Life-course body mass index trajectories and blood pressure in mid life in two British birth cohorts： stronger associations in the later-born generation

Leah Li，Rebecca Hardy，Diana Kuh，Chris Power

1. Centre for Paediatric Epidemiology & Biostatistics，

2. MRC Unit for Lifelong Health and Ageing，University College London，UK

專家推薦

Life-course body mass index trajectories and blood pressure in mid life in two British birth cohorts： stronger associations in the later-born generation

Leah Li1*，Rebecca Hardy2，Diana Kuh2，Chris Power1

1. Centre for Paediatric Epidemiology & Biostatistics，

2. MRC Unit for Lifelong Health and Ageing，University College London，UK

學(xué)科：流行病與衛(wèi)生統(tǒng)計(jì)學(xué)

推薦專家：鄒延峰副教授（安徽醫(yī)科大學(xué)公共衛(wèi)生學(xué)院）

推薦論文：Leah Li，Rebecca Hardy，Diana Kuh，et al. Life-course body mass index trajectories and blood pressure in mid life in two British birth cohorts： stronger associations in the later-born generation ［J］. International Journal of Epidemiology，Accepted 20 May 2015，

10.1093/ije/dyv106

·專家點(diǎn)評·

近年來肥胖和體質(zhì)指數(shù)（BMI）快速增加，它們對心血管疾病有何影響知之甚少。該研究在出生隊(duì)列中調(diào)查了生命過程中的BMI軌跡與成年人的血壓（BP）之間的關(guān)系。作者使用了1946年（樣本量4787）和1958年（樣本量16820）的兩個英國出生隊(duì)列。調(diào)查結(jié)果發(fā)現(xiàn)BMI軌跡和BP之間存在關(guān)聯(lián)，在后出生隊(duì)列中關(guān)聯(lián)更強(qiáng)；BMI軌跡和BP之間的關(guān)聯(lián)可能被其他因素影響，如飲食和吸煙。該研究結(jié)果為研究BMI軌跡和BP之間關(guān)聯(lián)及高血壓危險因素隨著時間推移的改變情況提供了重要的理論依據(jù)和指導(dǎo)意義。

優(yōu)點(diǎn)：1）前瞻性的大樣本研究；2）使用了兩個出生隊(duì)列；3）使用了Joint multivariate response models；4）數(shù)據(jù)分析詳實(shí)，合適的分層分析，結(jié)果展示清楚；

缺點(diǎn)：混雜因素對結(jié)果的影響還需進(jìn)一步控制，如對高血壓的治療情況可能對結(jié)果產(chǎn)生一定的影響。

總的來講，高血壓的危險因素可能隨著時間的推移而改變的情況應(yīng)該需要更多的研究來探討，在后出生隊(duì)列中BMI軌跡和BP之間關(guān)聯(lián)更強(qiáng)的可能原因也應(yīng)該進(jìn)一步探討，混雜因素應(yīng)該控制的更嚴(yán)。

Background： Little is known about the impact of recent increases in obesity and more rapid gains in body mass index（BMI）on cardiovascular risk factors. We investigated life-course BMI trajectories associations with adult blood pressure（BP）across two generations.

Methods： We used the the 1946 and 1958 British birth cohorts. Joint multivariate response models were fitted to longitudinal BMI measures ［7，11，16，20，26，36，43 and 50 y（years）： 1946 cohort，n=4787； 7，11，16，23，33 and 45 y： 1958 cohort，n=16820］ and midadult BP. We adopted linear spline models with random coefficients to characterize childhood and adult BMI slopes.

Results： Mean systolic BP（SBP）decreased from the earlier- to later-born cohort by 2.8 mmHg in females，not males； mean diastolic BP（DBP）decreased by 3.2-3.3 mmHg（both sexes）. Adult BMI was higher in the later- than the earlier-born cohort by 1.3-1.8 kg/m2，slopes of BMI trajectory were steeper from early adulthood and associations with adult BP were stronger. Associations between adult BMI and SBP were stronger in the later-born cohort. For males，childhood BMI slope was associated with SBP only in the later-born cohort； the association for adult BMI slope was stronger in the later-born cohort： correlation coefficient r=0.28 ［95% confidence interval（CI）： 0.25，0.33］ versus 0.13（0.06，0.20）. For females，childhood slope was associated with SBP in both cohorts； adult slope was associated with SBP only in the 1958 cohort ［r=0.34（0.31，0.37）］. Patterns of child-to-adult BMI associations were similar in relation to DBP.

Conclusions： BP did not increase between two generations born 12 y apart despite higher BMI levels. A stronger association between BMI trajectory and BP in the laterborn cohort suggests that BMI-related effects may have been offset by improvements in other factors linked to BP，such as diet and smoking.

Cohort study； blood pressure； BMI trajectories； joint modeling； life course

Key Messages

· Mean values of systolic or diastolic blood pressure（BP）did not increase between two generations born 12 years apart，despite the steeper slope for BMI changes from early adulthood and higher levels of body mass index（BMI）in mid adulthood in the later- than the earlier-born cohort.

· A stronger association between life-course BMI trajectories and higher BP levels in the later-born cohort suggests that BMI-related effects may have been offset by improvements in other factors linked to BP over successive generations in the UK.

責(zé)任編輯：王帥帥

* Corresponding author. Centre for Paediatric Epidemiology and Biostatistics，UCL Institute of Child Health，30 Guilford Street，London WC1N 1EH，UK. E-mail： leah.li@ucl.ac.uk.