Statistical Development and Implementation

Our research focuses on the development and implementation of advanced statistical methods to make informed decision, improve health outcomes and strengthen evidence-based practices.

We are particularly interested in advancing statistical methods to inform decision/makers and regulatory assessors where evidence cannot be derived from randomized studies due to ethical reasons or when decisions are mainly based on single arm studies. We are working on the methodological development of G-methods in a flexible Bayesian framework to address issues related to spatial correlation in the data for a better estimation of causal contrasts of interests.

We also provide support in the analyses of individual and cluster randomized trials, including trials with innovative design such as Trials within Cohorts (TwiCs), platform trials, and stepped wedge designs.

Key project

Efficient use of available data in clinical research

Innovative Statistical Approaches in Causal Inference

Randomized clinical trials provide the highest level of evidence of the effect of an intervention. By eliminating confounding bias and ensuring a balanced distribution of potential confounding factors, randomization ensures a fair comparison between groups of individuals. However, randomized clinical trials are increasingly expensive, time-consuming, and can sometimes pose ethical problems. Cohort data collected through routine visits are a ready-made source of information. Under certain assumptions, advanced statistical methods known as 'G-methods' are able to correct for selection bias present in observational data and produce results similar to randomized trials. We aim to contribute to the development, validation, and implementation of ‘G methods’. more

Team

Projects we support

Viral Load Triggered ART care in Lesotho

VITAL

A large cluster randomized trial Lesotho evaluating an automated differentiated service delivery model that uses viral load results, sub-population needs, and participant and provider preferences to triage participants into groups requiring different levels of care. more

Community Based Chronic Care Lesotho

ComBaCaL

Developing, testing, and sharing innovative approaches to improve and expand services to prevent and care for non-communicable diseases in rural Lesotho. Lay providers are supported through eHealth applications to manage and diagnose NCD's. more

Innovative Trial Designs

TwiCs and Platforms

Innovative trials designs have the potential to overcome frequent problems of traditional Randomized Controlled Trials (RCT). Two promising ones are TwiCs and platform trials. more

COrona VaccinE tRiAL pLatform

COVERALL

This multicenter RCT embedded in the Swiss HIV Cohort Study & Swiss Transplant Cohort Study provides crucial information about the efficacy and safety of the mRNA SARS-CoV-2 vaccines. more

Previous projects

Timing of Antiretroviral Initiation and Cancer in HIV

Immediate initiation of antiretroviral therapy (ART) regardless of CD4 cell count reduces risk for AIDS and non–AIDS-related events in asymptomatic, HIV-positive persons. more

CD4/CD8 Ratio and Cancer Risk in HIV

Measures of longitudinal immune dysfunction and risk of AIDS and non-AIDS defining malignancies in HIV-infected antiretroviral treated positive individuals. more

CD4/CD8 Ratio and Heart Disease in HIV

CD4/CD8 ratio a marker of immunosenescence and chronic inflammation for coronary heart disease and stroke events in the Swiss HIV Cohort Study. more

To top