweetened beverages. million deaths per year among tives for meaningful outcomes individual behavioral change is Cardiovascular disease risk fac- children globally.32 Another ex- (e.g., blood pressure and choles- the exception rather than the rule. tors (e.g., hypertension) are cur- ample is colonoscopy, which can terol control), and systems to en- For example, although clear, rently addressed at the individual significantly reduce colon cancer able improved performance are strong, and personalized smoking level through screening and med- and is only needed every 5 to 10 all essential to improve health cessation advice, even in the ab- ication. But even assuming perfect years for most people. Smoking care system performance. Elec- sence of pharmacological treat- treatment, this approach fails to cessation programs increase quit tronic health records have the ment, doubles quit rates among prevent almost half of the disease rates; life expectancy among men potential—if and only if they are smokers who want to stop and burden caused by elevated blood who quit at age 35 is almost 7 implemented with prevention should be the norm in medical pressure; cardiovascular risk in- years longer than for those who and accountability as guiding care, it still fails to help 90% of creases with systolic blood pres- continue to smoke.33 principles—to facilitate greatly im- those who are motivated to sure above 115 mm Hg, a level at Male circumcision, a minor proved preventive and chronic quit.47,48 which medical treatment is not outpatient surgical procedure, care.43 This goal is more likely to Nevertheless, educational inter- recommended currently.25,26 can decrease female-to-male be attained if electronic record ventions are often the only ones Changing the environmental con- HIV transmission by as much as keeping is implemented along with available, and when applied context so that individuals can easily 60%.34 Scale-up could potentially changes in both financial incen- sistently and repeatedly may have take heart-healthy actions in the prevent millions of HIV infections tives and physician practices to considerable impact. An example normal course of their lives can in sub-Saharan Africa.35,36 A sin- proactively support preventive of a successful evidence-based have a greater population impact gle dose of azithromycin or iver- care and control of chronic dis- educational intervention is trained than clinical interventions that mectin can reduce the prevalence eases.44 peer counselors advising men treat individuals. of onchocerciasis, a major cause of who have sex with men about For example, modern diets blindness.37 Counseling and Educational reducing HIV risk.49 contain many times the minimum Interventions daily requirement of sodium— Clinical Interventions The pyramid’s fifth tier repre- PROGRAM mostly from packaged foods and The fourth level of the pyramid sents health education (educa- IMPLEMENTATION restaurant meals—making it diffi- represents ongoing clinical inter- tion provided during clinical en- cult for individuals to control their ventions, of which interventions to counters as well as education in Comprehensive tobacco control intake.27 Reducing dietary sodium prevent cardiovascular disease other settings), which is per- programs, which contain elements can reduce hypertension at the have the greatest potential health ceived by some as the essence of that work at all levels of the population level.28,29 A healthier impact. Although evidence-based public health action but is gen- pyramid, illustrate the potential food environment can be created clinical care can reduce disability erally the least effective type of application of this paradigm and by decreasing salt in packaged and prolong life, the aggregate intervention.9 The need to urge the synergies among different foods. This is happening in the impact of these interventions is behavioral change is symptom- levels of intervention. People with United Kingdom, which intro- limited by lack of access, erratic atic of failure to establish con- low incomes and low educational duced four-year sodium reduction and unpredictable adherence, and texts in which healthy choices attainment have higher rates of targets,30 and in Finland, where imperfect effectiveness. Access are default actions. For example, smoking than do people with dietary sodium intake decreased can be limited even in systems that counterbalances to our obeso- higher incomes and education.50 approximately 25% in the past guarantee health coverage for genic environment include ex- Interventions that address social 30 years.31 all38 and is a much greater prob- hortations to increase physical determinants of health, such as lem in the United States and other activity and improve diet, which increasing a population’s educa- Long-Lasting Protective countries without universal health have little or no effect. More than tional and economic status, should Interventions care coverage.39,40 Nonadherence one third of US adults, or 72 therefore reduce smoking rates. The third level of the pyramid is especially problematic for million people, were obese in However, because these changes represents 1-time or infrequent chronic conditions that are 2006, a dramatic increase over often require fundamental social

592 | Commentaries | Peer Reviewed | Frieden American Journal of Public Health | April 2010, Vol 100, No. 4 COMMENTARIES

TABLE 1—Structural Approaches to Health Promotion for Communicable Disease, Noncommunicable Disease, and Injury Prevention

Approaches to Prevention Communicable Disease Noncommunicable Disease Injuries

Counseling and educational Behavioral counseling to reduce sexually Dietary counseling Counseling and public education to avoid interventions transmitted infections Counseling to increase levels of physical activity drinking and driving and encourage compliance Public education about avoiding with traffic laws lifestyle-mediated disease School-based programs to prevent or reduce violent behavior Clinical interventions HIV treatment to decrease viral load Treatment of hypertension and hyperlipidemia Methadone and buprenorphine treatment to and reduce transmission Aspirin therapy for people with coronary heart disease decrease opiate overdose Treatment of tuberculosis, resulting Screening and treatment of women older in decreased spread of infection than 65 years for osteoporosis to reduce fractures Long-lasting protective Immunizations Colonoscopy Brief behavioral counseling to reduce alcohol interventions Male circumcision in countries Treatment of tobacco addiction consumption with high HIV prevalence and significant Surgical sterilization, intrauterine device insertion, Home modification, such as installation of grab female-to-male transmission or other long-acting contraception to reduce bars and handrails, to prevent falls among Mass antibiotics to prevent or treat tropical maternal mortality the elderly diseases (e.g., onchocerciasis) Dental sealants Changing the context Clean water Trans fat elimination in processed food to reduce Road and vehicle design requirements to reduce Reduced indoor smoke pollution from cardiovascular disease crashes and protect pedestrians and bicyclists biomass cooking Sodium reduction in packaged foods and food Laws prohibiting the sale of alcohol to minors Ubiquitous condom availability served in restaurants to reduce cardiovascular and increased alcohol price disease Laws prohibiting driving at even low blood Fluoridation of water to prevent dental cavities alcohol levels Elimination of lead paint and asbestos exposures Effectively implementing laws to mandate helmet Increased unit price for tobacco, alcohol, and use by motorcyclists and motorcycle passengers sugar-sweetened beverages Occupational safety requirements Smoke-free workplaces Community and transit design to promote greater physical activity Socioeconomic factors Reduced poverty to improve immunity, Reduced poverty, increased education levels, and Reduced poverty levels to reduce drug use decreased crowding and environmental more nutritional options to reduce cardiovascular and violence, improved housing options, exposure to communicable microbes, and disease, some cancers, and diabetes and lowered vulnerability to extreme improved nutrition, sanitation, and housing weather conditions

change, they are generally not ad campaigns, particularly as motivated to quit and are treated intervention (tier 5), but if done within the traditional purview of part of a comprehensive tobacco will succeed.48 Education about effectively, such actions can tobacco control or public health control program, not only reduce the harms of smoking provides change the context by altering the programs. tobacco use by changing the people with information to help social norms related to tobacco Context-changing interventions, social context of smoking52 but them change their behavior. Other use (tier 2). such as increasing tobacco taxes, also provide in effect a social im- examples of this 5-tiered frame- establishing smoke-free work- munization against smoking that work applied to communicable PRACTICAL APPLICATION places, and changing the social persists over time. Clinical care disease, chronic disease, and in- OF THE HEALTH IMPACT norms regarding smoking through that includes cessation medica- jury prevention are given in Table PYRAMID hard-hitting antitobacco cam- tions can triple quit rates in in- 1. Inevitably, some programs blur paigns and elimination of adver- dividual smokers, but even the the distinctions between tiers. The health impact pyramid, tising and promotional cues to best systems treat only a small For example, mass media cam- a framework for public health smoke, are highly effective in re- proportion of smokers, and only paigns for tobacco control could action, postulates that addressing ducing tobacco use.51 Hard-hitting one third of those who are be viewed as an educational socioeconomic factors (tier 1, or

April 2010, Vol 100, No. 4 | American Journal of Public Health Frieden | Peer Reviewed | Commentaries | 593 COMMENTARIES

the base of the pyramid) has the children, and community rede- Acknowledgments 12. Marmot M. Social determinants of greatest potential to improve sign to encourage walking and The author thanks Kelly Henning for health inequalities. Lancet. 2005;365 valuable insight and input and Drew (9464):1099–1104. health. Interventions that change bicycling, although far more ef- Blakeman, Cheryl de Jong Lambert, Leslie 13. Mackenbach JP, Stirbu I, Roskam AJ, the context for individual behavior fective, are also politically more Laurence, and Karen Resha for assistance et al. Socioeconomic inequalities in health (tier 2) are generally the most difficult. with article preparation and research. in 22 European countries. N Engl J Med. effective public health actions; Interventions that address so- 2008;358(23):2468–2481. 1-time clinical interventions cial determinants of health have References 14. Wood D. Effect of child and family 1. World Economic and Social Survey poverty on child health in the United (tier 3), such as immunizations, the greatest potential public health 2007. New York, NY: United Nations; States. Pediatrics. 2003;112(3 part 2): can be more effectively applied benefit. Action on these issues 2007. Available at: http://www.un.org/ 707–711. than those requiring ongoing needs the support of government esa/policy/wess/wess2007files/ 15. Winkleby MA, Jatulis DE, Frank E, wess2007.pdf. Accessed December 12, care; and clinical interventions and civil society if it is to be Fortmann SP. Socioeconomic status and 2008. health: how education, income, and (tier 4) are generally, although successful.58 The biggest obstacle 2. Kinsella K, Gist Y. Gender and occupation contribute to risk factors for not inevitably, more effective to making fundamental societal aging: mortality and health. Suitland, MD: cardiovascular disease. Am J Public Health. than counseling and education changes is often not shortage of Bureau of the Census, Economics and 1992;82(6):816–820. Statistics Administration; 1998. Available (tier 5). funds but lack of political will; the 16. Centers for Disease Control and at: http://www.census.gov/ipc/prod/ Prevention. Cigarette smoking among Although the effectiveness of health sector is well positioned to ib98-2.pdf. Accessed December 12, adults—United States, 2002. MMWR 2008. interventions tends to decrease at build the support and develop the Morb Mortal Wkly Rep. 2004;53(20): higher levels of the pyramid, those partnerships required for 3. Panel on Population Projections, 427–431. Committee on Population, National at the top often require the least change.59 17. Frieden TR, Henning KJ. Public Research Council. Beyond Six Billion: health requirements for rapid progress in political commitment. Achieving To say that social and contex- Forecasting the World’s Population. global health. Glob Public Health. 2009; Bongaarts J, Bulatao RA, eds. Washing- social and economic change might tual changes are more effective 4(4):323–337. ton, DC: National Academy Press; 2000. require fundamental societal at improving public health is not 18. Sclar ED, Garau P, Carolini G. The 4. McKeown T. The Modern Rise of transformation. Contextual change to imply that other interventions 21st century health challenge of slums Population. New York, NY: Academic and cities. Lancet. 2005;365(9462): is often controversial, as evi- should be ignored. For different Press; 1976. 901–903. denced by disputes over smoke- public health problems, differ- 5. Ford ES, Ajani UA, Croft JB, et al. 19. Unger A, Riley LW. Slum health: Explaining the decrease in U.S. deaths free laws, restrictions on artificial ent interventions may be the from understanding to action. PLoS Med. from coronary disease, 1980-2000. N trans fat, and water fluorida- most effective or feasible in any 2007;4(10):1561–1566. Engl J Med. 2007;356(23):2388–2398. 53,54 tion. One-time interventions given context. Education to en- 20. Cutler D, Miller G. The role of 6. For a Healthy Nation: Return on public health improvements in health tend to be less controversial, al- courage condom use, although of Investments in Public Health. Hyattsville, advances: the twentieth-century United though immunization programs only limited effectiveness, can MD: US Public Health Service; 1994. States. Demography. 2005;42(1):1–22. that attempt to reach all members reduce HIV transmission and 7. Grizzell J. High reach/low cost 21. UN-Water. Global annual assess- of a society often meet resistance save lives. Changing the context health agenda programming. Pomona, CA: California State Polytechnic Uni- ment of sanitation and drinking water: arising from suspicion and to make condoms ubiquitously versity, Pomona; 2007. Available at: 2008 pilot report—testing a new report- disbelief.55 available and acceptable makes http://www.csupomona.edu/;jvgrizzell/ ing approach. Geneva, Switzerland: World Health Organization; 2008. Although the structure and fi- education about their use more healthagenda/intervention_pyramid.htm. Accessed May 15, 2009. Available at: http://www.unwater.org/ nancing of health care systems can effective. Comprehensive public downloads/GLAAS_2008_Pilot_Report. 8. Hamilton N, Bhatti T. Population pdf. Accessed December 3, 2009. be controversial, clinical care itself health programs should gener- Health Promotion: An Integrated Model of rarely is. While exceptions exist, ally attempt to implement mea- Population Health and Health Promotion. 22. Centers for Disease Control and Prevention. Ten great public health health education usually requires sures at each level of interven- Ottawa: Health Canada, Health Promo- tion Development Division; 1996. achievements—United States, 1900– minimal political backing. Hence tion to maximize synergy and 1999. MMWR Morb Mortal Wkly Rep. 9. Whitlock EP, Orleans CT, Pender N, 1999;48(12):241–243. the greater popularity of school- the likelihood of long-term Allan J. Evaluating primary care behav- based antismoking programs success. j ioral counseling interventions. Am J Prev 23. Uusitalo U, Feskens EJ, Tuomilehto J, et al. Fall in total cholesterol concen- (despite consistent evidence they Med. 2002;22(4):267–284. tration over five years in association with 56 10. Rose G, Khaw K-T, Marmot M. provide little to no benefit ) than changes in fatty acid composition of Rose’s Strategy of Preventive Medicine. About the Author cooking oil in Mauritius: cross sectional of proven tobacco control inter- New York, NY: Oxford University Press; Thomas R. Frieden is with the Centers for survey. BMJ. 1996;313(7064):1044– ventions such as taxation, smoke- 2008. Disease Control and Prevention, Atlanta, GA 1046. free environments, and compre- Correspondence should be sent to Thomas 11. Closing the gap in a generation: 24. Angell SY, Silver LD, Goldstein GP, R. Frieden, MD, MPH, Director, Centers for health equity through action on the social hensive marketing bans. Similarly, et al. Cholesterol control beyond the Disease Control and Prevention, 1600 determinants of health. Final Report of clinic: New York City’s trans fat restric- exhorting people to exercise more Clifton Rd, NE, MS D-14, Atlanta, GA the Commission on Social Determinants tion. Ann Intern Med. 2009;151(2):129– and eat less is politically popular, 30333 (e-mail: [email protected]). Reprints can of Health. Geneva, Switzerland: World 134. but taxation of soda and other be ordered at http://www.ajph.org by clicking Health Organization; 2008. Available at: 57 the ‘‘Reprints/Eprints’’ link. http://whqlibdoc.who.int/publications/ 25. Lewington S, Clarke R, Qizilbash N, sugar-sweetened beverages, This article was accepted December 8, 2008/9789241563703_eng.pdf. Peto R, Collins R, Prospective Studies bans on marketing junk food to 2009. Accessed October 18, 2009. Collaboration. Age-specific relevance of

594 | Commentaries | Peer Reviewed | Frieden American Journal of Public Health | April 2010, Vol 100, No. 4 COMMENTARIES

usual blood pressure to vascular mortality: eliminating ocular Chlamydia tracho- uspstf08/sti/stirs.htm#clinical. Accessed a meta-analysis of individual data for one matis with repeat mass antibiotic treat- November 13, 2008. million adults in 61 prospective studies. ments. JAMA. 2004;292(6):721–725. 50. Centers for Disease Control and Lancet. 2002;360(9349):1903–1913. 38. Shah SM, Cook DG. Socio-economic Prevention. Cigarette smoking among 26. Law M, Wald N, Morris J. Lowering determinants of casualty and NHS direct adults and trends in smoking cessation– blood pressure to prevent myocardial in- use. J Public Health (Oxf). 2008;30(1): United States, 2008. MMWR Morb Mor- farction and stroke: a new preventive 75–81. tal Wkly Rep. 2009;58(44):1227–1232. strategy. Health Technol Assess. 2003; 39. Lasser KE, Himmelstein DU, 51. The Role of the Media in Promoting 7(31):1–94. Woolhandler S. Access to care, health and Reducing Tobacco Use. Bethesda, MD: 27. Mattes RD, Donnelly D. Relative status, and health disparities in the National Cancer Institute; 2008. Tobacco contributions of dietary sodium sources. United States and Canada: results of control monograph 19. NIH Publicaton J Am Coll Nutr. 1991;10(4):383–393. a cross-national population-based survey. 07-6242. 28. Sacks FM, Svetkey LP, Vollmer Am J Public Health. 2006;96(7):1300– 52. Office on Smoking and Health. Best WM, et al. Effects on blood pressure of 1307. Practices for Comprehensive Tobacco reduced dietary sodium and the Dietary 40. Schoen C, Doty MM. Inequities in Control Programs—2007. Atlanta, GA: Approaches to Stop Hypertension access to medical care in five countries: National Center for Chronic Disease Pre- (DASH) diet. DASH–Sodium Collabora- findings from the 2001 Commonwealth vention and Health Promotion, Centers tive Research Group. N Engl J Med. 2001; Fund International Health Policy Survey. for Disease Control and Prevention; 344(1):3–10. Health Policy. 2004;67(3):309–322. 2007. 29. Danaei G, Ding EL, Mozaffarian D, 41. Sbarbaro JA. Compliance: induce- 53. Cross DW, Carton RJ. Fluoridation: et al. The preventable causes of death in ments and enforcements. Chest. 1979; a violation of medical ethics and human the United States. PLoS Med. 2009;6(4): 76(6 Suppl):750–756. rights. Int J Occup Environ Health. 2003; 9(1):24–29. e1000058. 42. Haynes RB, McKibbon KA, Kanani 30. He FJ, MacGregor GA. A compre- R. Systematic review of randomized trials 54. Frieden TR, Bassett MT, Thorpe LE, hensive review on salt and health and of interventions to assist patients to follow Farley TA. Public health in New York current experience of worldwide salt re- prescriptions for medications. Lancet. City, 2002–2007: confronting epidemics duction programmes. J Hum Hypertens. 1996;348(9024):383–386. of the modern era. Int J Epidemiol. 2008; 37(5):966–977. 2009;23(6):363–384. 43. Mostashari F, Tripathi M, Kendall 31. Laatikainen T, Pietinen P, Valsta L, M. A tale of two large community elec- 55. Garrett L. Preferring anarchy and Sundvall J, Reinivuo H, Tuomilehto J. tronic health record extension projects. class disparity. In: Betrayal of Trust: The Sodium in the Finnish diet: 20-year Health Aff (Millwood). 2009;28(2):345– Collapse of Global Public Health.New trends in urinary sodium excretion among 356. York, NY: Hyperion; 2000:268–485. the adult population. Eur J Clin Nutr. 44. Frieden TR, Mostashari F. Health 56. Thomas R, Perera R. School-based 2006;60(8):965–970. care as if health mattered. JAMA. 2008; programmes for preventing smoking. 32. WHO, UNICEF, World Bank. State 299(8):950–952. Cochrane Database Syst Rev. 2006;3: CD001293. of the World’s Vaccines and Immuniza- 45. Prevalence of overweight and obesity tion. 3rd ed. Geneva, Switzerland: World among adults: United States, 2003–2004. 57. Brownell KD, Frieden TR. Ounces Health Organization; 2009. Available at: Atlanta, GA: Centers for Disease Control of prevention: the public policy case for http://whqlibdoc.who.int/publications/ and Prevention; 2008. Available at: http:// taxes on sugared beverages. N Engl J Med. 2009/9789241563864_eng.pdf. www.cdc.gov/nchs/products/pubs/pubd/ 2009;360(18):1805–1808. Accessed January 14, 2010. hestats/overweight/overwght_adult_03. 58. Marmot M, Friel S, Bell R, Houweling 33. Taylor DH Jr, Hasselblad V, Henley htm. Accessed July 23, 2009. TA, Taylor S, Commission on Social SJ, Thun MJ, Sloan FA. Benefits of smok- 46. Ogden CL, Carroll MD, McDowell Determinants of Health. Closing the gap ing cessation for longevity. Am J Public MA, Flegal KM. Obesity among adults in in a generation: health equity through Health. 2002;92(6):990–996. the United States—no change since 2003- action on the social determinants of 34. Auvert B, Taljaard D, Lagarde E, 2004. Hyattsville, MD: National Center health. Lancet. 2008;372(9650):1661– Sobngwi-Tambekou J, Sitta R, Puren A. for Health Statistics; 2007. NCHS data 1669. Randomized, controlled intervention brief 1. Available at: http://www.cdc.gov/ 59. Friel S, Bell R, Houweling T, trial of male circumcision for reduction nchs/data/databriefs/db01.pdf. Accessed Marmot M. Calling all Don Quixotes and of HIV infection risk: the ANRS 1265 October 18, 2009. Sancho Panzas: achieving the dream of Trial. PLoS Med. 2005;2(11):e298. 47. Thorndike AN, Regan S, Rigotti NA. global health equity through practical ac- 35. Nagelkerke NJ, Moses S, de Vlas SJ, The treatment of smoking by US physi- tion on the social determinants of health. Bailey RC. Modelling the public health cians during ambulatory visits: 1994– Glob Health Promot. 2009(Suppl 1): impact of male circumcision for HIV pre- 2003. Am J Public Health. 2007;97(10): 9–13. vention in high prevalence areas in Africa. 1878–1883. BMC Infect Dis. 2007;7:16. 48. Fiore MC, Jaen CR, Baker TB, et al. 36. WHO/UNAIDS. New data on Treating Tobacco Use and Dependence: male circumcision and HIV prevention: 2008 Update. Clinical Practice Guideline. policy and programme implications. Rockville, MD: US Public Health Service; Geneva, Switzerland: World Health Or- 2008. ganization; 2007. Available at: http:// 49. US Preventive Services Task Force. libdoc.who.int/publications/2007/ Behavioral counseling to prevent sexually 9789241595988_eng.pdf. Accessed transmitted infections recommendation October 18, 2009. statement. Rockville, MD: Agency for 37. Melese M, Chidambaram J, Healthcare Research and Quality; 2008. Alemayehu W, et al. Feasibility of Available at: http://www.ahrq.gov/clinic/

April 2010, Vol 100, No. 4 | American Journal of Public Health Frieden | Peer Reviewed | Commentaries | 595