May 2003, Vol 93, No. 5 | American Journal of Public Health 771-774
© 2003 American Public Health Association
Is There a Gradient in Life Span by Position in the Social Hierarchy?
John N. Lavis, MD, PhD,
Christopher B. McLeod, MA,
Cameron A. Mustard, ScD and
Greg L. Stoddart, PhD
John N. Lavis is with the Centre for Health Economics and Policy Analysis and the Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada; the Institute for Work & Health, Toronto, Ontario; and the Population Health Program, Canadian Institute for Advanced Research, Toronto, Ontario. Christopher B. McLeod is with the Institute for Work & Health and the Centre for Health Economics and Policy Analysis, McMaster University. Cameron A. Mustard is with the Institute for Work & Health; the Population Health Program, Canadian Institute for Advanced Research; and the Department of Public Health Sciences, University of Toronto, Canada. Greg L. Stoddart is with the Centre for Health Economics and Policy Analysis, McMaster University; the Department of Clinical Epidemiology and Biostatistics, McMaster University; and the Population Health Program, Canadian Institute for Advanced Research.
Correspondence: Requests for reprints should be sent to John N. Lavis, MD, PhD, Health Sciences Centre, Rm 2D3, McMaster University, 1200 Main St West, Hamilton, ON, Canada L8N 3Z5 (e-mail: lavisj{at}mcmaster.ca).
 |
INTRODUCTION
|
|---|
Studies have repeatedly found a gradient in health by socioeconomic status, even in fairly homogeneous groups such as the UK civil service.1,2 These findings have led to the suggestion that there is a relationship between health and position in a hierarchy. Studies of social hierarchies in animals have also identified this relationship.3
Disentangling the relative importance of economic versus social hierarchies in humans, however, has been complicated by their degree of overlap. The British aristocracy, for example, although far more of a social hierarchy than an economic one, defines social rank by the type of peerage held. By definition, peers and baronets hold the highest position in an economic hierarchy of family members associated with the title through marriage or birth, and the staff who work for them. Only in situations where social hierarchy dictates matters such as privileges or treatment (e.g., in the House of Lords, at court) would peers or baronets perceive themselves as holding anything other than the highest position.
We hypothesized that there would be a gradient in the life spans of British peers and baronets by their position in the social hierarchy: dukes would live longer than marquesses, marquesses would live longer than earls, and so on down through viscounts, barons, Scottish lords of parliament, and baronets.
|
METHODS
|
|---|
We extracted usable birth and death information for 9529 male peers and baronets from Burke’s Peerage & Baronetage, a data source that records the lineage of all existing titles to their origin.4 We excluded women, because there were so few of them in the sample (n = 87). We calculated life span in years by subtracting year of birth from year of death. We coded position in the social hierarchy according to the highest peerage type a man held during his lifetime. We coded men who were awarded a peerage or baronetage for the first time (n = 1525) and peers or baronets who were awarded a new, higher title (n = 283) as first-title holders; others were coded as inherited-title holders (n = 7721).
For each group (first-title holders and inherited-title holders), we regressed life spans on peerage type. We accounted for a cohort effect by adjusting for century of birth (coded as "16th century or earlier," "17th century," "18th century," "19th century," or "20th century"). To rule out a "healthy survivor" effect, we accounted for selection related to the timing of when a peerage or baronetage was awarded over a man’s life span by adjusting for age at award. We accounted for selection related to the propensity to be called upon to fight in wars by excluding men who died of unnatural causes such as being killed in action or on duty (n = 238) and men who died in years in which Britain was at war (n = 1562).5 We were not, however, able to account for the heritable dimension of human longevity; any tendency toward shorter (or longer) life spans within a given peerage type would grow over time if there were high rates of intermarriage within that peerage type, especially for small groups such as many of the high-level peerages.
|
RESULTS
|
|---|
Baronets made up 50.4% of the sample; earls and barons made up an additional 37%. Earlier cohorts, from periods when life spans were shorter, were composed of a higher proportion of high-level peers than later cohorts. Some high-level peerages tended to be awarded earlier in life than some low-level peerages and the baronetages. Dukes and Scottish lords of parliament, the outliers among first-title holders, were awarded their peerages earlier in the historical period and were fewer in number than other peerage types (Figure 1
). Baronets were the outliers among inherited-title holders. First-title holders must live long enough to both perform the acts that lead to their award and actually have the peerage or baronetage awarded, whereas inherited-title holders must only live long enough to survive the incumbent. First-title holders lived longer than inherited-title holders in each of the last 5 centuries and for each peerage type, with the exception of dukes. A higher proportion of dukedoms were awarded early in the historical period we studied (i.e., when life spans were shorter), and dukes had the smallest difference between the ages at which first- and inherited-title holders were awarded their peerages.
No clear gradient existed in the life spans of British peers and baronets by their position in the social hierarchy before accounting for century of birth and selection processes (Figure 1
). Likewise, for inherited-title holders, there appeared to be no gradient in life span by position in the social hierarchy after accounting for century of birth and 2 types of selection processes (Table 1). If anything, peers of several peerage types had shorter life spans than peers in the lowest position in the social hierarchy (baronets). When peerage type was treated as a continuous variable, with baronets (the lowest position in the social hierarchy) coded as "1" and dukes (the highest position) coded as "7," peerage type was negatively associated with life span across all regression equations. When peerage type was treated as a series of 3 or 6 dummy variables, with baronets as the omitted comparison category, the negative associations for barons or barons together with Scottish lords of parliament (and, to a lesser extent, dukes or dukes together with marquesses) remained statistically significant or of borderline statistical significance across most regression equations. For first-title holders, there also appeared to be no gradient in life span by position in the social hierarchy after accounting for century of birth and 2 types of selection processes, with the exception of a negative association for earls, who had shorter life spans than baronets (results not shown).
View this table:
[in this window]
[in a new window]
|
TABLE 1 —Ordinary Least Squares Regressions for Male Inherited-Title Holders Using Life Span as the Dependent Variable
|
|
|
DISCUSSION
|
|---|
Contrary to our hypothesis, we found no gradient in life span by position in the social hierarchy. This finding is consistent with the observation that gradients in health status disappear at the top of the economic hierarchy.6,7 Our finding about the lack of a gradient among first-title holders is also consistent with the observation that extraordinary achievement is not always associated with longevity.8,9 After studying a unique population in which position in a social hierarchy is largely disentangled from position in an economic hierarchy, we conclude that among those at the top of a social hierarchy, relative position does not confer a survival advantage.
|
Acknowledgments
|
|---|
Lavis holds the Canada Research Chair in Knowledge Transfer and Uptake, and he receives partial salary support as Liberty Health Scholar from the Canadian Institute for Advanced Research. Mustard holds an Investigator award from the Canadian Institutes for Health Research.
We thank Tasha Sawler, who conducted the data extraction; Heidi Barnett, who oversaw the data extraction and conducted periodic quality checks; and Scott Trevithick, who helped us to establish an appropriate source for information for periods during which Britain was at war. All of these individuals are or were at the time employed by the Institute for Work & Health.
|
Footnotes
|
|---|
J. N. Lavis conceived of the original study, oversaw the data extraction and analysis, and drafted and revised the paper. C. B. McLeod conducted quality checks on the data, led the data analysis, and provided feedback on all drafts and revisions of the paper. C. A. Mustard and G. L. Stoddart helped to conceive of the original study, reviewed interim data analyses, and provided feedback on all drafts and revisions of the paper.
Peer Reviewed
Accepted for publication April 1, 2002.
|
References
|
|---|
1. Adler NE, Boyce T, Chesney MA, et al. Socioeconomic status and health. The challenge of the gradient. Am Psychol.1994;49:15–24.[Medline]
2. Marmot MG, Shipley MJ, Rose G. Inequalities in death—specific explanations of a general pattern? Lancet.1984;1:1003–1006.[ISI][Medline]
3. Sapolsky RM. Stress in the wild. Sci Am.1990;262:116–123.[ISI][Medline]
4. Mosley C, ed. Burke’s Peerage & Baronetage. 106th ed. Crans-près-Céligny, Switzerland: Burke’s Peerage (Genealogical Books) Ltd; 2000.
5. The New Encyclopædia Britannica. 15th ed. Chicago, Ill: Encyclopædia Britannica, Inc.; 1998.
6. Wolfson M, Rowe G, Gentleman JF, Tomiak M. Career earnings and death: a longitudinal analysis of older Canadian men. J Gerontol.1993;48:S167–S179.[ISI][Medline]
7. Backlund E, Sorlie PD, Johnson NJ. The shape of the relationship between income and mortality in the United States. Evidence from the National Longitudinal Mortality Study. Ann Epidemiol.1996;6:12–20.[ISI][Medline]
8. Redelmeier DA, Singh SM. Longevity of screenwriters who win an academy award: longitudinal study. BMJ.2001;323:1491–1496.[Abstract/Free Full Text]
9. Davey Smith G. Death in Hollywood. BMJ.2001;323:1441–1442.[Free Full Text]
Copyright © 2003 by the American Public Health Association
TOP
INTRODUCTION
METHODS
