Ask a healthcare professional — or a (friendly) stranger on the street for that matter — the question “what is health?”, and it is a foregone conclusion that every responder will commit the grave sin that high school grammar teachers have been railing against for centuries: they will answer a question with a question.
Indeed, it does not matter if they have a decade of medical school under their belt, or the closest they want to get to a hospital is binge-watching Grey’s Anatomy. The response in one form or another will be along the lines of: “well, what do you mean?”
What this illustrates, is that in order to understand what health actually is — and just as importantly, to understand what health is not — sufficient context is required. After all, health is multi-faceted and covers everything from physical fitness to psychological well-being. What’s more, what one individual essentially defines as “good health” or “bad health” may differ from another. For some, the absence of pain constitutes being healthy. For others, it is about feeling energized. Still others may bring in numbers, like BMI, blood pressure, hours of sleep, and so on.
In a similar sense, hospitals that generate information from their various systems and obtain it from third parties (e.g. post-acute care organizations), must recognize that data on its own has no context. It is raw material that can lead to faster and smarter decisions. But this only happens when hospitals know what questions to ask of the massive amount of data in their ecosystem — i.e. when they supply the right context — so that information can translate into intelligence.
Here are some example questions with simplified, yet realistic scenarios:
Scenario: The data says that a hospital has incurred two hospital-acquired DVTs in the past year. But does this number include all surgeries, or just hip and knee replacements? Does it cover all hospitals in the health system, or just one facility?
Scenario: A hospital re-admission rate is 10 percent over the past quarter, and 20 percent over the past year. Also within the past year, a significant change was made in care manager assignments. To what extent is the improvement in re-admission rate related to the change? And if there is a connection, when did improvement start and at what velocity?
Scenario: A hospital reaches its patient safety goal for the month. But is this number improving? Or dropping off? And at what rate?
Scenario: A hospital experiences a reduction in revenues for the month. Is this due to delay in getting bills out? Or are patients going elsewhere for service? The corrective action is very different for these two scenarios!
Scenario: A physician receives low marks for patient satisfaction. But is this based on 10 surveys or 100? Without enough data points, the sample size is too small, and conclusions are between questionable and spurious.
Whether their goal is to reduce re-admission rates that currently exceeds acceptable norms, identify top performing physicians to establish best practices, or achieve any other priority or objective, hospitals need context to make faster, smarter and safer decisions. And really, isn’t that the purpose of healthcare data in the first place?
At Polaris, we have the expertise, insight and tools that enable hospitals to transform their raw information into actionable intelligence — and ultimately achieve measurable improvements and sustainable results.
To learn more about our solutions, technologies and approach, contact the Polaris team today.