Development and validation of new flexible statistical methods for etiological and prognostic studies

我的研究计划将继续侧重于发展和验证用于分析卫生数据的新统计方法。特别是，与我的学员和合作者一起，我们将致力于制定新的方法，能够解释导致各种疾病发生或影响疾病进展的过程的复杂性。还将开发专门的方法，以便更有效地检测药物的潜在不良反应。这个问题越来越具有社会重要性，因为最近被广泛报道的非常流行的止痛药被发现使心脏病发作和心脏病死亡的风险增加了一倍，这戏剧性地说明了这一点。检测这些不良反应并准确评估其影响具有挑战性，因为药物使用的剂量、模式和持续时间因受试者和时间而异。此外，目前尚不清楚在药物使用和其负面影响之间是否存在以及存在何种延迟。然而，最近在计算资源和统计理论方面的加速进展将使我们能够制定灵活的方法，避免限制性假设，并反映潜在和影响人类健康的过程的复杂性。

My research programme will continue focusing on the development and validation of new statistical methods for the analysis of health data. In particular, with my trainees and collaborators, we will aim at elaborating new methods that will be able to account for the complexity of the processes leading to the occurrence of various diseases or affecting disease progression. Specialized methods will also be developed to allow a more efficient detection of potential adverse effects of medications. This issue is of increasing societal importance as it has been recently dramatically illustrated by highly publicized withdrawals of very popular pain relief drugs that were found to double the risks of heart attacks and deaths due to heart disease. Detection of such adverse effects and accurate assessment of their impact are challenging because dose, patterns and duration of drug use vary across subjects and over time. Moreover, it is unclear whether and what delay may exist between drug use and its negative effects. However, recent accelerated progress in both computational resources and statistical theory will allow us to develop flexible methods that avoid restrictive assumptions and reflect the complexity of the processes underlying and affecting human health.