Part 2: Questions Before Queries

Part 2: Questions Before Queries

Posted Feb 20, 2018

Questions Before Queries: 5 Approaches to Improving the Report Request Discussion
Part 2 of the Series: Healthier By the Numbers: Musings in Pursuit of Data Interpretation and Explanation

|by Kempton Presley, M.P.H., M.S.
VP, Business Information Solutions & Client Performance

Everyone wants answers, but few understand the questions. When evaluating report requests, information supply does not always meet demand. We do not take enough time to understand the fundamental problems driving information requests. As such, outputs that follow complex information requests often don’t quite meet the needs of the requestors. Scarcity of time, perspective, and expertise can yield suboptimal responses characterized by silence, premature conclusions, overpromises, or reactionary defensiveness. No matter the flavor of bad answer, the unintended consequences range from extra, non-essential work for internal reporting teams to potentially disastrous clinical outcomes at the population level.

Logically, an accurate diagnosis of the question is a critical precedent to the construction of a comprehensive answer. However, it is always very tempting to start “solutioning” before knowing the optimal information destination. For a developer, this might involve sitting down at the keyboard and coding prior to receiving comprehensive requirements or fully understanding the problem. Here are a few ways to improve the interactions that precede the report construction adventure:

  1. Push For Clarity When Possible

Obscurity of the fundamental problems embedded in an information request creates a barrier to optimal report creation. Whether an internal customer or an external client, sometimes the requesters do not know their exact goals. Careful not to overly mirror the style of Trebek or Socrates, the best response to a question may actually involve more questions. Getting the requester to provide as much detail as possible regarding the intended output brings discipline to the entire process. Asking for a recap of the request in writing can ensure the requester thinks through the problems he or she is trying to solve. Taking it a step further, if the requester has the depth of knowledge to outline the data fields or specific charts in demand, he can help solve a big part of the mystery before the development work begins.

  1. Resist Oversimplification

“Hitting the easy button” is always an easy response. As a result, the contents, timing and overall value of reports often fall out of line with constituent expectations.  Many times, without looking at the data itself, a false sense of cause and effect can direct us to answers that seem obvious; but unproven theories are merely hypotheses. For an account representative, this could mean agreeing to a proposed timeline for a set of new reports without knowing the data required to produce the answer.  For a hospital executive, this could mean questioning accuracy of a new dashboard because the contents don’t mirror income statements and discharge reports. In each case, further discovery and expectation management before the reporting commitment would increase the likelihood of recipient satisfaction.

  1. Avoid Making False Parallels

A set of answers that worked in one situation may not work for another.  A compounding factor that improves the likelihood of under-delivery is the assumption that one solution fits many problems. This phenomenon happens all the time in reporting. A business analyst might agree to a request with the false sense of security that the query written to answer one client’s needs would work for those of another client. Without knowing if the data is available and how it is stored, this individual sets herself. Similarly, beware of the Endowment Effect[1]; while using existing resources is great, avoid the appeal of using existing reports just because you already have them built. Being 3 degrees off with an approach to the problem can lead to a destination in the wrong hemisphere!

  1. Don’t Overcomplicate

While oversimplification can be dangerous, so can taking a path that makes the request more complex than it needs to be. At times, a client wish list or a set of nice-to-have insights can create science experiments that get out of control. On the reporting side, a sophisticated data analyst can create “scope creep” when a particular request fuels his pet research project.  Having a good grasp of the deadlines and resources available for the particular requests provides the necessary constraints to ensure pragmatism.

  1. Stop Treating Symptoms

Frequently report requests arise out of a reactionary need. Without the luxury of long-term thinking or the visibility into an overarching reporting strategy, the demand and supply of reporting happens without intentionality. The information transaction is relatively superficial as the requesters are treating the symptoms of their problem (or someone else’s problem) without understanding the core challenge. Why not isolate the source so that you can start working on a comprehensive and permanent cure!  Achieving that position of strength involves discipline in the report triage process, evaluation of the request content patterns, and an understanding of the way the databases and reporting platforms are structured. Using the tactics above to zoom out of the individual daily reporting request minutia, you can evaluate the data and reporting needs of your constituents and build frameworks that will satisfy and persist.

The delivery of reporting and insights that pinpoint the greatest needs requires a disciplined and deliberate approach to data procurement, information analysis, and report construction; but you have to know what problem you are trying to solve. That’s why it is important to get the questions right first.

In the next post, I will review the process of taking the questions, once properly understood, and mapping out the key ingredients and context for the answers. From there, I will walk through some of the recipes that help us convert data to information and information to understanding.

_________________________________________________________________________________________________

[1] In psychology and behavioral economics, the endowment effect (also known as divestiture aversion and related to the mere ownership effect in social psychology) is the hypothesis that people ascribe more value to things merely because they own them. https://en.wikipedia.org/wiki/Endowment_effect

About PharmMD: PharmMD is a leading healthcare technology company providing data-driven medication adherence platform solutions that deliver value-based care outcomes. PharmMD partners with health plans, providers and other risk-bearing entities seeking to improve patient and employee healthcare through better medication adherence. Founded by healthcare and pharmaceutical innovators, PharmMD’s outcomes-driven solutions bolster initiatives in quality improvement and related regulatory compliance through better medication adherence. For more information, please visit www.pharmmd.com.