Dr. Sadmir Ciketic, Research Fellow in Health Economics at King’s College London has been using Palisade’s @RISK since his postgraduate days. His research in health economics and public health involves economic evaluation of healthcare programs and quantifying risks and uncertainties to evaluate the economic impact of medical interventions, such as the costs and benefits/harms of new medications, new or existing treatments and medical technologies.
According to Dr. Ciketic, Palisade’s DecisionTools Suite provides good estimates for the adverse outcomes and risks associated with new interventions or medical drugs introduced into the market. The Suite’s @RISK especially comes in handy for complex simulations and parameter uncertainties in modeling economic evaluations of medical interventions for making policy decisions. The software allows him to change the different assumptions of critical estimates in practice—for example, using a different distribution or a different method of sampling and simulation of results. Typically, quality of life estimates are derived from clinical trials and systematic reviews and as such these data already have a varying degree of confidence associated with them. This is especially of importance in solving the economic problem of limited resources in the public health arena. This is where the DecisionTools Suite and @RISK join in. Through Monte Carlo simulation, models can account for a whole range of possible outcomes, and indicate how likely they are to occur, providing crucial estimates that can be interpreted and used in policy and decision-making.
“In my current area of work for King’s College London, I mainly use quantitative analysis,” said Dr. Ciketic. “The two most common examples of the distributions that I use include the beta distribution for the quality of life estimates and gamma distribution for cost estimates. The use of these distributions is justified due to the fact that quality of life parameters in the model are binomial data—mainly probability parameters in the model ranging in values from 0 to 1, while the cost data are non-negative and in many cases skewed. Due to the fact that cost is represented in Bayesian statistics (discrete) by a Poisson distribution, the gamma distribution is a conjugate to the Poisson and as such is constrained on the interval 0 to positive infinity.”
Dr. Ciketic also attends presentations and events organized by Palisade on a regular basis. He said, “There is always something new to learn and something that is very useful in relation to @RISK, and all the other DecisionTools Suite tools. It’s a complete suite of programs that are suited for many professions from science, finance, government, health and many other areas. I like the fact that Palisade has this broad level of use that is relevant in all aspects of life. ”