Research

• My main research interests lie within the area of: Bayesian Methods in Health Economic Evaluation. 

In particular, I use Bayesian methods to model utility measures using mixtures of distributions. Although utility measures, such as QUALYs, are usually modelled using a single distribution, however, often observed data show multimodality. Therefore, it may be necessary to use mixture models to properly account for this.

FOR THE SPECIALIST:

The EQ-5D-3L is a self-reported questionnaire which is frequently used for the measurement of someone’s health and to identify their health state. The derived utility scores of the UK EQ-5D-3L were computed using frequentist techniques on a Time-Trade-Off (TTO) assessment and they represent point estimates of the quality of each health state. These point estimates tacitly ignore the uncertainty of the estimates due to the variability inherent in the underlying data. In order to address this, I propagate the uncertainty of the UK EQ-5D-3L scores by constructing a Bayesian hierarchical model which accounts for the responses of the survey participants and assigns a probability distribution to the utility score of every feasible EQ-5D-3L state experienced by a group of people such as clinical trial subjects. This provides the useful advantage of doing sensitivity analysis without making any further distributional assumptions about the utility scores of the EQ-5D-3L health states experienced by the clinical trial subjects. Furthermore, due to the frequent multi-modality in the data, mixtures of distributions are used to properly account for this issue. Similar methods can be applied to the EQ-5D-3L scores of other countries and to other health instruments such as the SF-6D. 

FOR THE NON-SPECIALIST:

There is an ongoing interest for the improvements of economic evaluation methods. Although a typical research field of economists, nowadays it has become an active and growing research area of statisticians who have been attracted by the potential for interesting modelling, especially in the context of Bayesian statistics. In a world of potentially infinite medical advances but with only finite resources, decisions are made by the National Institute for Health and Care Excellence (NICE) regarding which medical interventions in the UK worth the value for money. The EQ-5D-3L is a two-pages self-report questionnaire which is frequently used for the measurement of someone’s health and the identification of their health state; it is a generic instrument so it is easy to be used across different disease areas and it is the preferred tool of NICE for the measurement of the health-related quality of life. These tools are used to conduct cost-effectiveness analysis and determine if a medical intervention is worth the value for money. The original tariff of the utility scores of the EQ-5D-3L states uses point estimates for the description of the utility of each health state. I use statistical techniques to provide probability distributions in order to update the tariff of the EQ-5D-3L utility scores so that I account for the variability inherent in the underlying data.

My work has been presented in international conferences and published in scientific journals.

• Effectiveness of interventions to facilitate prompt referral to memory clinics in the UK    

Several UK residents who have dementia do not receive timely diagnosis. I assess if the use of GP's personal letter (with the inclusion of an evidence-based leaflet) can help to increase patient presentation to general practice and increases timely dementia diagnosis.

Check the article published in the British newspaper "Independent" about this part of my research work.

• Parametric model comparison for survival data, in the context of Health Economics Evaluation

The main tools of interest are: Akaike Information Criterion (AIC), Deviance Information Criterion (DIC), Schwarz's Bayesian Information Criterion (BIC), Bayes factor and variants.

Their efficiency is examined by conducting simulation experiments, and then they are applied to clinical data to investigate different model extrapolations.

• Gazyva & Bendamustine maintenance budget impact model

Construction of a budget impact model in an Italian context for medical interventions focused on patients with indolent non-Hodgkin's lymphoma (iNHL). 

An innovative treatment regime.