Pathway Analysis of Treatments in Healthcare

ongoing

“What is the drug of choice for the given condition?” is the most important question raised by patients and clinicians at the point of care. The use of systematic reviews to answer this question, accompanied often with pairwise or network meta-analysis, ensures that decisions are informed by the best available evidence. However, in many healthcare areas we are moving towards the era of treatment pathways, in which not only a single treatment but a sequence of treatments can be the optimal choice for patients. Despite the important advances in methodology of evidence synthesis, treatment adaptations, such as modifying dose, switching, adding or discontinuing treatments are rarely informed by evidence. The overarching research aim of this project is to develop and test a framework for deriving the optimal treatment pathway. To this aim, I have set the following objectives: a) develop the statistical framework to compare alternative treatment pathways and b) evaluate the framework by developing methods to make it useful and applicable to clinical practice. I have devised the concept that treatment pathways can be seen as a combination of components, in which, however, sequence needs to be taken into account. As a first step, I will thus use component network meta-analysis. I will subsequently incorporate advances from clinical trials, dose-response, time-course models and prediction models, among others. I will develop PATH using as an example schizophrenia, where I will model treatment switching, increasing dose and/or adding treatment and will produce relative effects between all alternative treatment pathways. PATH will provide a much-needed methodological toolkit to healthcare practitioners, patients and policy makers, while it has the potential to have a profound impact on drug manufacturing; manufacturers can be informed by the position of similar drugs in the best treatment pathways, so that they can target their newly to be developed drugs accordingly.