In my ongoing effort to remember what I have read, some notes on Atul Gawande’s book, “The Checklist Manifesto: How to Get Things Right” [1].

1. A Book About Checklists (Sort Of)
2. The Solution to Defining Problems
3. Algorithm of a Problem
4. Complexity & Nightmares
5. How Do I Fail You: Let Me Check the Ways?
6. A (Good) Checklist to the Rescue
   1. What Makes a Good Checklist?
   2. Why do we hate checklists?
   3. The Definitive (Check) List of Why We Hate Checklists
7. References and Further Reading

A Book About Checklists (Sort Of)

As a surgeon, Dr. Gawande describes the benefits the lowly checklist has made in operating theatre. He has been compared to Malcom Gladwell in his writing-style which I think is apt. Beyond being a medical-Gladwell, he makes the case for using checklists to reduce errors, costs and improve outcomes.

He also discusses what makes for a good checklist or how to spot a bad one. Looking beyond these practicalities, this book is about management issues such as decision making. For example, how to make a quick decision when the plane you are flying is falling from the sky and you have 180 seconds to decide where to land it.

The Solution to Defining Problems

Checklists come in all shapes and sizes. When I travel, I use a generic packing list in which new items are constantly added. For a specific trip, I suppress all but the few that I need to throw in a duffle bag. At the other extreme is a short set of instructions for a pilot of what do when the engines have stopped because they are stuffed-full of Canada Geese [2].

In both cases, the checklist is supposed to reduce/eliminate a problem: forgetting a tooth brush or restarting the engines. But problems are not one size fits all. Gawande refers to the work of Zimmerman and Gloubermann who defined problems [adapted, 3]; The company Clever Checklist added the 4th definition and the applicability to checklists [8].

Simple problems encompass some basic issues of technique and terminology, but once mastered there is a very high assurance of success. CHECKLISTS can assist with procedural steps.

Complicated problems contain subsets of simple problems but are not
merely reducible to them. Their complicated nature is often related not only to the scale of a problem but also to issues of coordination or specialized expertise. Complicated problems, though generalizable, are not simply an assembly of simple components. CHECKLISTS can assist with documenting unexpected outcomes.

Complex problems can encompass both complicated and simple subsidiary problems, but are not reducible to either (Goodwin 1994) since they too have special requirements, including an understanding of unique local conditions (Stacey 1992), CHECKLISTS have limited applicability without breaking this type of problem down into smaller components.

Chaotic Problems are neither knowable nor predictable, such as the weather. Many pieces act together in a system but the pattern and relationships may be beyond our ability to comprehend. CHECKLISTS have no applicability due to the inability to repeat results.

Glouberman and Zimmerman, 2002

Algorithm of a Problem

Complex situations (such as an operation or a day in an ICU) involve hundreds of different types of simple and complicated problems. The math behind whether a situation is complex or not might be something like this:

• [Technique + Terminology + Methods + Resources + Environment] X # Simple Problems X
• [Scale + Coordination + Specialization] X # Number of Complex Problems X
• [Local Conditions + Need for Experience + Intuition]
• = Complex Situation

Gawande noted that in the operating room, each of these elements is growing in its own fashion. New surgical Techniques, new Resources such as diagnostic tools, more complex procedures (Scale), which require more Specialization, all of which still require the correct application of Intuition and Experience by skilled and highly trained surgeons.

Complexity & Nightmares

Simple, Complicated and Complex – that was the case history behind a parent’s worst nightmare. A 3-year-old girl who fell into a pond while out walking with her parents in an Austrian Forest. Gawande describes how a small, regional hospital did the impossible – save this ‘drowned’ child through a coordinated series of interventions.

Watch from 2:15 to 10:34 of a video presented by Microsoft Research in which Dr. Gawande recounts the above case history [4]. In the final minutes of the case, he explains how the hospital was able to do the impossible.

How Do I Fail You: Let Me Check the Ways?

Gawande notes that failure comes in two favours: Lack of knowledge or Ineptitude.

Knowledge Failure: Because we lack the conceptual framework, technical expertise, or language (e.g. calculus) to solve a problem. Gawande argues that this was the state of affairs throughout human history until the last ~100 years.

A Simple Problem is unsolvable because of a lack of technique, terminology, resources, knowledge, etc. Because a society lacked an understanding of germ theory, it failed to install sewers which led to cholera outbreaks.

Ineptitude Failure: Despite having the knowledge (or at least a portion of it), we fail because we are inept in applying the resources available. A skillful doctor kills a patient for want of washing her hands.

Ineptitude can act on its own (e.g. an infection) or small ineptitudes can compound leading to failure. A checklist is designed to act as a fire break between both types. The simple question of ‘have you washed your hands’ is a matter of life or death. [5]

A (Good) Checklist to the Rescue

So, is that it? We adopt a checklist and we can then fly planes, are no longer inept or can save drowned children? Not quite. Before thinking about a check list, we must have mastered the skills and resources required to solve Simple, Complicated and Complex problems. If so, then what makes for a good checklist?

What Makes a Good Checklist?

Gawande visits Boeing in Seattle to find out how to write a good checklist, one useful when your airplane engines are full of Canada Geese. The following points are critical:

1. Research and Tested: All of the following attributes are useless unless the checklist is tested in the field (or at least a flight simulator).
2. Adapted and Updated: Circumstances will change necessitating an update to a checklist. In addition, every checklist should be re-tested periodically (yearly, bi-annual, etc.).
3. Owned: A checklist must be owned by one individual who ensures it is tested and updated.
4. Context-Relevant: Building a skyscraper affords a longer luxury of time and thus the length of a checklist. Alternatively, having your engines filled with purée of goose and the ground 180 seconds away requires brevity.
5. Organized into Pause Points: The point a user must stop to run through a set of checks before proceeding. In the surgical checklist Gawande developed with the World Health Organization, there were three: pre-anesthetic, pre-incision and end of procedure.
6. Keep tight and focused on ‘Killer-Items’: A checklist should only include the things most likely to kill you, your airline passengers, or your patient. These are the things most often forgotten or missed.
7. Simple Language, Uncluttered and on One Page: Ideally a checklist is printed on one side of a piece of paper. Ambiguity is eliminated and text is written in both upper and lower case font.
8. Less (Time) Is More: Returning to Context-Relevant, the ideal checklist associated with a Pause-Point can be completed in 60-90 seconds. Every additional second returns diminishing returns. A team evaluating investments can perhaps afford a patience testing 300 second checklist but a flight crew listening to screaming passengers will appreciate one that takes fewer than 60 seconds.

Why do we hate checklists?

Gawande confesses that he hates using checklists but can enumerate countless times when he was glad that he has adopted their use. Given that the evidence is striking, exactly why do checklists seem so unappealing?

This is perhaps the weakest area of Gawande’s book. One explanation he provides is that a checklist runs contrary to one’s sense of craftsmanship and joy in the work. While plausible, this explanation is not compelling. His other explanations are sprinkled throughout the book rather than in a nice list, such as this one:

The Definitive (Check) List of Why We Hate Checklists

People hate checklists because of individual self-perception, cognitive biases, organizational culture and risk/reward of doing so.

1. Self Perception: as discussed above, a checklist seems to diminish our sense of mastery over a task or the heroism it entails when we are successful.
2. Hubris Biases: Personal ‘blind spots’ about ourselves [7].
   • Over-Confidence or Self-Assessment Bias: the reason 66% of all drivers are above average. To be good at something you need confidence but the more confidence you have the more likely you will overlook things that will make you bad at doing something.
     • Confirmation Bias: using information in a way that confirms or supports one’s prior beliefs or values. If you are a confident surgeon, you may miss the obvious signs that it is the RIGHT knee that needs to be replaced.
   • The Optimism Bias: underestimating the probability of undesirable outcomes and overestimating favorable and pleasing outcomes. Canada Geese flying into my engines is an impossibility.
3. Organizational Culture: some cultures manage failure better than others.
   • The military, from whom flight checklists originated, does a good job (mostly) of integrating new information and failure in its planning where as surgeons traditionally buried their mistakes.
   • The regulatory framework of the aviation industry supports standardization. Aviation is new and until the last few decades subject to enormous technological changes.
   • Surgery has had a slower evolutionary progress from the days of the barber-surgeons.
4. Risk-Cost Equation: a final factor is whether or not those using it are likely to die or be responsible for a catastrophic disaster.
   • Pilots, sky and scuba divers use checklists because a single point of failure can result in death or disability.
   • Gawande discusses how builders use checklists because a sky scrapper collapsing is hard to miss.
   • Move away from scenarios in which thousands of people (or just you) die and we can see those cognitive biases, and a poor culture may make a checklist not worth the bother.

Gawande frames why we hate checklists in 11 more minutes of the video made by Gawande for Microsoft [7, 31:18 to 42:07). In this clip, he describes how a checklist augments rather than reduces our abilities. He also closes the story of how a rural hospital in Austria prevented a parent’s worst nightmare – partly because of a checklist.

Have you read this book and what are your thoughts? Having spent most of my career creating, imposing and thinking about documentation [9]; it was a great perspective but left me wanting more.

References and Further Reading

1. The Checklist Manifesto: How to Get Things Right: Gawande, Atul: 9780312430009: Books – Amazon.ca.
2. Which is what happened to Sully Sullenberger on January 15, 2009.
3. Glouberman, S, Brenda Zimmerman, and Commission on the Future of Health Care in Canada. Complicated and complex systems: what would successful reform of Medicare look like? Saskatoon, Sask.: Commission on the Future of Health Care in Canada, 2002.
4. The Checklist Manifesto – YouTube.
5. Dr. Ignaz Philipp Semmelweis was vilified for suggesting that doctors wash their hands between patients. This is a tragic story in which he was unable to convince his peers of the benefits of the practice and he died a broken man Ignaz Semmelweis – Wikipedia.
   • A happier story is that of Dr. Robert Storrs who published his findings on the benefits of hand washing seven years before Semmelweis – although not in a major medical journal.
   • Unlike Semmelweis, Dr. Storrs had the support of his peers in this local practice and thus he concepts of hygiene were more widely adopted. [Adapted from Grogan, Suzie. Death. Disease & Dissection; The Life of a Surgeon-Apothecary, 1750–1850. Pen and Sword, 2017. Chapter 7].
6. Tuckman’s stages of group development – Wikipedia.
7. There are a gazillion references on cognitive biases on the internet. I like this list from Warren Buffet’s business partner: Medium – The 25 Cognitive Biases by Charles Munger.
8. “1. The Problem – Clever Checklist Docs.” Accessed January 16, 2022. https://cleverchecklist.com/docs/making_checklists_that_work/problem/.
9. For a selection, see discussions on Knowledge-Management in various blogs.
10. Other Youtube videos worth checking out:
    1. The Importance & Value of the CHECK LIST – YouTube.
    2. Making A Checklist, And Checking It – YouTube