Inside Angle
From 3M Health Information Systems
For the good of specificity
We know that being more specific in coding can affect and potentially increase reimbursement depending on the type of visit and conditions. Today, however, I would like to focus on another aspect of being more specific—tracking conditions which enables greater and more accurate data analytics. I call this “For the good of specificity.”
Let’s take asthma to illustrate the meaning of this phrase. In ICD-10, Asthma has many specific codes. In fact, not only can asthma be classified as “Mild intermittent,” “Mild persistent,” “Moderate persistent,” “Severe persistent,” or “Other and unspecified,” but within each of these classifications, the condition of asthma can be made even more specific! Though specificity can matter for DRG reimbursement of asthma, if a hospital is using HCCs as their reimbursement model, then being more specific won’t affect reimbursement in the same way. In this scenario, why should the coder bother with more specificity? It’s because of the good specificity can do!
Imagine the kind of power to improve care if doctors and healthcare professionals could look at populations as a whole and figure out trends in data for certain types of conditions. For example, are people with “Moderate persistent” asthma more likely to have other types of common conditions than those with “Mild persistent”? Perhaps one type of specific code under “Mild persistent” correlates with patients who are more likely to develop another condition—I’m not a doctor or healthcare provider, but I do know data analytics and the good that can come from specific data! Without consistent and accurate specificity, fewer interactions and correlations can be found in data.
Using data in the way I described above has already happened in some organizations and will continue to grow in the years to come. In fact, a recent article in HealthIT Analytics titled “HIM Pros Must Shift from Coding to Health Informatics, Analytics” discusses the future of HIM departments and coding as an industry. The article refers to a white paper from AHIMA which recommends HIM professionals “prepare for a future of informatics and big data analytics by acquiring new skills sets.” Specificity will play an important part in the future of HIM and also of health care., There usually is a difference in reimbursement as we become more specific, but we are also opening the door for greater data analytics in coding and in health care for the future.
Clarissa George is a business intelligence specialist at 3M Health Information Systems.
2 thoughts on “For the good of specificity”
Leave a reply
Clarissa George
Business Intelligence Specialist
Clarissa earned her BS in Business Administration with an emphasis in Marketing from Brigham Young University and joined 3M Health Information Systems in 2013. She enjoyed different roles in the…
Good post. Between this and your last one in March on specificity as well, I’m beginning to wonder how payment models will change for office visits too.
As a biller for a group practice, I know that the old fee-for-service payment system is not going to be here forever, but do you also see payments possibly differing based on specificity for outpatient visits as well? (For example, the data you used on A-fib led to to think the same for many other conditions I code on a day to day basis.)
Admittedly the biggest problem is just getting the providers to be more specific in their chart notes. It’s very common for them to just say “asthma” despite all the variations you list. I tend to stick with J45.998 “other asthma” unless they give a certain type.
As always, a good read. I love a good coding theory discussion!
Thanks for the comment Andrew, of course it’s possible that specificity could affect outpatient as well. It’s conceivable even that within an organization implementing bundled payments, that organization could look at the specificity of coding in helping disburse payments. Your second point with the asthma example is very interesting and in fact is something we’ve looked at in our data and hope to talk about in a future blog post. In looking at the data, it becomes quite clear that some of the unspecified codes are not widely used across any enterprise while others vary widely. The variation shows that it must be *possible* for the documentation to be specific enough but it’s either not widely documented specifically or that coders do tend to “just stick with” the code they’re used to using.
Thanks for reading!!