Data Driven Decision Making for Hospitals Part 3 – Critical Thinking

In Part 2 of this series, we covered a framework for the data driven decision making process used for defining the desired state after making a decision. In this post, we’ll cover the need for critical thinking while progressing through the decision making process. Critical thinking can be applied to any problem, so think of the problems in your hospital as you read this to envision how to apply critical thinking.

The Greek philosopher Socrates is credited with developing the dialectic method of opposing views. Challenging the perceived “truth” by posing alternative views and then supporting those positions with data and establishing a new “truth” can be applied to any decision process.

The critical thinking process can be broken down into a series of steps:

  1. Recognize the problem and break it down into subsets.
  2. Prioritize importance of solving subset problem statements.
  3. Collect currently available data.
  4. Generate a solution statement and alternatives, recognize bias.
  5. Expand data collection based on solution statements and alternatives.
  6. Examine relationships.
  7. Draw conclusions and state new “truth”

Fundamental to the critical thinking process is the ability to generate a clear problem statement. This means that the question must accurately describe the problem in terms that can readily be quantified or qualified. The problem statement must be specific. For instance, the statement “Improve Operating Room Efficiency” is too nondescript for a data driven approach.  Rather, “stratifying patients into complex and routine groups and scheduling based on past operating room time consumption” provides more context and leads to the collection of the appropriate data. Another sub question would be “determine the factors that classify a patient as routine or complex”.

Often the problem statement becomes too complex to address as a single question. In this case the problem must be broken down into objective subcomponents that address the stakeholders in the external and internal environment as demonstrated by the above mentioned problem statement.

Each of these statements will generate an array of decisions that must be addressed using critical thinking. As questions are answered, the results form the basis leading to how the decision is made. These “meta decisions” require specific data collection that will provide sufficient information on which to base the overall decision.  

The ability to expand into multiple domains as the problem is broken down into smaller components requires input from subject matter experts in associated domains.  Often these are experts familiar with the reach and development around a specific problem statement (e.g. who determines “routine vs complex”), however, the team may require other expertise such as legal and regulatory, sales and marketing, or staffing.  All of those impacted by and who are required to execute the outcome of the decision play a role in providing the necessary data and more importantly context for the decision.  The process is enabled by the use of machine data collection techniques such as Data Fusion and Natural Language Processing that can aggregate the necessary data required for the decision. It is not a requirement that everyone agrees with the decision since organizations generally have a responsibility matrix for these activities, but they must support the decision moving forward.

As with any decision there are alternatives. As part of the critical thinking process it is important to evaluate alternatives. Boundaries are sometimes placed on the generation of alternatives. From the purist point of view regarding critical thinking, this is generally frowned upon. However in the business world there are often practical considerations that come to play when posing alternatives. Practical considerations may be based on core competencies, financial capabilities, or resource restrictions that cannot be easily overcome. For example, the desire to offer a new type of treatment may be constrained by a lack of appropriate staff and equipment. The market analysis may not support the investment. During the decision-making process these barriers should be made clear and the reasons why they cannot be overcome.

It is important in the decision-making process that teams or ultimately the responsible individual obtain input from a diverse set of stakeholders prior to making their decision. The strength of the team is in its diversity of thought about the problems. This helps avoid decision bias. This is often referred to in the medical setting as “hearing hoofbeats and assuming horses rather than zebras”. The decision-maker, to be effective, must recognize that he most likely has a preconceived belief of what the decision should be based on intuition. Recognizing that bias exists allows the decision-maker to compensate by obtaining input from others.  The decision maker should be self-conscious about his decision.

The cognitive process of the decision maker is influenced by perception.  This will influence what the decision maker perceives as “truth”.  An example of such bias can be found in the following illustration.

ile:Two silhouette profile or a white vase.jpg

The decision maker may see the vase while others see two faces. Regardless, the data presented is the same.  How the decision maker processes the data differs and leads to different interpretations of “truth”.  Managing and leveraging these differences using a data driven process leads to more effective decisions.

Additionally, machines cannot deal with ambiguity.  Regardless of the claims associated with Artificial Intelligence, all machine learning systems work within boundaries. The more ambiguous the problem statement, the more unfocused the data collection and meaningless the results.  It is the responsibility of Management to drive ambiguity out of the decision making process using critical thinking.  Partnering with the subject matter expert drives the collection of facts that enable the use of data.

Next, we’ll delve into the taxonomy of problem types and associated approaches to find their solutions.

NoviLens is an appliance for data driven decision making. NoviLens handles the tedium of the computing components of the process and frees up the decision maker to focus on their problem solving tasks.

Visit to learn more about applying NoviLens in a hospital setting, and join our email list to get updates on new features and case studies. If you’d like to learn how NoviLens can help your hospital, contact us for a 30 minute demo at