Design and Discovery
Eight months ago, I started a new project. We want to improve the way a large organization delivers age-friendly care in an ambulatory setting. We outlined an initial design and tried it with six volunteer providers in May for four weeks. We couldn’t reach our initial goal to have clinicians achieve perfect age-friendly care in at least 50% of eligible visits. We encountered several challenges in communication and feed-back.
In July, we recruited a diverse group of 23 providers to try a revised design for eight weeks. Our feedback methods were still too burdensome for busy providers. While eight providers improved performance relative to baseline as measured by conformance to an ideal age-friendly visit, our revised design still failed to achieve our initial 50% goal. Our second test convinced us to focus on a subset of 100 providers already named by the organization as ‘age-friendly champions’—three of four champions in the group of 23 were among the eight providers with improved performance.
We’re now adjusting the summer’s design and plan to test again with 24 champions, starting in January. Our redesign work includes reviewing 2021 performance of ideal age-friendly visits to shape our predictions and selection of participants.
As I look back at our experience since April, I see how our tests have revised and deepened our initial understanding. We’re closer to a better design; the third test cycle should help us discover new ways to make it easy for all providers to deliver age-friendly care.
Spear’s Insight
I referred to Steve Spear’s 2009 book The High Velocity Edge in my previous post.
In Chapter 4 of his book, Spear describes the system redesign and performance at Alcoa under leadership of Paul O’Neill, CEO and then chairman, 1987-2000.
Just after starting as CEO in 1987, O’Neill famously set a goal of zero injuries in a speech to investors:
“Every year, numerous Alcoa workers are injured so badly that they miss a day of work… Our safety record is better than the general American workforce, especially considering that our employees work with metals that are 1,500 degrees and machines that can rip a man’s arm off. But it’s not good enough. I intend to make Alcoa the safest company in America. I intend to go for zero injuries.” (speech accessed here 22 November 2021.)
Over the next 13 years, O’Neill led Alcoa toward the safety goal through transformation of expectations and application of relentless problem-solving.
Spear summarizes a key lesson from his study of the Alcoa management system under O’Neill:
“No one team can design a perfect system in advance, planning for every contingency and nuance. However, as Alcoa realized, people can discover great systems and keep discovering how to make them better.”[emphasis in the original text]
“[There is an] inherent impossibility of anticipating the myriad interactions among the components that make up complex systems of work. Despite all the effort put into up-front design, something will always be overlooked.”
“If you had to depend on a single explanation for Alcoa’s success, it would be that Alcoa gave up depending on designing perfect processes and committed itself to discovering them instead.” (p. 92-93 The High-Velocity Edge)
Testing as Universal Skill and Expectation
Complex systems are characterized by interactions and relationships that are a priori difficult or impossible to anticipate.
Like my ambulatory care example at the start of this post, most of the systems I encounter in my healthcare work qualify as complex.
Spear’s insight implies we should abandon the illusion that planning or design will be a one-time occurrence in work with complex systems. Instead, we should commit to discovery of effective designs and plans through explicit testing using Plan-Do-Study-Act. In other words, we should budget and coach people to regularly test their way to discovering ways to provide better services with less strain and waste.
In this view, testing is not a skill and intervention to be reserved for special improvement projects. Rather, testing is a way of working in complex systems that we should expect everyone to master.
