MTEK Sciences develops cutting-edge statistical software to increase efficiency in a variety of settings.
Highly Efficient Clinical Trial Simulator
The highly efficient clinical trial (HECT) simulator is an online application that allows users to perform simulations of Bayesian platform adaptive trials to evaluate the pros and cons of candidate designs. The key design features in the application are early stopping for superiority or futility, adaptation of allocation ratios according to the accumulation of information throughout the trial, as well as the flexibility of dropping or adding treatment arms. The application incorporates a wide range of other features to optimize efficiency and ethics of the clinical trial including a number of time and cost related inputs and outputs.
Synthetic Control Arm Application
This web application addresses the need for control information in single arm trials where resources will not allow for a randomized control. This has applications in trials involving rare diseases, as well as those in which early evidence of treatment efficacy is too strong to permit a control group. The tool takes advantage of well-established causal inference methods to borrow information from historical control data, while verifying a healthy balance of all potential prognostic factors across the treatment arms. Numeric, dichotomous, count and time-to-event outcomes can be handled.
Dynamic Treatment Decision Model
Dynamic Treatment Decision Models (DTDMs) are a new way to inform legislators and planners in developing countries about how and where to allocate finite resources to maximize health gain. Traditional resource allocation information tools such as cost -effectiveness league tables have two significant limitations; firstly they do not take adequate account of the context in which they are used and secondly, they are static in nature and therefore do not account for the impact of other interventions in parallel.
We are developing a series of models – starting with child health – that allow for significantly more specific and reliable context relevant data on cost-effectiveness (both relative and absolute) and the ability to assess these estimates in combinations, rather than the rather unrealistic approach of seeing each intervention as a one-off investment.
DTDMs use regression-based algorithms to determine the scale and strength of the drivers of effectiveness of intervention in context and then apply them to contexts (at specific points in time) to ensure the costs and effectiveness are most specific to any one context at any one time. In addition, as they account for the simultaneous uptake of other interventions, they reflect the relative value of a new intervention under assumptions on level of take up of other interventions simultaneously, allowing for real time evaluation of combinations or ‘baskets’ of interventions.
For more information regarding any of our software, please