Helping Leaders Blink Correctly Split-second decisions have patient - PDF document

Patient Safety Helping Leaders Blink Correctly Split-second decisions have patient safety implications. than esthetic judgment. Now I real- problematic. It is not uncommon, for In 1983, the J. Paul Getty Museum in California was approached by an ize I was wrong.” example, for individuals to blink and quickly engage in thin slicing when art dealer claiming he had a very rare statue called a kouros (an ancient This story appears in Malcolm presented with performance improve- Gladwell’s 2005 award-winning book, ment or financial data. We see trends Greek statue of a standing nude youth often thought to represent the Blink: The Power of Thinking Without where no trends exist, conclude that Thinking (Little, Brown), which the data have shifted when in fact they idea of youth) that dated back to the sixth century B.C. As only about 200 details fascinating stories of how indi- display nothing more than random viduals make split-second decisions by variation or spend an inordinate kouri exist and most are damaged, the Getty was interested in adding engaging in what Gladwell calls “thin amount of time trying to explain a slicing.” Thin slicing “refers to the single high or low data point while this rare and supposedly fully intact statue to its collection. Before writing ability of our unconscious to find pat- ignoring the rest of the data. terns in situations and behavior based a check for more than $10 million, however, Getty’s curator wanted to on very narrow slices of experience,” In order to blink correctly, like Getty according to Gladwell. board member Zeri did, healthcare be sure the kouros was authentic. leaders need to develop skills in four Sometimes these thin slices lead indi- key areas: The museum’s research staff con- ducted a 14-month study and deter- viduals to make accurate assessments, as in the Getty board member’s one • Understanding the messiness of mined the statue was the real thing. But just before the acquisition was look at the kouros. But at other times improving healthcare completed, Getty board member thin slicing leads people to make • Determining why they are Frederico Zeri took one look at the incorrect decisions, some of which measuring statue and said it “didn’t look right.” can lead to tragic consequences. For • Understanding and depicting What was the problem? “It was example, Gladwell tells the story of variation fresh,” he said. The statue turned out how four New York City police offi- • Translating data into to be a fake. cers thin sliced an unfolding situation information and killed a young man from Guinea How, after researchers spent 14 as he pulled out his wallet to show The first two skills are discussed in months studying the kouros and the officers his identification card. this article, and the second two skills gathering a considerable amount of They thought Amadou Diallo was will be addressed in a future issue. data, did they arrive at an inaccurate pulling out a gun. conclusion? How did one man, with Understanding the Messiness of a quick look at the same statue, know We blink and thin slice all the time. Improving Healthcare it was a fake? The experience led the In healthcare, especially, we engage in The complexity of healthcare chal- curator to conclude, “I always consid- thin slicing when it comes to analyz- lenges cannot be adequately under- ered scientific opinion more objective ing data, and that approach is usually stood with simple models or theories. 88 Reprinted from Healthcare Executive MAY/JUNE 2010 ache.org

Patient Safety Rarely does a single variable drive an are five independent variables (age, in patient outcomes. There may be gender, current health status, coordi- outcome. But it is surprising how variables we are not even considering often we blink as though this is the nation of care and communication), (e.g., the presence of a family support indicated by the Xs. Each independent case—that is, X leads to Y. An alter- system) that have more of an impact native to this perspective, proposed in variable by itself has a direct effect on on the outcome than do the variables the outcome. Notice, however, that the book Causal Models in the Social we have identified. These unac- Sciences , edited by H.M. Blalock Jr. this model becomes messy from the counted-for variables are identified by 10 possible interactions between the the residuals (the Rs) in the model. (Aldine, 1971), is causal modeling. This process offers a more accurate five independent variables (e.g., four of these interactions are X 1 X 2 , X 1 X 3 , framework for blinking (and think- Blinking at even simple problems can ing) about the complexity of the X 1 X 4 and X 1 X 5 ). lead us to think that the solutions should be quick and easy. (“Just fix problems we face. These interactions create a complex set it!”) Good leadership begins with blinking accurately and realistically Using causal modeling, the outcome of relationships as we attempt to measure or dependent variable (Y) in untangle, for example, the combined about the nature of the problems we seek to improve. the chart below could be a patient effect of age and gender on patient out- assessment score such as health status comes. The model becomes even mess- or an outcome from a hospital admis- ier when you realize it may not Determining Why You Are sion. Note that for outcome Y there adequately account for all the variation Measuring The act of measuring healthcare pro- cesses and outcomes provides an oppor- Direct and Indirect Effects in a Causal Model tunity to blink in many different ways. Yet, many people blink as though all In this example, there are numerous direct and indirect effects between the indepen- measurement is basically the same. dent variables (Xs) and the dependent variable (Y). For example, X 1 and X 4 both have direct effects on Y, plus there is an indirect effect due to the interaction of X 1 and X 4 L. Solberg, G. Mosser and S. conjointly on Y. McDonald, in “The Three Faces of Performance Measurement: R 1 Age Improvement, Accountability and X 1 R 4 Research,” published in the Journal Coordination on Quality Improvement in 1997, pro- Patient assessment of care vide a useful context for thinking score (could be health outcomes, functional about how we blink when it comes X 4 status or satisfaction) to measurement and define what they call the three faces of perfor- R Y Gender Y mance measurement: accountability, Source: Institute for Healthcare Improvement R 2 X 2 research and improvement. Healthcare organizations regularly engage in and X 5 use all three approaches to perfor- mance measurement. Communication The leadership challenge, therefore, is R= residuals or error X 3 R 5 terms representing the to be clear about the purpose of your effects of variables not Current measurement efforts and avoid being, included in the model health status R 3 as Solberg and colleagues state, “coun- terproductive by mixing measurement Time 1 Time 2 Time 3 for accountability or research with 90 Reprinted from Healthcare Executive MAY/JUNE 2010 ache.org