Executive Managing Director at Healthbridge, Luis da Silva, shares his insights into how doctors can navigate ICD-10 coding, improve accuracy, efficiency and avoid Z56.3.

Why is ICD-10 coding such a big subject?

The release of ICD-10 meant that it was possible to capture much more detailed and precise healthcare information than ever before. It is an extensive diagnostic classification that defines the universe of diseases, disorders, injuries and other health related conditions.

ICD-10 also became the international standard for reporting diseases, conducting research, monitoring outbreaks, trends and population health at a global level. So there was a clear advantage to migrating to ICD-10 but it was still a contentious issue.

From a practitioner’s point of view, doctors had to change over from the old coding set to the new, much larger ICD-10 coding set which, in itself, came with real challenges. It was also an unavoidable change for a few key reasons, particularly in the South African private healthcare context.

The first reason is that ICD-10 is regulation. The Medical Schemes Act stipulates that all claims must contain relevant ICD-10 codes if you want to get paid. Secondly, Prescribed Minimum Benefits (PMBs) are identified by ICD-10 coding and therefore used to determine what is and isn’t covered under PMBs.

But coming back to the practitioners’ point of view, using ICD-10 gets extremely complicated when we look at the anatomy and structure of ICD-10 coding. There are no less than 68 000 ICD-10 codes to work with and this alone is the basis of the real complexities for practitioners.

Can you describe some of these complexities?

If you calculate the sheer number of procedure codes and NAPPI codes, in addition to ICD-10 codes, there are theoretically trillions of possible claim code combinations. However in practice, practitioners tend to construct their claims in a sequential manner and in so doing, this naturally limits the possible selections that follow.  If we break down how most GPs construct their claims, they’re left with over 201 000 possible combinations. This is better but still overwhelming for a GP to navigate in a 15 minute appointment while treating a patient at the same time.

At Healthbridge, we wanted to understand how the complexities of ICD-10 coding impact medical billing, among other things. We analysed a sample of 500 Healthbridge clients to find the correlation between ICD-10 codes captured and what GPs typically treat at their practices. One finding that the data revealed was that ICD-10 codes used at an individual per practice level varied greatly.

This is important when we look at issues of compliance and accuracy. Accurate ICD-10 coding is required from an ethical and regulatory perspective but it is also essential when it comes to medical billing. If you are not using the right ICD-10 codes, it will most likely reflect in the rejected claims from medical schemes, and your practice revenue.

But we’ve talked about the volume of ICD-10 combinations and checking and double checking the accuracy of every code will have an impact on productivity, so we have to take that into account too.

How can medical professionals improve the accuracy of ICD-10 coding without losing time and productivity?

It is absolutely possible to mitigate losses in productivity while still improving ICD-10 coding and medical billing. Here are a few key tactics to consider:  

• Attend ICD-10 training. There are numerous online and in-person training courses to consider. The minimum criteria you can expect when choosing a course is whether or not it is CPD accredited and hosted by a reputable institution. Beyond that, engage in training that can add real value to your day-to-day use of ICD-10 and help you improve your productivity.
• Use (or update) claim templates to reflect the ICD-10 codes typically used at your practice. Templates are a powerful way to maximise productivity and the accuracy of medical billing. Updating your claim templates eliminates the tens of thousands of unnecessary ICD-10 coding combinations and reduces the risk of errors in medical billing, too. Review the templates inside your EMR and practice management solution to implement or update as necessary.   
• Embrace automation designed to boost productivity and efficiency. Practitioners who choose to use digital solutions in their practices have a distinct advantage in productivity and efficiency. Your chosen solution should make ICD-10 coding straightforward.
• Performs tests with medical schemes. Take a look at your data to see what claims are rejected and whether there is a correlation with the ICD-10 coding. Making sure that the codes you are using in your practice management and billing systems correspond with what is acceptable to a medical aid. This is also another reason to update your claim templates in your system.

Your service provider should be able to perform these tests for you, but it’s important to do so to ensure accurate medical billing and reimbursement from medical schemes.

Where could AI help?

What we have at the moment in South Africa and many other parts of the world is a scenario whereby technology has to fit seamlessly into a consultation setting. But often doctors have to give the patients their ‘undivided’ attention and still capture codes and patient records electronically – which inherently divides their attention. Some think that the use of digital systems within the consultation setting could be more seamless and that’s possibly where AI comes in.

What machines are much better at doing then we are, is processing large amounts of data to find patterns. The more we harness that, the more doctors will be able to capture the patient encounter the way it naturally flows.  AI can be leveraged in the background to review the patient’s history, assimilate data, develop a hypothesis and suggest a possible course of action. I think that’s what we can look forward to in the coming years in healthcare in South Africa.