Fitness versus fatness: Disentangling their effects on disease outcomes and estimating the population burden of disease

超重和缺乏运动是心血管疾病和糖尿病的两个主要风险因素。随着人口超重水平的不断提高，各国政府越来越多地倡导旨在提高体力活动水平或以其他方式减轻体重的公共卫生措施。关于这些因素中的哪些是不良健康结果的关键预后因素，已经进行了大量研究，但由于研究结果持续缺乏明确性，导致对预防健康战略和改变人口生活方式的建议的方向存在不确定性。此外，特别是超重的风险，在科学和非专业文献中都被指责被夸大了。这通常是由于很难适当地处理时变的混杂因素，这些因素也是中间因素(如高血压)。到目前为止，还没有研究对缺乏体力活动和超重对心血管事件和糖尿病的联合影响进行仔细的纵向模拟，同时考虑到高血压或动脉粥样硬化等因素的影响，这些因素既影响体力活动和超重，也影响疾病，同时也是这些危险因素的后果。众所周知，标准统计方法在这些情况下会产生有偏见的估计，但我们将应用最近开发的统计技术来提供对这些影响的更好的估计。在统计建模阶段之后，我们将对糖尿病、心血管疾病和死亡率与给定的超重和缺乏体力活动水平相关的负担进行建模。这些将与超重和缺乏活动的人口水平相结合，以确定可归因于这些风险因素的当前和未来疾病负担的比例。

Overweight and physical inactivity are two of the major risk factors for cardiovascular disease and diabetes. With increasing population levels of overweight, governments are increasingly advocating public health measures aimed at increasing physical activity levels or otherwise decreasing weight. There has been much research concerning which of these factors is the key prognostic factor for adverse health outcomes, but an ongoing lack of clarity of research findings has led to uncertainty as to the direction of recommendations for preventive health strategies and population lifestyle changes. In addition, the risks of overweight, in particular, have been accused of being exaggerated in both the scientific and lay literature. This is often due to the difficulty of dealing appropriately with time varying confounders which are also intermediate factors (such as hypertension). To date, no studies have performed careful longitudinal modelling of the joint effects of physical inactivity and overweight on cardiovascular events and diabetes while taking into account the effects of factors, such as hypertension or atherosclerosis, that influence both physical activity and overweight as well as disease, and simultaneously are consequences of these risk factors. Standard statistical methods are known to produce biased estimates in these situations but we will apply more recently developed statistical techniques to provide much improved estimation of these effects. After the statistical modelling stage, we will model the burden of diabetes, cardiovascular disease and mortality associated with given levels of overweight and physical inactivity. These will be combined with population levels of overweight and inactivity to identify the fraction of the current and future burden of disease attributable to these risk factors.