Contribution of modifiable risk factors to social inequalities in type 2 diabetes: Prospective Whitehall II cohort study

Silvia Stringhini, A. Tabák, Tasnime N. Akbaraly, Séverine Sabia, Martin J. Shipley, Michael G. Marmot, Eric J. Brunner, G. David Batty, Pascal Bovet, Mika Kivimäki

Research output: Contribution to journalArticle

84 Citations (Scopus)

Abstract

Objective: To assess the contribution of modifiable risk factors to social inequalities in the incidence of type 2 diabetes when these factors are measured at study baseline or repeatedly over follow-up and when long term exposure is accounted for. Design: Prospective cohort study with risk factors (health behaviours (smoking, alcohol consumption, diet, and physical activity), body mass index, and biological risk markers (systolic blood pressure, triglycerides and high density lipoprotein cholesterol)) measured four times and diabetes status assessed seven times between 1991-93 and 2007-09. Setting: Civil service departments in London (Whitehall II study). Participants: 7237 adults without diabetes (mean age 49.4 years; 2196 women). Main outcome measures: Incidence of type 2 diabetes and contribution of risk factors to its association with socioeconomic status. Results: Over a mean follow-up of 14.2 years, 818 incident cases of diabetes were identified. Participants in the lowest occupational category had a 1.86-fold (hazard ratio 1.86, 95% confidence interval 1.48 to 2.32) greater risk of developing diabetes relative to those in the highest occupational category. Health behaviours and body mass index explained 33% (-1% to 78%) of this socioeconomic differential when risk factors were assessed at study baseline (attenuation of hazard ratio from 1.86 to 1.51), 36% (22% to 66%) when they were assessed repeatedly over the follow-up (attenuated hazard ratio 1.48), and 45% (28% to 75%) when long term exposure over the follow-up was accounted for (attenuated hazard ratio 1.41). With additional adjustment for biological risk markers, a total of 53% (29% to 88%) of the socioeconomic differential was explained (attenuated hazard ratio 1.35, 1.05 to 1.72). Conclusions: Modifiable risk factors such as health behaviours and obesity, when measured repeatedly over time, explain almost half of the social inequalities in incidence of type 2 diabetes. This is more than was seen in previous studies based on single measurement of risk factors.

Original languageEnglish
Article numbere5452
JournalBritish Medical Journal
Volume345
Issue number7875
DOIs
Publication statusPublished - Sep 22 2012

Fingerprint

Type 2 Diabetes Mellitus
Cohort Studies
Health Behavior
Incidence
Body Mass Index
Biomarkers
Blood Pressure
Risk Adjustment
Social Class
Alcohol Drinking
HDL Cholesterol
Triglycerides
Obesity
Smoking
Outcome Assessment (Health Care)
Prospective Studies
Confidence Intervals
Exercise
Diet

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Contribution of modifiable risk factors to social inequalities in type 2 diabetes : Prospective Whitehall II cohort study. / Stringhini, Silvia; Tabák, A.; Akbaraly, Tasnime N.; Sabia, Séverine; Shipley, Martin J.; Marmot, Michael G.; Brunner, Eric J.; Batty, G. David; Bovet, Pascal; Kivimäki, Mika.

In: British Medical Journal, Vol. 345, No. 7875, e5452, 22.09.2012.

Research output: Contribution to journalArticle

Stringhini, S, Tabák, A, Akbaraly, TN, Sabia, S, Shipley, MJ, Marmot, MG, Brunner, EJ, Batty, GD, Bovet, P & Kivimäki, M 2012, 'Contribution of modifiable risk factors to social inequalities in type 2 diabetes: Prospective Whitehall II cohort study', British Medical Journal, vol. 345, no. 7875, e5452. https://doi.org/10.1136/bmj.e5452
Stringhini, Silvia ; Tabák, A. ; Akbaraly, Tasnime N. ; Sabia, Séverine ; Shipley, Martin J. ; Marmot, Michael G. ; Brunner, Eric J. ; Batty, G. David ; Bovet, Pascal ; Kivimäki, Mika. / Contribution of modifiable risk factors to social inequalities in type 2 diabetes : Prospective Whitehall II cohort study. In: British Medical Journal. 2012 ; Vol. 345, No. 7875.
@article{e1035d6d601b479497a05ae31f30ea20,
title = "Contribution of modifiable risk factors to social inequalities in type 2 diabetes: Prospective Whitehall II cohort study",
abstract = "Objective: To assess the contribution of modifiable risk factors to social inequalities in the incidence of type 2 diabetes when these factors are measured at study baseline or repeatedly over follow-up and when long term exposure is accounted for. Design: Prospective cohort study with risk factors (health behaviours (smoking, alcohol consumption, diet, and physical activity), body mass index, and biological risk markers (systolic blood pressure, triglycerides and high density lipoprotein cholesterol)) measured four times and diabetes status assessed seven times between 1991-93 and 2007-09. Setting: Civil service departments in London (Whitehall II study). Participants: 7237 adults without diabetes (mean age 49.4 years; 2196 women). Main outcome measures: Incidence of type 2 diabetes and contribution of risk factors to its association with socioeconomic status. Results: Over a mean follow-up of 14.2 years, 818 incident cases of diabetes were identified. Participants in the lowest occupational category had a 1.86-fold (hazard ratio 1.86, 95{\%} confidence interval 1.48 to 2.32) greater risk of developing diabetes relative to those in the highest occupational category. Health behaviours and body mass index explained 33{\%} (-1{\%} to 78{\%}) of this socioeconomic differential when risk factors were assessed at study baseline (attenuation of hazard ratio from 1.86 to 1.51), 36{\%} (22{\%} to 66{\%}) when they were assessed repeatedly over the follow-up (attenuated hazard ratio 1.48), and 45{\%} (28{\%} to 75{\%}) when long term exposure over the follow-up was accounted for (attenuated hazard ratio 1.41). With additional adjustment for biological risk markers, a total of 53{\%} (29{\%} to 88{\%}) of the socioeconomic differential was explained (attenuated hazard ratio 1.35, 1.05 to 1.72). Conclusions: Modifiable risk factors such as health behaviours and obesity, when measured repeatedly over time, explain almost half of the social inequalities in incidence of type 2 diabetes. This is more than was seen in previous studies based on single measurement of risk factors.",
author = "Silvia Stringhini and A. Tab{\'a}k and Akbaraly, {Tasnime N.} and S{\'e}verine Sabia and Shipley, {Martin J.} and Marmot, {Michael G.} and Brunner, {Eric J.} and Batty, {G. David} and Pascal Bovet and Mika Kivim{\"a}ki",
year = "2012",
month = "9",
day = "22",
doi = "10.1136/bmj.e5452",
language = "English",
volume = "345",
journal = "British Medical Journal",
issn = "0959-8146",
publisher = "BMJ Publishing Group",
number = "7875",

}

TY - JOUR

T1 - Contribution of modifiable risk factors to social inequalities in type 2 diabetes

T2 - Prospective Whitehall II cohort study

AU - Stringhini, Silvia

AU - Tabák, A.

AU - Akbaraly, Tasnime N.

AU - Sabia, Séverine

AU - Shipley, Martin J.

AU - Marmot, Michael G.

AU - Brunner, Eric J.

AU - Batty, G. David

AU - Bovet, Pascal

AU - Kivimäki, Mika

PY - 2012/9/22

Y1 - 2012/9/22

N2 - Objective: To assess the contribution of modifiable risk factors to social inequalities in the incidence of type 2 diabetes when these factors are measured at study baseline or repeatedly over follow-up and when long term exposure is accounted for. Design: Prospective cohort study with risk factors (health behaviours (smoking, alcohol consumption, diet, and physical activity), body mass index, and biological risk markers (systolic blood pressure, triglycerides and high density lipoprotein cholesterol)) measured four times and diabetes status assessed seven times between 1991-93 and 2007-09. Setting: Civil service departments in London (Whitehall II study). Participants: 7237 adults without diabetes (mean age 49.4 years; 2196 women). Main outcome measures: Incidence of type 2 diabetes and contribution of risk factors to its association with socioeconomic status. Results: Over a mean follow-up of 14.2 years, 818 incident cases of diabetes were identified. Participants in the lowest occupational category had a 1.86-fold (hazard ratio 1.86, 95% confidence interval 1.48 to 2.32) greater risk of developing diabetes relative to those in the highest occupational category. Health behaviours and body mass index explained 33% (-1% to 78%) of this socioeconomic differential when risk factors were assessed at study baseline (attenuation of hazard ratio from 1.86 to 1.51), 36% (22% to 66%) when they were assessed repeatedly over the follow-up (attenuated hazard ratio 1.48), and 45% (28% to 75%) when long term exposure over the follow-up was accounted for (attenuated hazard ratio 1.41). With additional adjustment for biological risk markers, a total of 53% (29% to 88%) of the socioeconomic differential was explained (attenuated hazard ratio 1.35, 1.05 to 1.72). Conclusions: Modifiable risk factors such as health behaviours and obesity, when measured repeatedly over time, explain almost half of the social inequalities in incidence of type 2 diabetes. This is more than was seen in previous studies based on single measurement of risk factors.

AB - Objective: To assess the contribution of modifiable risk factors to social inequalities in the incidence of type 2 diabetes when these factors are measured at study baseline or repeatedly over follow-up and when long term exposure is accounted for. Design: Prospective cohort study with risk factors (health behaviours (smoking, alcohol consumption, diet, and physical activity), body mass index, and biological risk markers (systolic blood pressure, triglycerides and high density lipoprotein cholesterol)) measured four times and diabetes status assessed seven times between 1991-93 and 2007-09. Setting: Civil service departments in London (Whitehall II study). Participants: 7237 adults without diabetes (mean age 49.4 years; 2196 women). Main outcome measures: Incidence of type 2 diabetes and contribution of risk factors to its association with socioeconomic status. Results: Over a mean follow-up of 14.2 years, 818 incident cases of diabetes were identified. Participants in the lowest occupational category had a 1.86-fold (hazard ratio 1.86, 95% confidence interval 1.48 to 2.32) greater risk of developing diabetes relative to those in the highest occupational category. Health behaviours and body mass index explained 33% (-1% to 78%) of this socioeconomic differential when risk factors were assessed at study baseline (attenuation of hazard ratio from 1.86 to 1.51), 36% (22% to 66%) when they were assessed repeatedly over the follow-up (attenuated hazard ratio 1.48), and 45% (28% to 75%) when long term exposure over the follow-up was accounted for (attenuated hazard ratio 1.41). With additional adjustment for biological risk markers, a total of 53% (29% to 88%) of the socioeconomic differential was explained (attenuated hazard ratio 1.35, 1.05 to 1.72). Conclusions: Modifiable risk factors such as health behaviours and obesity, when measured repeatedly over time, explain almost half of the social inequalities in incidence of type 2 diabetes. This is more than was seen in previous studies based on single measurement of risk factors.

UR - http://www.scopus.com/inward/record.url?scp=84866604182&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866604182&partnerID=8YFLogxK

U2 - 10.1136/bmj.e5452

DO - 10.1136/bmj.e5452

M3 - Article

C2 - 22915665

AN - SCOPUS:84866604182

VL - 345

JO - British Medical Journal

JF - British Medical Journal

SN - 0959-8146

IS - 7875

M1 - e5452

ER -