Playing with models in loosely coupled decision networks
Posted Oct 22 2008 6:27pm
In this post, I present two concepts which, when combined, have the potential to transform our healthcare system in profoundly positive ways.
The first is “ loosely-coupled decision networks ” in which people from multiple locations and with different roles, responsibilities and experiences work together to make decisions beyond the knowledge or skills of any individual. Collaboration among people with wide diversities of knowledge, ideas and points of view provides a larger collection of intellectual resource, and offers access to a greater variety of non-redundant information and knowledge on which to base decisions. Compare this to a tightly-coupled network that limits participation to people within the same discipline, department, region, etc. and with people who have access to the same information sources and who share similar experiences. In the loosely-coupled social networks are the greatest opportunities for stimulating multifaceted discussions, out-of-the box thinking, and creative solutions.
The second concept is “ sharing & playing with models. ” There are many different types of models used in healthcare, including models for defining health problems/diagnoses (e.g., ICD and DSM codes) and treatments (e.g., CPT and ABC codes ), for assessing and managing clinical and financial issues/risk (e.g., retrospective encounter and claims data analyses), for evaluating performance (e.g., variance analysis and risk-adjustment), for deciding the interventions to render and procedures to follow (e.g., clinical guidelines and pathways), for testing hypotheses and assumptions, for paying for care (e.g., HSA/HDHP and traditional indemnity insurance), for rationing care (e.g., QALY ), and so on. When people share and play with models, they compare models and test them for their ability to reflect reality accurately; they manipulate the models to represent different scenarios, such as “what if” scenarios about the probability of future occurrences; and they discuss the assumptions and results the models produce. When they find models that disagree or generate invalid results, they examine the fundamental assumptions built into the models, looking for logical flaws and inconsistencies, questioning the authors' perception of reality, and debating about the assumptions and practical value of the model. By challenging their assumptions, useful counterintuitive insights often emerge, innovative thought is sparked, new questions arise, relationships are developed, the influence of an organization’s culture and politics are revealed, and compelling and unexpected management issues are discovered. This means that sharing and playing with models is an effective path to innovation, risk management, and value creation.
Conclusion: By encouraging people in loosely coupled decision networks to share and play with models, radical innovation is fostered by disrupting of status quo, which enables the models upon which decisions are made to evolve continuously. The bottom line is that connecting diverse groups of people and giving them the ability to model-play would produce continually improving models; and using these model to support decision would result in safer, higher quality, more cost-effective care.