Testing Interventions
In line with our mission to establish the Community Health Council, we seek to rigorously test interventions implemented at the direction of the Council. Our first intervention is the High-Need, High-Cost patients care coordination program run by PCIC.
Randomized control trials (RCT) have been accepted as the gold standard of testing in both medicine and public health. The RCT design is able to control both measurable and unmeasurable confounding and is best able to establish a causal link between an intervention and the observed outcome or lack thereof.
If feasible and ethical, RCTs have come to be expected as an initial basic standard for an intervention to be adopted into practice. At PCIC, we expect to test every intervention at implementation through randomization, if it is feasible and ethical to do so. The first RCT will seek to evaluate our High-Need, High-Cost patients care coordination intervention.
However, it is not always prudent, feasible, or ethical to use the randomized design. In addition, invaluable information can be learned from the success of an intervention that has already been implemented and is being evaluated in a context closer to real-world functionality of the intervention. In other words, the rigorous and highly controlled environment of a randomized design often over-estimates the potential of an intervention in impacting the outcome of interest. Designs that approximate randomization, will, hence, be leveraged.
Testing Intervention Techniques
-
Propensity score matching is one of several approaches that seeks to approximate a randomized design. We are currently using the propensity score approach to evaluate the success of our High-Need, High-Cost patients interventions.
-
We are currently in the design phase of our first randomized control trial that will evaluate the effect of PCIC's High-Need, High-Cost patients care coordination intervention. Our effort follows in the footsteps of the Camden Coalition evaluation of their intervention for complex patients.