Interpreting health information and low value care using data science techniques
Identify the most robust statistical and data science methods for addressing challenges of comparing clinicians and interpreting healthcare information with respect of quality, cost and outcomes.
Risk adjustment is a common issue among medical providers who are cautious in supporting any initiative regarding transparency and information sharing. Medical providers indicate that factors influencing the complexity of a procedure, for example patient complexity factors such as age and obesity, must be considered when analysing such variation in order to form an accurate basis for comparison. Patient factors are a significant driver of cost and resource utilisation.
This project reviewed existing public models of risk adjustment and recommend techniques, approaches and communication frameworks suitable for public and semi-public (i.e. via intermediaries) transparency and information sharing purposes.