Editor's Choice |
Equal, but different? Ecological, individual and instrumental approaches to understanding determinants of health
Editing a journal, even a relatively minor one like the International Journal of Epidemiology, can be hard work, but some things make it very rewarding. The chance to reprint a report from the first Framingham study—a community based prospective investigation of tuberculosis control, initiated in 19171—and commentaries from George Comstock about the study itself,2 Meryvn Susser and Zena Stein about Donald Armstrong, the instigator of the study,3 and William Kannel and Daniel Levy, relating it to the better known Framingham heart disease study initiated a generation later4—is an example of one of these rewards. In the case of the transition from Framingham I (TB) to Framingham II (coronary heart disease) the move was certainly from the community to the individual, perhaps in line with the general tendency of epidemiological thinking over this period.
A second example of the joys of editing is the ability to host an extended debate on an important topic, as exemplified by the exchange initiated by Jose Tapia Granados' analysis of the effect of short-term economic growth on health.5 To cut a long story short, this exchange6–12 relates to the use of data on secular trends in health outcomes (mainly mortality) in relation to economic indicators to estimate whether short-term economic growth improves or harms health. As Tapia Granados and other commentators point out, formal analytical interest in this issue has existed since the first decades of the 20th century, and while the methods have improved greatly, some of the interpretive issues are similar to those raised in an exchange between Joseph Eyer13 (one of whose papers ‘Prosperity as a cause of death’ admirably adopted a title that summarised its content14) and Richard Cooper15 a quarter of a century ago. The continued focus on this is completely justified, as these are important issues about population health. They are also issues than can ultimately only be addressed by the use of aggregate data.
Analyses of population aggregates—either through secular trend data or through the comparison of health outcomes between areas—are basic epidemiological approaches to understanding the determinants of population health, and ones exemplified in pioneering texts such as the first edition of Jerry Morris' ‘Uses of Epidemiology’16 50 years ago. In recent years such ecological analyses played an important role in the development of the fetal origins of adult disease hypothesis, as discussed in the cohort profile for the Hertfordshire cohort study, which was set up to follow up on the findings from these aggregate analyses.17 Our second cohort profile also starts with a discussion of ecological analyses of cause-specific death rates within China, which identified clear and important environmental influences on disease risk that are now being investigated in a large-scale cohort study involving half a million people.18 A different scale of aggregate experience is illustrated in our Photoessay,19 which considers how social fragmentation is reflected in very specific features of place.
Ecological studies have other advantages, one of which is that they provide estimates of causal effects that are not attenuated by measurement error (which is discussed in another context by Frost and White in this issue).20 However, they are subject to confounding, as Yoav Ben-Shlomo21 discusses in his editorial. An approach that can help here is the use of instrumental variables (discussed in different contexts in our pages several times in recent years22,23), which have even been applied to such seemingly difficult issues as the one tackled by Houweling et al.24 regarding the association between wealth and child mortality.25 Classical epidemiological designs such as using information on the company providing water supply as an index of water quality (as utilized by John Snow) are, essentially, applications of this approach, a fact that links the instrumental variable and ecological analyses.
A final use of aggregates in this issue of the International Journal of Epidemiology is the entertaining comparison of longevity of popes and artists by Carrieri and Serraino.26 These authors suggest that the longer life expectancy of popes reflects the lives characterized by social instability, high-risk behaviours and geographical mobility (and thus infection risk) of the artists. Certainly the life of one of the artists in Carrieri and Serriano's sample—Michelangelo Merisi, better known as Caravaggio, illustrates these threats to longevity. He often lived in poverty, who was fond of alcohol (see his self-portrait as the god of wine—known as ‘sick little Bacchus’, reproduced here, Figure 1), left the protection and comfort of the house of one patron because he was fed up with the (healthy) diet of salad, travelled constantly, engaged in frequent fights (one leading to murder), was sexually reckless, and who ultimately ‘died as wretchedly as he lived’aged 39.27 However, the assumption of better behaviour by the popes is perhaps unjustified. Of the 41 popes who succeeded Pope John VIII in 872 when his attendants beat him to death, a third had unnatural deaths, some at the hands of their successors.28 The incessant copulator John XII was accused ‘of homicide, perjury, sacrilege, [and] incest with your relatives, including two of your sisters’. No wonder it was popularly considered that the antichrist would first appear as a pope. While painting the ceiling of the Sistine chapel in the early 1500s Michelangelo despaired of his patron, the syphilitic father of (at least) three, Pope Julius II, in verse:
Of chalices they make helmet and swordClearly Michelangelo would have recognized the dangers of assuming that certain aggregate groups, such as popes, were free from the vices of other aggregate groups, such as artists.
View larger version (85K):
[in this window]
[in a new window]
Sick Bacchus by Caravaggio. Reproduced with permission from Archivio fotografica Soprintendenza Speciale per il Polo Museale Roma, Italy
And sell by the bucket the blood of the LordHis cross, his thorns are blades in poison dipped
And even Christ himself is of patience stripped27
References
1 Armstrong DM. The medical aspects of the Framingham community Health and Tuberculosis Program Journal. Am Rev Tuberc 1918;2:195–206. (Reprinted Int J Epidemiol, 2005;34:1183–1187.)
2 Comstock GW. Commentary: The first Framingham Study—a pioneer in community-based participatory research. Int J Epidemiol 2005; 34:1188–90.
3 Susser M, Stein Z. Commentary: Donald Budd Armstrong (1886–1968)—pioneering tuberculosis prevention in general practice. Int J Epidemiol 2005;34:1191–93.
4 Kannell WB, Levy D. Commentary: Medical aspects of the Framingham Community Health and Tuberculosis Demonstration. Int J Epidemiol 2005;34:1187–88.
5 Tapia Granados JA. Increasing mortality during the expansions of the US economy, 1900–1996. Int J Epidemiol 2005;34:1194–202.
6 McKee M, Suhrcke M. Commentary: Health and economic transition. Int J Epidemiol 2005;34:1203–06.
7 Ruhm CJ. Commentary: Mortality increases during economic upturns. Int J Epidemiol 2005;34:1206–11.
8 Catalano R, Bellows B. Commentary: If economic expansion threatens public health, should epidemiologists recommend recession? Int J Epidemiol 2005;34:1212–13.
9 Brenner MH. Commentary: Economic growth is the basis of mortality rate decline in the 20th century—experience of the United States 1901–2000. Int J Epidemiol 2005;34:1214–21.
10 Neumayer E. Commentary: The economic business cycle and mortality. Int J Epidemiol 2005;34:1221–22.
11 Edwards RD. Commentary: Work, well-being, and a new calling for countercyclical policy. Int J Epidemiol 2005;34:1222–25.
12 Tapia Granados JA Response: On economic growth, business fluctuations, and health progress. Int J Epidemiol 2005;34:1226–33.
13 Eyer J. Reply to Dr Cooper. Int J Health Serv 1979;9:161–68.[Medline]
14 Eyer J. Prosperity as a cause of death. Int J Health Serv 1977;7:125–50.[ISI][Medline]
15 Cooper R. Prosperity—of the capitalist variety—as a cause of death. Int J Health Serv 1979;9:155–59.[Medline]
16 Morris JN. Uses of Epidemiology. Edinburgh: Livingstone, 1957.
17 Syddall HE, Aihie Sayer A, Dennison EM et al. Cohort Profile: The Hertfordshire Cohort Study. Int J Epidemiol 2005;34:1234–42.
18 Chen ZM, Lee L, Chen J et al. Cohort Profile: The Kadoorie Study of Chronic Disease in China (KSCDC). Int J Epidemiol 2005;34:1243–49.
19 Zyada A. Visualizing sense of community and social fragmentation. Int J Epidemiol 2005;34:1255–56.
20 Frost C, White IR. The effect of measurement error in risk factors that change over time in cohort studies: do simple methods overcorrect for ‘regression dilution’? Int J Epidemiol 2005;34:1359–68.
21 Ben-Shlomo Y. Editorial. Real epidemiologists don't do ecological studies? Int J Epidemiol 2005;34:1181–82.
22 Greenland S. An introduction to instrumental variables for epidemiologists. Int J Epidemiol 2000;29:722–29.
23 Thomas DC, Conti DV. Commentary: The concept of ‘Mendelian Randomization’ Int J Epidemiol 2004;33:21–25.
24 Houweling TAJ, Kunst AE, Looman CWN, Mackenbach JP. Determinants of under-5 mortality among the poor and the rich: a cross-national analysis of 43 developing countries. Int J Epidemiol 2005; 34:1357–65.
25 Pritchett L, Summers LH. Wealthier is healthier. J Hum Resources 1996;31:841–68.[CrossRef][ISI]
26 Carrieri MP, Serraino D. Letter. Longevity of popes and artists between the 13th and the 19th century. Int J Epidemiol 2005;34:1435–36.
27 Hellwig K. Michelangelo Merisi da Caravaggio. Munich: Prestel, 2005.
28 Moynahan B. The Faith: A History of Christianity. London: Pimlico, 2003.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. March and E. Susser The eco- in eco-epidemiology Int. J. Epidemiol., December 1, 2006; 35(6): 1379 - 1383. [Full Text] [PDF]
|
|
|
|
|
|
J. A HANLEY, M. P. CARRIERI, and D. SERRAINO Statistical fallibility and the longevity of popes: William Farr meets Wilhelm Lexis Int. J. Epidemiol., June 1, 2006; 35(3): 802 - 805. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||