Complicated versus complex clinical problems
One of the biggest challenges today in health care is that most clinical organizations are being run with the wrong organizational assumptions about the core problem.
Health care managers wish that physicians would just act differently. As a partner at McKinsey and Company, I saw in multiple payor and provider senior management meetings the sentiment that “if only the physicians would do [fill in the blank], we would be able to solve problem [ fill in the blank]. Management is searching for the right playbook and rules set that if the physician follows all else will be well.
Physicians believe that most management systems don’t understand their workflow or the specific challenges in patient care. You can see this most readily, in the ongoing and warranted uproar about meaningful use standards in electronic health records and how it has created more work for little values. The Health Care Blog has some provocative viewpoints here and here.
I think a core issue is that management does not appreciate that high quality care for patients with multiple problems is inherently a complex problem rather than a complicated problem.
In their practical overview on how to drive improvement in health care , Value by Design: Developing Clinical Microsystems to Achieve Organizational Excellence, a team of Dartmouth quality improvement leaders articulate the difference between complicated and complex problems:
Complicated problems are classic engineering problems such as building a rocket or open heart surgery. They lend themselves to detailed engineering solutions and high specialization. However, once you master the solution, it is readily scalable. The key aim of solutions for complicated problems is reliability.
Complex problems are those that have multiple systems and feedback loops between humans. When you try to solve the problem, the problem can fight back and change. Think of the work done in raising a child – there is no one playbook. The same goes for changing lifestyle behaviors or in creating a custom solution for a patient not responding well to traditional medicine. The key aim of solutions for complex problems is resilience.
Problem Type Simple Complicated Complex
Prototype Yes/ No functionality If / Then algorithms Leveraging relationships
Elements Known Knowable Partly known, can change
Outcome Predictable Largely predictable Inherently unpredictable on an individual basis, improved predictability on a population level
Sample clinical solutions Automating tasks, routinizing safety, checklist, forcing functions, monitoring Care algorithms, standardized education and treatment Building relationships, shared decision making, N of 1 experiments and customization, management of frailty and multiple conditions
Clinician autonomy Low Variable High
Aim Reliability Reliability Resiliency
Non clinical examples Following a recipe Building a rocket Raising a child
Expertise None required Specialized fields Helps, not sufficient
Integration Parts and quantities clearly defined Parts separated then coordinated Parts cannot be separated from the whole
The first step towards addressing the epidemic of chronic disease in the US and the world is recognizing that is a complex problem at multiple levels – the individual level, the patient-clinician level, the community level and the policy / business level. Solving this will require resilient organizations and resilient solutions.