3 real-life examples of how coding automation improves primary care practices

Times are tough for physicians. A shortage of doctors, exacerbated by the pandemic, has led to a higher patient-to-physician ratio. More patients equals more administrative work for doctors. This translates into less time to see patients and more stress on physicians. And it often leads to increased coding or documentation mistakes that cut into a practice’s bottom line. How can practices improve the situation? Coding automation may be the answer.

You may think that coding automation primarily impacts coders. While that is partially true, the technology has far-reaching effects on practices and a direct impact on the health and happiness of physicians. How is this possible? It all starts with a doctor's administrative duties.

According to one study, 58% of physicians list administrative duties as the number one cause of burnout. How bad is it? Research published in 2022 found that primary care physicians would need to spend 2.6 to 3.2 hours per day on documentation and inbox management to provide guideline-recommended care for a panel of 2,500 patients. These administrative burdens have a negative impact on physicians' well-being and effectiveness – contributing to frustration, work-life imbalance, and reduced productivity.

Coding automation can help alleviate this burden. Not only does the AI technology free physicians from spending time on coding, it also improves the quality and velocity of the revenue cycle for the practice, flagging incomplete or missing documentation earlier for staff to correct. This more efficient approach results in more accurate coding and revenue capture, as well as more time for physicians to spend with patients.

3 examples of how coding automation ensures proper revenue capture in primary care

As AI coding is more comprehensive and specific than its human counterpart, it helps practices obtain appropriate reimbursement for all their work. Below are some real-life, common examples of how this benefit manifests in primary care practices.

Example 1: Documenting an annual checkup and problem visit in the same appointment

During an annual wellness checkup, it’s not unusual for a patient to present with a new problem or get new treatment for a chronic illness. An annual wellness visit and a problem visit can be separately reportable, but the physician may forget to document this accordingly.

AI, however, will catch this error and correctly code for both visits when appropriate. For example, in this scenario, an established patient may justify code G0439 for the annual wellness visit and code 99213, or another evaluation and management (E/M) code, for the problem visit. With relative value units (RVUs) of 3.84 and 2.68, respectively, capturing both services represents compensation around $216 using the 2023 CMS conversion factor of $33.06 per RVU.

Example 2: Coding for vaccinations

Both doctors and coders know that vaccine coding is complex. Vaccinations involve two different code types: a code for the type of vaccine and a code for its administration.

For vaccines, the codes vary depending on the specific type of vaccine. For administration, the codes can differ due to the method of administration (oral, intramuscular, etc.), patient age and whether a consultation was given, and the number of components.

If this doesn’t sound confusing enough, the codes can also vary based on the manufacturer, formulation of the vaccine, and updates to coding guidelines. Even if a doctor is detail-oriented, it can be easy for coding mistakes to happen. But not with AI. Coding automation will capture all of these details and code for them to ensure you receive proper reimbursement.

Example 3: Assigning the correct E/M code level

In addition to their heavy workload, physicians have a limited amount of time for documentation after a patient encounter, with pressure to meet the next patient waiting. This can result in E/M code errors, especially if the patient’s condition is complex and requires extensive treatment.

Coding automation can be a reimbursement savior here. It can pick up on specific elements – such as point of care tests, multiple conditions addressed, or new prescriptions – that a doctor may forget to account for, enabling correct coding of the encounter and reducing the effort that doctors need to spend on documentation. The difference is non-trivial. For example, while a 99213 entails 2.68 RVUs, a more extensive case justifying 99215 involves 5.31 RVUs, or about $87 more.

Though physicians may be busier than ever, coding automation provides a smarter way to reduce their administrative burden while simultaneously improving revenue capture. For practices struggling to keep up with patient demand and billing, AI coding may be just what the doctor ordered.

Taylor Ross is the Strategy and Operations Lead at Fathom, a health technology company that uses deep learning AI to automate medical coding for a wide range of specialties and practices. She leads coding quality and verification, working cross-functionally with customer success, engineering, and product development to support healthcare organizations through onboarding, production, and ongoing quality assurance. Taylor holds a Certified Professional Coder (CPC) certification through AAPC and a Certified Coding Specialist (CCS) certification through AHIMA, and a bachelor’s degree from the University of Pennsylvania.