Construct Traditional Industrial QI Complex Adaptive System QI
Differences Differences Differences
Inspiration for Change Top-down from administration often using analyses of large datasets. Bottom-up from self-organizing self-learning independent, but interconnected, agents.
Selecting Measures of Success System administration or payers select approved single-disease (often) outcome targets vetted by influential organizations (National Quality Forum, e.g.) Single-disease measures are often recognized as inappropriate for complex systems trying to provide patient-centered care. Health services or health system measures, such as better use of clinic vs. urgent/emergency facilities, may be more meaningful.
Setting Target Goals Administration selects numerical goals often based on best practices or results reported by “centers of excellence.” Six sigma level outcomes often expected. There is no absolute numerical target to define success, only a sense of making the current situation better. Front-line clinicians realize their efforts could be influenced by numerous forces not in their immediate control, making a single numerical target meaningless.
Selecting Specific Processes to Improve Front-line personal are expected to rigidly and consistently apply best practices that may change slightly in subsequent PDSA* cycles. Agents that evolve in their understanding of an issue could change every aspect of the project, including who else is asked to help, the processes that are changed, and even the ultimate goals of the project.
Naming QI Priorities System administration or payers declare a list of possible improvements and prioritizes QI projects based on practical, financial, and political factors. Priorities are set by front-line caregivers and their immediate working partners also based on numerous factors, but largely driven by their own sense of what is most important at that time.
Organizational Buy-In and Reporting Regular organizational meetings with all members expected to participate and receive regular reports of the most recent analyses. The top levels of the organization may or may not be asked to buy-in and participate in QI activities for a particular project. Involved agents self-learn, self-organize, and evolve in their understanding of the contributing agents and forces influencing their goals.
Recognition and Rewards Organizations and their subunits are extrinsically recognized and rewarded to help motivate front-line workers to improve processes. Successful projects may be rewarded post hoc, but the motivation to improve systems is more intrinsic.
Similarities Similarities Similarities
Data Data are needed at several times over the life of the project, which will often require support from IT or financial personnel (billing or encounter data). Traditional QI utilizes reports updated on a regularly scheduled basis, complex QI perhaps less so. Data are needed at several times over the life of the project, which will often require support from IT or financial personnel (billing or encounter data). Traditional QI utilizes reports updated on a regularly scheduled basis, complex QI perhaps less so.
Resources Except for the relatively rare and simple “low-hanging fruit” successes, resources are required to enable people improving care to have the time and space to work on the specific QI project separate from direct patient care, often for both the immediate project and its long-term stability. Except for the relatively rare and simple “low-hanging fruit” successes, resources are required to enable people improving care to have the time and space to work on the specific QI project separate from direct patient care, often for both the immediate project and its long-term stability.
Multi-disciplinary Solutions most always include more than one type of clinician on the healthcare team and often include non-clinicians. Solutions most always include more than one type of clinician on the healthcare team and often include non-clinicians.
Processes not People The classic Deming philosophy that poor quality is almost always the result of poor processes, not bad employees, is true for both traditional and complex QI work. The classic Deming philosophy that poor quality is almost always the result of poor processes, not bad employees, is true for both traditional and complex QI work.