When a clinician gets a piece of data, a series of questions are asked prior to accepting the data for clinical use. The need for these questions has increased as interoperability has expanded.
Situation: receipt of clinical data
|
Questions
|
Action
|
|
Is the data of interest available?
|
If no, then order a test to get the data.
|
|
Is the data reliable?
|
If no, then order a test get reliable data.
|
|
Is the data relevant to the patient's current situation (too old, not fasting, etc)?
|
If no, then order a test to get relevant data.
|
|
Is the data unexpected for the patient's clinical situation?
|
If no, then troubleshoot.
|
|
Does the data need to be transformed or normalized?
|
If yes, then appropriately process the data.
|
If the data successfully navigates this gauntlet, then it can be used for clinical decision making.
Troubleshooting may include:
(1) repeating the test with the same specimen
(2) performing the test with a new specimen
(3) testing by another method
(4) check for interference due to drug or specimen feature (hemolysis, lipemia, etc)
Transformation of data may involve:
(1) unit conversion
(2) dividing by body weight or body surface area
(3) creating a Z-score
(4) interpret based on age
