There is a straightforward way  to think through the problem of managing complex, dynamic systems for high performance and the avoidance of realized risk.

For examples, healthcare professionals regularly assess risk in complex, dynamic systems by determining where vulnerability exists and acting in anticipation of failure, to preclude it, rather than in reaction to failure, when correction is difficult, costly, and incomplete, if at all possible.

The specific example I’m writing about, of course, is how well trained and disciplined healthcare professionals treat patients.  Working against ‘normal’ for temperature, blood pressure, and a host of neurological and chemical measures, you:

See Problems When and Where They Occur
-1- start with a slate of what ‘normal’ measures
-2- measure for departures (the more vulnerable the ’system,’ the more intense the monitoring),
Solve Problems When They Are Seen
-3- engage your understanding of chemistry and biology to explain deviations
-4- develop treatments
-5- follow up to determine if the treatments are working as predicted/expected.

If not, you repeat as necessary.

Those exceptionally capable of managing complex systems for exceptional levels of performance do exactly they same thing.

See Problems When and Where They Occur
-1- The predict ‘normal’ in advance of doing work–in the form of prespecifications, :

• what outcomes will be generated by what ‘pathway’ (flow) of connected work-activities.

-2- The see ‘abnormality,’ in the form of departures from these predictions, through ‘built in test’ diagnostics that determine if:

• Output: what is being generated is different than the predicted results.
• Flow on pathways: work is preceding in a sequence or is being done by people other than indicated on the specified pathway.
• Handoffs over connections: exchanges of information, materials, and services match the content, format, and timing expected in the design.
• Activity-methods for doing individual tasks are adhering to the pre-specification or if something is occurring by content, location, timing, or result different than expected.

Solve Problems When They Are Seen
-3- When any of these tests indicates a departure, the examine the outputs, pathways, handoffs, and activities to see is mis designed or mis performed
-4- Develop ‘countermeasures’ (treatments) that are predicted/expected to restore the health of the operating system.
-5- Follow up to determine if the countermeasures are improving quality, cost, safety, timeliness, security, responsiveness.

This pattern of define ‘normal,’ detect ‘abnormal’ (at excruciatingly low thresholds) and containing and treating are fundamental to the success of Alcoa in achieving near perfect workplace safety (see The High Velocity Edge, chapter 4) and the US Navy’ nuclear program in achieving perfect safety (no injuries or environmental contamination) since the first nuclear powered warship was launched in 1954 (chapter 5).

The well known medical error drama, “First Do No Harm,” is replete with negative examples of these failures.  From the very first encounters in the obstetrician’s office–misdirected delivery people, inability to schedule appointments, through the crescendo climax of lost code teams and mis stored medications, the system fails because small departures from normal go unseen, unsolved, so remain in place to coalesce into terrible harm.

We launch an IHI Open School course on October 14th illustrating these concepts.

Yours,
Steve Spear

www.TheHighVelocityEdge.com
Twitter @StevenJSpear

Related posts:

1. Designing, Operating, and Improving Complex Systems: Common Challenges–>Common Responses
2. Incident Reporting Systems: Inadequate tool for quality and safety…
3. Bob Herbert: A Less Than Honest Policy–Sen. Healthcare ‘Reform’
4. Factory Efficiency Comes to Hospitals…NY Times July 11, 2010
5. David Brooks: “Where Wisdom Lives” way overstates Medicare problem…