Annals of Oncology Advance Access originally published online on October 25, 2005
Annals of Oncology 2006 17(1):157-166; doi:10.1093/annonc/mdj018
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2005 European Society for Medical Oncology
Adult height in relation to mortality from 14 cancer sites in men in London (UK): evidence from the original Whitehall study
1 MRC Social & Public Health Sciences Unit, University of Glasgow, Glasgow; 2 Department of Epidemiology and Public Health, University College London, London; 3 Department of Social Medicine, University of Bristol, Bristol, UK
* Correspondence to: Dr G. D. Batty, MRC Social & Public Health Sciences Unit, University of Glasgow, 4 Lilybank Gardens, Glasgow G12 8RZ, UK. Tel: +44-141-357-7520; Fax: +44-141-337-2389; E-mail: david-b{at}msoc.mrc.gla.ac.uk
 |
Abstract |
|---|
 Top Abstract introductionmaterials and methodsresultsdiscussionReferences |
|---|
Background: Adult height has been related to organ-specific malignancies in relatively few studies. Findings are discrepant for some sites and several studies are subject to a series of methodological limitations.
Materials and methods: We examined the association of adult height with death attributed to 14 cancer sites using data from the original Whitehall cohort. This is a prospective study of 18 403 middle-aged, non-industrial, London-based, male government employees who were examined in the late 1960s and then followed up for mortality for a maximum of 35 years.
Results: There were 11 099 deaths during follow-up, 3101 (28%) of which were ascribed to cancer. Cox proportional hazards regression models revealed modest effects for height in relation to site-specific cancers. Following adjustment for covariates that included employment grade (an indicator of socioeconomic position), body mass index and smoking habit, increased height was associated with elevated mortality rates for cancer of combined sites [hazards ratio per 5 cm increase in height (95% confidence interval); P for trend across height categories: 1.05 (1.03, 1.08); P <0.001], lung [1.13 (1.06, 1.20); P <0.001], prostate [1.07 (0.99, 1.15); P = 0.08], kidney [1.20 (0.99, 1.46); P = 0.08], skin [1.35 (1.06, 1.70); P = 0.02] and leukaemia [1.11 (0.96, 1.28); P = 0.02].
Conclusions: Amongst other explanations, the weak positive height–cancer gradients apparent herein may be ascribed to early life exposures that correlate with adult height, such as high caloric intake.
Key words: cancer, cohort study, height, Whitehall
|
introduction |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionReferences |
|---|
Over the last decade there has been a re-emergence of interest in the early life origins of chronic disease in adulthood [1
]. There is, however, a paucity of studies with extended follow-up of children with which to examine these associations. Accordingly, investigators have related indices of early life factors assessed in cohorts of middle-aged persons to subsequent health experience. These indices include social factors, such as adult recall of childhood socioeconomic position [2], and physical characteristics, chiefly height [3]. Height may be regarded as a marker of early life illness, nutrition or psychosocial stress [3], and has the advantage of remaining relatively stable from early to late adulthood.
In industrialised populations, an inverse association between height and all-cause mortality is apparent. Thus, taller persons experience lower death rates in cohorts drawn from Britain [4, 5], Scandinavia [6–9], Finland [9], the USA [10] and South Korea [11]. That these associations are generally weak may be ascribed to the apparent opposing gradients for height in relation to two major chronic diseases among older persons: cardiovascular disease and all cancers combined. Thus, while the former association is inverse [5, 12], the latter has been shown to be positive [13]. Within subtypes of malignancy, a heterogeneous relation with height is also apparent [14]. Although greater stature is associated with elevated rates of carcinoma of the colorectum, lung, prostate and breast (in women), there is a suggestion of an inverse relation with neoplasm of the stomach and oesophagus in some studies [14]. For other cancer subtypes, findings are either discrepant (e.g. renal, pancreas) or there are too few studies from which to draw clear conclusions about links with height (e.g. bladder, haematopoetic, skin) [14].
In addition to this weak evidence base for height in relation to some malignancies, several existing studies are subject to a series of methodological shortcomings. First, the majority utilise a case–control design in which the problem of selection and information bias is more acute than in prospective cohort studies. Secondly, in the prospective studies that do exist, low statistical power is commonplace owing to a limited number of cases. Also, investigators often focus on one type of malignancy, which limits conclusions about specificity of association [14] and therefore insights into causality. Thirdly, many datasets lack information on potentially important confounding variables such as socioeconomic circumstances [15], adiposity and, critically for many malignancies, tobacco use [14]. Importantly, when available adjusting for such covariates seems to attenuate the height–cancer effect [14].
In the Whitehall study, over 18 000 male, middle-aged, London-based government employees participated in a medical examination in the late 1960s that included measurement of their height and a range of collateral data [16]. In extended (maximum 35 year) mortality surveillance of this cohort there have been over 3000 cancer deaths. We are therefore able to address these issues of prospective study scarcity and methodological shortcomings. In earlier (maximum 27 year) follow-ups of this cohort, reduced risks of total mortality [4] and cardiovascular disease [17], and elevated rates of some cancers [17, 18], were apparent in taller men. The present analysis allows us to examine the link between height and a much greater range of organ-specific malignancies than previously examined in this study.
|
materials and methods |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionReferences |
|---|
In the Whitehall study, data were collected on 18 403 non-industrial, London-based, male government employees aged from 40 to 64 years when examined between September 1967 and January 1970, representing a 74% response. This involved the completion of a study questionnaire and participation in a medical examination, both of which have been described in detail elsewhere [16
]. In brief, the questionnaire included enquiries regarding civil service employment grade (an indicator of socioeconomic position [19]), smoking habits, intermittent claudication, angina, chronic bronchitis, marital status, physical activity, unexplained weight loss in the preceding year and prescribed drug use. Forced vital capacity, ischaemia, fasting plasma cholesterol, post-challenge 2-h blood glucose, blood pressure, and triceps skinfold thickness were all determined using standardised protocols. Height was measured with the subject wearing shoes and standing with their back to a measuring rod. Readings were taken to the nearest half inch below the point of measurement [20]. For the purposes of the present analyses, all readings were converted to centimetres.
ascertainment of cancer mortality
The records of 18 245 men (99.1% of cohort members) were traced and flagged using the procedures of the National Health Service Central Registry until 31 December 2002. Among the 11 710 men who died, 91.6% of death certificates were coded according to the 8th revision of the International Classification of Diseases (ICD) [21], 7% according to the 9th revision [22] and 1.4% according to the 10th revision [23]. The category of all malignant neoplasms (ICD-8: 140–208; ICD-9: 140–209; ICD-10: C00–C97)—referred to as ‘all-cancers’—was divided into individual organs. In ascending order of ICD-8/9 code, these were: oesophagus (ICD-8/9: 150; ICD-10: C15); stomach (ICD-8/9: 151; ICD-10: C16); colon (ICD-8/9: 153; ICD-10: C18); rectum (ICD-8/9: 154; ICD-10: C19); liver (ICD-8/9: 155–156; ICD-10: C22–C24); pancreas (ICD-8/9: 157; ICD-10: C25); trachea, bronchus and lung (ICD-8/9: 162; ICD-10: C33–C34; referred to as ‘lung cancer’); skin (ICD-8/9: 172–173; ICD-10: C43–C44); prostate (ICD-8/9: 185; ICD-10: C61); bladder (ICD-8/9: 188; ICD-10: C67); kidney (ICD-8/9: 189; ICD-10: C64–C66, C68); brain (ICD-8/9: 191; ICD-10: C71); lymphoma (ICD-8/9: 200–203; ICD-10: C81–C90); and leukaemia (ICD-8: 204–207; ICD-9: 204–208; ICD-10: C91–C95).
We also grouped cancers into those for which tobacco smoking appears to have an aetiological role, and those for which it does not. This was based on the hypothesis that any influence of greater stature on smoking-related cancers would be obliterated by the substantially greater carcinogenic effects of tobacco smoke. We therefore anticipated a null or inverse relation between height and malignancies related to smoking (increased levels of smoking are apparent in shorter individuals), and a positive relation between height and malignancies not related to smoking. The following cancers were designated as being related to tobacco smoking: lip, oral cavity and pharynx (not including salivary gland; ICD-8/9: 140, 141, 143–149; ICD 10: C00–C06, C09–C15); oesophagus (see above); pancreas (see above); respiratory and intrathoracic organs (ICD-8: 160–163; ICD-9: 160–165; ICD-10: C30–C39); and urinary tract (ICD-8/9: 188–189; ICD-10: C64–C68). All other cancers were denoted as not being related to smoking. While updating these analyses an error was discovered in the coding of smoking-related cancers utilised in a report based on an earlier (27 year) follow-up of the Whitehall study [18]. In this previous publication [18], the following cancers were incorrectly assigned as being related to smoking: stomach (ICD-8/9: 151); small intestine, including duodenum (152); colon (153); rectum (154); liver (155); gallbladder (156); retroperitoneum and peritoneum (158); other and ill-defined sites within the digestive organs and peritoneum (159). This error has been corrected in the present analyses and we comment on the implications of this for the results in the Discussion.
data analyses
A total of 17 353 men (94.3% of the original cohort) had data for height and all potential covariates, and these form the sample for the analyses. The cause of death for 41 of these men was unknown and they were excluded from all analyses of cancer mortality outcomes. In all analyses we quantified the relation of cancer with height, categorised into four groups of similar sample size (<171, 171–175.9, 176–180.9 and 
181 cm) and as a continuous measure expressed in 5 cm (increasing) increments.
In analyses of baseline characteristics according to height categories, the prevalence of the former were adjusted for age (in 5-year groups) by the direct standardisation method. Trends in these proportions were examined using the Mantel–Haenszel test. For continuous variables, least squares means were used to present the age-adjusted means, with tests for trend across height groups computed by fitting a linear trend term. Mortality rates in each height category were calculated using person-years at risk. These rates were standardised for age at entry by the direct method using the study population as standard.
Hazard ratios and accompanying 95% confidence intervals (CIs) were computed for the relation of height with each mortality outcome using Cox's proportional hazards regression model [24] with follow-up period as the time scale. In these analyses, indicator variables were used for analyses according to the four height categories. P values for trend across these categories were computed by fitting a single linear term for the height categories. Analyses of height as a continuous measure used a single linear term. Statistical models were initially adjusted for age and then for other potential covariates. For the purposes of statistical adjustment, age, triceps skinfold thickness, plasma cholesterol, forced vital capacity (used in lung cancer analyses only), body mass index (BMI) and systolic blood pressure were fitted as continuous variables. The remaining factors were fitted as categorical variables.
During the baseline study, the physical activity enquiries on the questionnaire were modified. Levels of this behaviour were determined from either an item about travel activity [25] (administered to approximately two-thirds of men) or leisure activities [26] (administered to the remainder). Analyses of the height–total cancer relationship indicated that there was no confounding effect due to questionnaire type. Smoking status was grouped into four categories (never, ex-smoker, current pipe or cigar smoker, and current cigarette smoker) together with additional adjustment for the number of cigarettes smoked per day in current smokers. Existing disease at study entry was defined as a positive response to enquiries regarding a range of health conditions: use of blood pressure-lowering medication, unexplained weight loss in the preceding year, ischaemia [27], diabetes [28], intermittent claudication, physician-diagnosed heart problems or high blood pressure (one question), dyspnoea, and bronchitis.
In cohort studies of height and disease risk, morbidity, including cancer, whether detected or undetected at study induction, may cause a reduction in height. Referred to as shrinkage, examples include obstructive lung disease which may lead to kyphosis, a pronounced outward curvature of the spine. This would have the effect of decreasing the magnitude of an underlying positive relation (generally apparent in studies where cancer is the outcome of interest) and increasing the magnitude of an inverse relation (generally apparent in studies where total mortality and, particularly, cardiovascular disease are the outcomes of interest). To address this issue of reverse causality we adopted two approaches. First, to explore the role of clinically manifest illness we adjusted for disease at study entry (see above definition). Secondly, to explore the role of occult disease, we excluded deaths in the first 10 years of mortality surveillance and repeated our analyses. In so doing, we reasoned that a large proportion of deaths attributable to cancer, if present at study induction, would have occurred within this time frame. All statistical analyses were conducted using SAS computer software [29].
|
results |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionReferences |
|---|
In Table 1 the associations between height and a range of baseline characteristics are presented. In general, the more favourable levels of each characteristic were apparent in the taller men. The only exception was triceps skinfold thickness, which was positively related to height (Ptrend <0.001). In comparison to their shorter colleagues, taller men were younger, had greater lung function, lower BMI and cholesterol levels, and a reduced prevalence of existing disease at baseline, impaired glucose tolerance and diabetes (all Ptrend <0.001). Taller men were also less likely to be physically inactive, smoke cigarettes, or be without a partner (all Ptrend <0.001). There was a marked gradient in employment grade across the height groups, with the proportion of workers of low socioeconomic position in the tallest group, half that of men in the shortest group (Ptrend <0.001).
|
In Tables 2–5
hazards ratios with accompanying 95% CIs for height in relation to selected mortality outcomes are presented. There were 11 099 deaths during follow-up, 3101 (28%) of which were ascribed to cancer. Types of cancer are presented in decreasing order of incidence. Total mortality, included as an outcome for the purposes of completeness, was inversely related to stature [hazard ratio per 5 cm increase in height (95% CI): 0.96 (0.94, 0.97)], however, this gradient was effectively eliminated following full adjustment [0.99 (0.98, 1.01)]. In age-adjusted analyses (Table 2), height was positively associated with cancer of combined sites [1.03 (1.00, 1.05)], but not lung cancer mortality [1.00 (0.95, 1.06)]. However, a stronger positive association for carcinoma of the lung was seen following adjustment for socioeconomic position and other potential confounding and mediating variables, but was weakened somewhat after individuals who died in the first 10 years of follow-up were excluded from the analysis. There was also a suggestion that taller men experienced an elevated risk of prostate cancer, although statistical significance at conventional levels was lacking in several of the analyses.
|
|
|
|
In Table 3, stature showed no relation with either deaths due to lymphoma or pancreatic cancer. The suggestion of a reduced rate of stomach cancer deaths in taller men apparent in age-adjusted analyses [0.92 (0.83, 1.02)] was essentially lost when adjustment was made for all covariates. Point estimates for carcinoma of the bladder were raised in taller men, but in a non-stepwise manner (Ptrend = 0.48 in an age-adjusted analysis). These effects became stronger after deaths in the first 10 years of surveillance were dropped from the analysis. A similar observation was made when oesophageal cancer and leukaemia were the outcomes of interest (Table 4). A positive, non-incremental height–kidney cancer relation was apparent, although statistical power in this analysis was low owing to a small number of cases. There was little evidence of an association between stature and malignancy of the liver. Although based on a small number of deaths as evidenced by the very wide CIs, mortality rates from skin cancer (Table 5) were markedly elevated in taller men, showing by far the steepest gradient with height of all the cancer subtypes [1.36 (1.08, 1.71)]. When we combined site-specific cancers into those for which smoking has, and has not, been implicated in their aetiology, we found a similar positive relation with height for each after adjustment for all covariates.
We computed a test for heterogeneity in the height–cancer gradient across the fourteen site-specific end points, and found no indication of heterogeneity in either the age- (P = 0.18) or multiply-adjusted analyses (P = 0.43). With the suggestion of change in the strength of the height–malignancy association for some cancer sites after the exclusion of deaths in the first 10 years of mortality surveillance, we examined whether there was any effect modification according to follow-up duration. For 13 of the 14 individual cancer sites there was no such evidence (P value for non-proportionality of hazards ratios in multiply-adjusted analyses ranged between 0.24 and 0.91). The only exception was oesophageal cancer (P <0.001).
We also explored effect modification across strata of other key collateral data. These were smoking (stratified as: never, former, current), overweight/obesity (BMI <25 kg/m2 or 25 kg/m2) and prevalent disease at study induction (absence versus presence). We chose to examine only all-cancers as an outcome, because there was little evidence of statistical heterogeneity of the height–outcome gradient across the cancer subtypes (see above), suggesting our findings would hold for site-specific end points. Using tests for heterogeneity in the height–cancer association across strata of these variables, no evidence of effect modification was found (smoking: P = 0.27; adiposity: P = 0.28; prevalent disease: P = 0.97). Tables of these results are available from G.D.B. upon request.
|
discussion |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionReferences |
|---|
In the present study of a large, well characterised cohort of London-based, male government employees in which >3000 cancer deaths had occurred during follow-up, the relation of height to the 17 cancer mortality outcomes (14 sites) was essentially either positive or null. Thus, there was evidence that taller men experienced elevated mortality rates from cancer of combined sites, cancers related and unrelated to tobacco smoke, and cancer of the lung, the prostate, the bladder, the kidney, the skin and leukaemia. That the positive height–cancer gradients we found were generally of low magnitude is congruent with the conclusions of a recent review [14
]. In a separate analysis using data from the most recent follow-up of the Whitehall study, we have shown height to be inversely related cardiovascular disease and stroke, a finding that also generally accords with other studies [30–32].
comparison with other studies
A comprehensive review suggests that prostate and lung cancer are among the few malignancies to be consistently linked to stature in men (the association is positive) [14]. Subsequent reports do not, however, find an association [11, 33]. In the analyses reported herein, the direct height–lung cancer relation was further strengthened after controlling for covariates that included smoking, an adjustment that has not been made in all studies [17, 34, 35]. Similar observations were made when prostate cancer mortality was the outcome of interest. In contrast to these and several other neoplasms, oesophageal cancer is one of the few organ-specific carcinomas to demonstrate inverse relations with stature [17, 36, 37], although this finding is not universal [11, 38, 39]. While an inverse association was also found in an earlier (maximum 20 year) follow-up of the present cohort [17], in this current analysis based on a longer period of surveillance with a larger number of deaths, no overall height–oesophageal carcinoma association is apparent. As implicated by the findings of this earlier report and confirmed in further analyses here, the relation between this malignancy and height was dependent upon the duration of follow-up.
Although several investigators have related height to colorectal cancer [5, 39–41], given the apparent dissimilar aetiology of the neoplasms that comprise this end point [42], calls have been made to report separate exposure–disease effects for colon and rectal cancer. However, in the present study, in keeping with some others [38, 43, 44], we found no evidence to support a differential effect for height. Lymphoma and leukaemia are also often collapsed into a single outcome (haematopoetic cancers) in studies of stature and cancer. In the present analysis, a clear height–lymphoma gradient was not apparent, while there was evidence of a positive, incremental relation when leukaemia death was the outcome of interest. Other than an earlier follow-up of the present cohort [17], to our knowledge, only data from a Scottish study has been utilised to examine the height–leukaemia link [5] and there was no evidence of association. While the graded stature–cancer effect found for leukaemia was not apparent when we correlated height with either carcinoma of the bladder or kidney, several of the point estimates were nonetheless significantly raised. Of the three cohort studies [34, 38, 39] identified by Gunnell et al. [14] to have examined the influence of height on bladder cancer, none found an association. We are not aware of any publications subsequent to this review.
The literature reporting on the height–renal cancer relation is also modest in size and reveals mixed results, with two [38, 45] out of three cohort studies [38, 39, 45] finding a positive association, as we have. In one study where a sample of Norwegian men and women [45] was two orders of magnitude larger than any other previous investigation of risk, a positive, monotonic height–renal cancer gradient was found in men. Although collateral data were scarce in this study—there were, for instance, no data on socioeconomic position [15]—a subgroup analysis on those in which information on smoking history had been collected showed similar effects following statistical adjustment for this behaviour.
When we dichotomised cancers into those related to smoking and those not, similar positive associations with height were apparent for both outcomes after adjustment for covariates. An earlier follow-up of the Whitehall cohort found a differential association with height for these cancer groups [18], whereby a positive gradient with height was seen for cancers unrelated to smoking but not those related to smoking. Reasoning that the discrepant results offered by these respective reports might be a reflection of this coding error (see Materials and methods for description), we re-analysed data for the earlier follow-up. In analyses adjusted for age, employment grade and smoking, the association of height with both groups of cancers was, in fact, positive and of similar magnitude (Ptrend = 0.09 in both groups) to those found in the present paper.
plausible mechanisms
Rather than tall stature in itself ‘causing’ cancer, it is much more likely that stature is an environmental and/or genetic marker of risk [3]. In the present study we were unable to investigate genetic factors. With regard to environmental factors, by adjusting for socioeconomic position, BMI and smoking, it was apparent that these indices, at least when assessed in middle-age, are unlikely to have generated any height–cancer effects. Other possible explanations for a height–cancer relation include the link between stature and an array of early life exposures such as health status, social conditions and energy intake. Given the suggestion that childhood illness and poverty correlate with an increased risk of some adult cancers [46], but are themselves inversely related to adult height, they are largely ruled out as explanations for the positive height–malignancy gradients apparent in the present study. One exception may be stomach cancer, which was negatively related to height in age-adjusted analyses herein, as it was in multiply-adjusted models in another cohort [5]. In this Scottish study [5], the authors advanced Helicobacter pylori infection as the underlying cause of this gradient. Associated with suboptimal growth [47], this biological organism has also been identified as a predictor of stomach cancer in some [48], but not all [49], reports.
In the positive height–cancer gradients that predominate in the present analyses, stature may be a proxy of childhood food intake. It has long been recognised that animals who are fed a calorie-constrained, but otherwise micronutrient-balanced diet, subsequently experience lower cancer incidence than their overfed counterparts [50, 51]. In perhaps the only study of humans to date, this finding was replicated in individuals who participated in a inter-war dietary survey as children who were then followed over several decades into later life [52]. It has been proposed that underlying this relation may be reduced levels of insulin-like growth factors (IGF), which correlate directly with caloric intake (in animals) [53], height in children [54] and risk of colorectal and prostate cancer in adult humans [55]. IGF levels [56] and height [5, 12] are also inversely related to coronary heart disease in some studies. In support of an IGF–malignancy link, we found a positive stature–prostate cancer gradient; however, no association was observed for colorectal cancer.
A separate explanation for the positive height–cancer effects evident in the present analysis simply posits that taller people have larger bodies and therefore more cells that may potentially undergo malignant transformation to frank cancer. A priori, this was our hypothesis when we explored the link between stature and skin cancer. While our finding of a strong positive relation is supported by two [57, 58] of the three [38, 57, 58] prospective cohort studies that have also examined this association, it should nonetheless be viewed with caution owing to the very low number of deaths on which it is based (there were only two skin cancer deaths in the shortest group of men).
alternative explanations
Plausible non-causal explanations for some of the associations reported herein are chance, bias and confounding. A large number of statistical tests were necessarily conducted given the high number of health end points, so raising the problem of multiple comparisons where some positive results surface by chance alone. While selection bias is a potential concern in cohort studies, this is unlikely to be so in the Whitehall study, where loss to follow-up was very low (<1%). The principal impediment to interpretation of an association in observational epidemiology is confounding [59]. In contrast to some other studies [36, 45, 60], we controlled for a number of important adult risk factors for cancer: smoking, adiposity and socioeconomic disadvantage, thereby minimising confounding as a concern. It was notable that for some cancers (e.g. lung and all those related to smoking), a relation with height only emerged after adjustment for covariates, particularly employment grade. Given the positive association between height and employment grade, and the negative association between grade and some malignancies, such negative confounding is to be expected [18, 52].
conclusions
In this large, well-characterised cohort of government employees, there was evidence that taller men experienced elevated mortality rates from cancer of combined sites, the lung, the prostate, the bladder, the kidney, the skin and leukaemia. Exposures in early life which correlate with adult height, such as caloric intake, may, at least partially, explain these associations.
|
Acknowledgements |
|---|
We thank the civil servants who gave of their time to participate in the baseline survey in the 1960s, which was funded by the then Department of Health and Social Security and the Tobacco Research Council. M.J.S. is currently supported by the British Heart Foundation; M.G.M. by the Medical Research Council (MRC) and C.L. by a UK MRC Research Training Fellowship. When work on this manuscript began, G.D.B. was supported by a University of Copenhagen Senior Research Fellowship in Epidemiology; he is now funded by a Wellcome Advanced Training Fellowship.
Received for publication May 6, 2005. Revision received August 1, 2005. Accepted for publication August 8, 2005.
|
References |
|---|
|
TopAbstractintroductionmaterials and methodsresultsdiscussionReferences |
|---|
1. Kuh D, Ben Shlomo Y. A Lifecourse Approach to Chronic Disease Epidemiology. Oxford: Oxford Medical Publications 2004.
2. Galobardes B, Lynch JW, Davey Smith G. Childhood socioeconomic circumstances and cause-specific mortality in adulthood: systematic review and interpretation. Epidemiol Rev 2004; 26: 7–21.
3. Gunnell D. Can adult anthropometry be used as a ‘biomarker’ for prenatal and childhood exposures? Int J Epidemiol 2002; 31: 390–394.
4. Marmot MG, Shipley MJ, Rose G. Inequalities in death-specific explanations of a general pattern? Lancet 1984; 1: 1003–1006.[CrossRef][Web of Science][Medline]
5. Davey Smith G, Hart C, Upton M. Height and risk of death among men and women: aetiological implications of associations with cardiorespiratory disease and cancer mortality. J Epidemiol Community Health 2000; 54: 97–103.
6. Engeland A, Bjorge T, Selmer RM, Tverdal A. Height and body mass index in relation to total mortality. Epidemiology 2003; 14: 293–299.[CrossRef][Web of Science][Medline]
7. Allebeck P, Bergh C. Height, body mass index and mortality: do social factors explain the association? Public Health 1992; 106: 375–382.[CrossRef][Web of Science][Medline]
8. Waaler HT. Height, weight and mortality. The Norwegian experience. Acta Med Scand Suppl 1984; 679: 1–56.[Medline]
9. Jousilahti P, Tuomilehto J, Vartiainen E et al. Relation of adult height to cause-specific and total mortality: a prospective follow-up study of 31,199 middle-aged men and women in Finland. Am J Epidemiol 2000; 151: 1112–1120.
10. Liao Y, McGee DL, Cao G, Cooper RS. Short stature and risk of mortality and cardiovascular disease: negative findings from the NHANES I epidemiologic follow-up study. J Am Coll Cardiol 1996; 27: 678–682.[Abstract]
11. Song YM, Davey Smith G, Sung J. Adult height and cause-specific mortality: a large prospective study of South Korean men. Am J Epidemiol 2003; 158: 479–485.
12. Langenberg C, Shipley MJ, Batty GD, Marmot MG. Adult socioeconomic position and the association between height and coronary heart disease mortality: findings from 33 years of follow-up in the whitehall study. Am J Public Health 2005; 95: 628–632.
13. Giovannucci E, Rimm EB, Liu Y, Willett WC. Height, predictors of C-peptide and cancer risk in men. Int J Epidemiol 2004; 33: 217–225.
14. Gunnell D, Okasha M, Davey Smith G et al. Height, leg length, and cancer risk: a systematic review. Epidemiol Rev 2001; 23: 313–342.
15. Batty GD. Confounding effect of socioeconomic position in the study of height in relation to prostate cancer risk. Br J Cancer 2004; 90: 1875.[Web of Science][Medline]
16. Reid DD, Hamilton PJS, McCartney P et al. Cardiorespiratory disease and diabetes among middle-aged male civil servants. Lancet 1974; i: 469–473.
17. Leon DA, Davey Smith G, Shipley M, Strachan D. Adult height and mortality in London: early life, socioeconomic confounding, or shrinkage? J Epidemiol Community Health 1995; 49: 5–9.
18. Davey Smith G, Shipley M, Leon DA. Height and mortality from cancer among men: prospective observational study. BMJ 1998; 317: 1351–1352.
19. Davey Smith G, Leon D, Shipley MJ, Rose G. Socioeconomic differentials in cancer among men. Int J Epidemiol 1991; 20: 339–345.
20. Batty GD, Shipley MJ, Jarrett RJ et al. Obesity and overweight in relation to organ-specific cancer mortality in London (UK): findings from the original Whitehall study. Int J Obes 2005; In press.
21. Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death, 8th revision. Geneva: World Health Organisation 1967.
22. Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death, 9th revision). Geneva: World Health Organisation 1977.
23. International Statistical Classification of Diseases and Related Health Problems, 10th revision. Geneva: World Health Organisation 1992.
24. Cox DR. Regression models and life-tables. J R Stat Soc 1972; 34: 187–220.
25. Batty GD, Shipley M, Marmot M, Davey Smith G. Physical activity and cause-specific mortality in men: further evidence from the Whitehall study. Eur J Epidemiol 2002; 17: 863–869.
26. Davey Smith G, Shipley MJ, Batty GD et al. Physical activity and cause-specific mortality in the Whitehall study. Public Health 2000; 114: 308–315.[Web of Science][Medline]
27. Batty GD, Shipley MJ, Marmot M, Davey Smith G. Leisure time physical activity and coronary heart disease mortality in men symptomatic or asymptomatic for ischaemia: evidence from the Whitehall study. J Public Health Med 2003; 25: 190–196.
28. Batty GD, Shipley MJ, Marmot M, Davey Smith G. Physical activity and cause-specific mortality in men with Type 2 diabetes/impaired glucose tolerance: evidence from the Whitehall study. Diabet Med 2002; 19: 580–588.[CrossRef][Web of Science][Medline]
29. SAS Institute Inc. SAS/STAT® User's Guide, Version 6, Vol. 1 & 2. Cary, NC: SAS Institute Inc. 1989.
30. Morris JN, Marr JW, Clayton DG. Diet and heart: a postscript. BMJ 1977; 2: 1307–1314.
31. Wannamethee SG, Shaper AG, Whincup PH, Walker M. Adult height, stroke, and coronary heart disease. Am J Epidemiol 1998; 148: 1069–1076.
32. Yarnell JW, Limb ES, Layzell JM, Baker IA. Height: a risk marker for ischaemic heart disease: prospective results from the Caerphilly and Speedwell Heart Disease Studies. Eur Heart J 1992; 13: 1602–1605.
33. Rodriguez C, Patel AV, Calle EE et al. Body mass index, height, and prostate cancer mortality in two large cohorts of adult men in the United States. Cancer Epidemiol Biomarkers Prev 2001; 10: 345–353.
34. Albanes D, Jones Y, Schatzkin A et al. Adult stature and risk of cancer. Cancer Res 1988; 48: 1658–1662.
35. Drinkard CR, Sellers TA, Potter JD et al. Association of body mass index and body fat distribution with risk of lung cancer in older women. Am J Epidemiol 1995; 142: 600–607.
36. Engeland A, Tretli S, Bjorge T. Height and body mass index in relation to esophageal cancer; 23-year follow-up of two million Norwegian men and women. Cancer Causes Control 2004; 15: 837–843.[CrossRef][Web of Science][Medline]
37. Tran GD, Sun XD, Abnet CC et al. Prospective study of risk factors for esophageal and gastric cancers in the Linxian general population trial cohort in China. Int J Cancer 2005; 113: 456–463.[CrossRef][Web of Science][Medline]
38. Whittemore AS, Paffenbarger RS Jr, Anderson K, Lee JE. Early precursors of site-specific cancers in college men and women. J Natl Cancer Inst 1985; 74: 43–51.[Web of Science][Medline]
39. Tulinius H, Sigfusson N, Sigvaldason H et al. Risk factors for malignant diseases: a cohort study on a population of 22,946 Icelanders. Cancer Epidemiol Biomarkers Prev 1997; 6: 863–873.[Abstract]
40. van Wayenburg CA, van der Schouw YT, van Noord PA, Peeters PH. Age at menopause, body mass index, and the risk of colorectal cancer mortality in the Dutch Diagnostisch Onderzoek Mammacarcinoom (DOM) cohort. Epidemiology 2000; 11: 304–308.[CrossRef][Web of Science][Medline]
41. Kaaks R, Toniolo P, Akhmedkhanov A et al. Serum C-peptide, insulin-like growth factor (IGF)-I, IGF-binding proteins, and colorectal cancer risk in women. J Natl Cancer Inst 2000; 92: 1592–1600.
42. Wei EK, Giovannucci E, Wu K et al. Comparison of risk factors for colon and rectal cancer. Int J Cancer 2004; 108: 433–442.[CrossRef][Web of Science][Medline]
43. Chyou PH, Nomura AM, Stemmermann GN. A prospective study of weight, body mass index and other anthropometric measurements in relation to site-specific cancers. Int J Cancer 1994; 57: 313–317.[Web of Science][Medline]
44. Le Marchand L, Wilkens LR, Mi MP. Obesity in youth and middle age and risk of colorectal cancer in men. Cancer Causes Control 1992; 3: 349–354.[CrossRef][Web of Science][Medline]
45. Bjorge T, Tretli S, Engeland A. Relation of height and body mass index to renal cell carcinoma in two million Norwegian men and women. Am J Epidemiol 2004; 160: 1168–1176.
46. Davey Smith G, Hart C, Blane D, Hole D. Adverse socioeconomic conditions in childhood and cause specific adult mortality: prospective observational study. BMJ 1998; 316: 1631–1635.
47. Patel P, Mendall MA, Khulusi S et al. Helicobacter pylori infection in childhood: risk factors and effect on growth. BMJ 1994; 309: 1119–1123.
48. Forman D, Newell DG, Fullerton F et al. Association between infection with Helicobacter pylori and risk of gastric cancer: evidence from a prospective investigation. BMJ 1991; 302: 1302–1305.
49. Webb PM, Yu MC, Forman D et al. An apparent lack of association between Helicobacter pylori infection and risk of gastric cancer in China. Int J Cancer 1996; 67: 603–607.[CrossRef][Web of Science][Medline]
50. Hart RW, Turturro A. Dietary restrictions and cancer. Environ Health Perspect 1997; 105 (Suppl 4): 989–992.
51. Shimokawa I, Higami Y. Effect of Dietary Restriction on Pathological Processes. Modulation of Aging Processes by Dietary Restriction. London: CRC Press 1994.
52. Frankel S, Gunnell DJ, Peters TJ et al. Childhood energy intake and adult mortality from cancer: the Boyd Orr Cohort Study. BMJ 1998; 316: 499–504.
53. Ruggeri BA, Klurfeld DM, Kritchevsky D, Furlanetto RW. Caloric restriction and 7,12-dimethylbenz(a)anthracene-induced mammary tumor growth in rats: alterations in circulating insulin, insulin-like growth factors I and II, and epidermal growth factor. Cancer Res 1989; 49: 4130–4134.
54. Juul A, Bang P, Hertel NT et al. Serum insulin-like growth factor-I in 1030 healthy children, adolescents, and adults: relation to age, sex, stage of puberty, testicular size, and body mass index. J Clin Endocrinol Metab 1994; 78: 744–752.[Abstract]
55. Renehan AG, Zwahlen M, Minder C et al. Insulin-like growth factor (IGF)-I, IGF binding protein-3, and cancer risk: systematic review and meta-regression analysis. Lancet 2004; 363: 1346–1353.[CrossRef][Web of Science][Medline]
56. Laughlin GA, Barrett-Connor E, Criqui MH, Kritz-Silverstein D. The prospective association of serum insulin-like growth factor I (IGF-I) and IGF-binding protein-1 levels with all cause and cardiovascular disease mortality in older adults: the Rancho Bernardo Study. J Clin Endocrinol Metab 2004; 89: 114–120.
57. Thune I, Olsen A, Albrektsen G, Tretli S. Cutaneous malignant melanoma: association with height, weight and body-surface area. a prospective study in Norway. Int J Cancer 1993; 55: 555–561.[Web of Science][Medline]
58. Veierod MB, Thelle DS, Laake P. Diet and risk of cutaneous malignant melanoma: a prospective study of 50,757 Norwegian men and women. Int J Cancer 1997; 71: 600–604.[CrossRef][Web of Science][Medline]
59. von Elm E, Egger M. The scandal of poor epidemiological research. BMJ 2004; 329: 868–869.
60. Engeland A, Tretli S, Bjorge T. Height, body mass index, and prostate cancer: a follow-up of 950000 Norwegian men. Br J Cancer 2003; 89: 1237–1242.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. D. Batty, F. Barzi, M. Woodward, K. Jamrozik, J. Woo, H. C. Kim, H. Ueshima, R. R. Huxley, and for the Asia Pacific Cohort Studies Collaboration Adult height and cancer mortality in Asia: the Asia Pacific Cohort Studies Collaboration Ann. Onc., November 4, 2009; (2009) mdp363v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G D Batty, Y Wang, S W Brouilette, P Shiels, C Packard, J Moore, N J Samani, and I Ford Socioeconomic status and telomere length: the West of Scotland Coronary Prevention Study J Epidemiol Community Health, October 1, 2009; 63(10): 839 - 841. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Y. Lee, F. Barzi, M. Woodward, G D. Batty, G. G Giles, J. W. Wong, K. Jamrozik, T. H. Lam, H. Ueshima, H. C. Kim, et al. Adult height and the risks of cardiovascular disease and major causes of death in the Asia-Pacific region: 21 000 deaths in 510 000 men and women Int. J. Epidemiol., August 1, 2009; 38(4): 1060 - 1071. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Batty, M. Kivimaki, D. Morrison, R. Huxley, G. D. Smith, R. Clarke, M. G. Marmot, and M. J. Shipley Risk Factors for Pancreatic Cancer Mortality: Extended Follow-up of the Original Whitehall Study Cancer Epidemiol. Biomarkers Prev., February 1, 2009; 18(2): 673 - 675. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zuccolo, R. Harris, D. Gunnell, S. Oliver, J. A. Lane, M. Davis, J. Donovan, D. Neal, F. Hamdy, R. Beynon, et al. Height and Prostate Cancer Risk: A Large Nested Case-Control Study (ProtecT) and Meta-analysis Cancer Epidemiol. Biomarkers Prev., September 1, 2008; 17(9): 2325 - 2336. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Batty, K Modig Wennerstad, G Davey Smith, D Gunnell, I. Deary, P Tynelius, and F Rasmussen IQ in early adulthood and later cancer risk: cohort study of one million Swedish men Ann. Onc., January 1, 2007; 18(1): 21 - 28. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||