Apr 26 2019
Reading through The Checklist Manifesto
I love making lists: lists of things I like, lists of things I don’t, lists of things I need to do, lists of things I never want to agree to again. Part of this is because I am forgetful and easily distracted. It helps to offload the important things into a place I can reference, and the act of writing them reiterates that I want to remember these things.
Although I use a habit tracker in the Clear Habits journal, it’s really a checklist of all the things I want to have done in the day that help keep my mental and physical health on track. Seeing lapses in the checklist can help explain why a week has gone from bad to worse, and the checklist gives me an easy-to-follow structure for how to come out of the slump. I like this, and I want to find ways to think about using checklists at work and for other parts of my life.
Dr. Atul Gawande’s The Checklist Manifesto dives into the why’s and how’s of good checklists. I don’t know that I’ve ever seen a bad checklist, but most of the time they are for my own reference. What’s more difficult is to design a checklist for others, particularly when you aren’t there to answer any questions that come up.
Gawande, A. (2009). _The checklist manifesto: How to get things right. _New York: Metropolitan Books.
Instead of typing up notes by chapter, I ended up reading through the whole book and using sticky-notes for interesting points I came across. I didn’t think I’d have so many for what I thought was essentially a gift book, but here they are.
The entire book is peppered with anecdotes of checklist users, but naturally there is a strong series of medical stories. As a fan of Grey’s Anatomy and ER, I loved it. Within these anecdotes are also moments of theory and insight garnered from philosophers and other academics on the nature of human behavior. Gawande uses these to theoretically ground explanations of the checklist. For instance, the introduction uses the idea of “necessary fallibility,” accredited to Samuel Gorovitz and Alasdair MacIntyre. This concept states that “some things we want to do are simply beyond our capacity.” It continues to argue that for the things within our control, they may fail from ignorance or ineptitude.
I like using this binary for thinking about failure, whether it’s from not knowing something (which could possibly be known for the next attempt), or from not having executed it well (which can also be changed). It also helps me steer thoughts from what went wrong to thinking about how the things went wrong could be categorized and thus compensated for next time. This theory also allows Gawande to provide the purpose of the checklist: a tool that incorporates knowledge alongside the means to compensate for ineptitude.
Using the example of airplane test pilots, Gawande describes the checklist as an appropriate tool if the knowledge required is too complicated for one person to know, yet can be simplified/made brief enough to fit on an index card. It’s not all high-level stuff either; this should specifically include the dumb stuff that is easy to assume done. Gawande also describes two difficulties faced in a complex situation: the fallibility of human memory and attention, and the variable nature of context in that all of the steps might not matter. It can be thought of in the sense of what are the minimal steps/actions/factors needed for something to be viable. What is the lowest common denominator? Those should be on the checklist. Yes, there should be checklists even for the mundane stuff.
The other concern Gawande answers is what good is a checklist if you don’t use it? Using the example of the hospital and routine acts, he advocates for the use of checklists in everyday operations. Normalizing the checklist means it will be used to ensure the results it’s intended to achieve. He points out that this also has the affect of raising the baseline to a higher standard, bringing back the oft cited quotation (with James Clear and others) that we fall to the level of our standards. To create the checklist, you also have to be very clear about what actions you want it to reinforce. The result can often highlight infrastructural needs as well. Gawande also points out that a checklist is not always a collection of things—it can also include people to notify. While he has the much more dramatic example of a Search & Rescue team, it also includes the use of what I would call the “phone tree” wherein one person is assigned to contact one group of people, who are then assigned to contact another group of people and so on.
While checklists are amazing, Gawande also notes their limitations. They are used to reinforces processes in the face of a complex situation. However, not all complex situations are well-posed enough to be faced with a checklist. Gawande cites Brenda Zimmerman and Sholom Glouberman for distinguishing between types of problems: simple, complicated, and complex. Simple problems are highly repeatable, easily broken down into steps. Complicated problems may be composed of simple problems, but are not as easily repeated due to the number of factors involved. Complex problems have so many unique variables that it’s very difficult to predict the results. Keeping these types of problems in mind, Gawande compares the function of a checklist to the engineering concept of “forcing functions” which are “straightforward solutions that force the necessary behavior.” If a certain behavior during a complex situation can be forced, then it becomes easier to predict and arrive the desired result.
A large portion of the book also describes Gawande’s process of building a checklist to improve patient outcomes for surgeons across the world through the World Health Organization. He explains how they arrived at this plan, the research carried out before hand, and how they arrived at the elements for the checklist itself. One a visit to the construction site for a skyscraper, Gawande describes the number of teams, offices, and plans required for such an endeavor. One of the plans used has a distinction between physical tasks and communication tasks. Although we think of checklists as one thing, it’s interesting to consider how a checklist can contain subsets that work together. The other benefit of the checklist goes back to a previous chapter, in using it as a place to indicate knowledge. In the skyscraper example, the checklist also functions as a type of extended memory, (in my opinion similar to Andy Clarke and other phenomenologists study of phones and other digital tools). The information is stored there, and thus it is no longer necessary to devote memory to it.
In this chapter, Gawande turns to examples of decentralized vs. centralized organizations. The “command-and-control” concept leaves no room for improvisation. He posits it as a “contradictory mix of freedom and expectation—expectation to coordinate, for example, and also to measure progress toward common goals.” The checklist is a means to enhance judgment through the use of procedure.
Part of the process of using the checklist is in forcing behaviors, even mundane ones such as a conversation. Having added a checklist item for introducing doctors and nurses to each other at the beginning of a case, Gawande notes that this turns into an “activation phenomenon” as discovered at Johns Hopkins, in that by “giving people a chance to say something at the start seemed to activate their sense of participation and responsibility and their willingness to speak up.” Part of this can also be seen in work toward inclusion: there’s a difference between inviting someone to a party and asking them to dance. Passive vs active inclusion.
Looking at the kinds of behaviors a checklist can force, Gawande calls back to the aviation checklists to talk about “pause points.” In the operating room, these pause points allow users to pivot to the communication plan, and re-confirm the steps of the plan. Gawande also reiterates that building a checklist takes thought, it takes trial, and it takes revision. Put it in action, and if it isn’t forcing the right behaviors, then change it. In Ch. 8, Gawande uses another example to illustrate how you can create a checklist out of repeated mistakes to avoid them in the future.
Continuing into characteristics of a checklist, Gawande goes back to his aviator, Boorman, to discuss good and bad checklists. As Boorman explains:
Good Checklist
Bad Checklist
To take it a step further, Boorman also talks about two types of checklists: DO-CONFIRM and READ-DO. The first requires that each step be done as the team members know how, then they go back and use the checklist to confirm that everything was completed. The latter requires that you do the task as you check it off the list.
Part of the difficulty in Gawande’s project was in promoting the checklist, and using it to initiate a culture change. Part of it lies in “embracing a culture of teamwork and discipline.” However, he also cautions against over-zealous planning, which can then result in lists for everything that never get used and defeat the point of efficiency. I love that he later refers to the “virtues of regimentation.” It’s interesting that this is the term that leads into a discussion of how investors evaluate deals. Those who use a checklist—a rigorous, standardized process—often gain more than those who go purely by instinct or are lured into a seemingly good prospect. Even knowing that the use of this checklist will increase their gains, others will deliberately not choose to use it because it’s seen as more time-consuming and less thrilling. It also discounts “rockstar” status or those needing to be seen as special, critical, and irreplaceable as a checklist allows you to improve outcomes without requiring an increase in skill.