The impact of AI on coding and documentation in dermatology

By Tiffany E. McFarland, RHIT, Analyst, Coding & Reimbursement, March 1, 2025

Academy coding staff address important coding topics each month in DermWorld Coding Consult. Read more Derm Coding Consult articles.

Artificial intelligence (AI) is rapidly transforming the health care industry, and its impact is particularly notable in dermatology. One significant area of change is in medical coding and documentation, a crucial process for translating diagnoses and treatments into standardized codes for billing and insurance purposes. Traditionally a time-intensive task, medical coding has been streamlined by advanced AI tools, such as computer-assisted coding (CAC) and autonomous medical coding (AMC).

These technologies present new opportunities to enhance efficiency and accuracy in medical code assignment by leveraging varying degrees of automation and the required level of human intervention. However, they also introduce unique challenges for dermatologists and non-physician clinicians (NPCs) that must be carefully navigated.

Differences in AI-driven coding technologies

AI-driven technologies in health care exhibit notable differences in their approach and application. For instance, AMC leverages AI, machine learning (ML), and deep learning (DL) to enhance coding automation. By employing continuous learning, AMC refines accuracy and adapts to evolving coding guidelines and regulations, reducing the need for manual adjustments. This approach allows AMC to assign codes with minimal human intervention, streamlining the coding process.

Conversely, CAC, as defined by the American Health Information Management Association (AHIMA), primarily utilizes natural language processing (NLP) or machine learning (ML) to identify key terms and phrases in clinical records that suggest specific codes. CAC systems independently analyze and extract information from documentation and propose codes that users can accept, reject, or modify, offering flexibility to add or delete codes as needed. This interactive process has the potential to improve both productivity and coding accuracy.

In addition, the AMA CPT Editorial Panel has created and added CPT^® Appendix S: AI taxonomy for medical services & procedures in the coding manual where they provide guidance for classifying various AI applications (e.g., expert systems, machine learning, algorithm-based services) for medical services and procedures into one of these three categories: assistive, augmentative, and autonomous.

The AMA further states that AI as applied to health care may differ from AI in other public and private sectors (e.g., banking, energy, transportation). It is important to note that there is no single product, procedure, or service for which the term “AI” is sufficient or necessary to describe its intended clinical use or utility; therefore, the term “AI” is not defined in the code set.

The AMA further states that the term “AI” is not intended to encompass or constrain the full scope of innovations that are characterized as “work done by machines.” Classification of AI medical services and procedures as assistive, augmentative, and autonomous is based on the clinical procedure or service provided to the patient and the work performed by the machine on behalf of the physician or other qualified health care professionals (QHP).

AMA AI taxonomy categorization & level of autonomy

Autonomous coding is particularly advantageous for routine procedures, such as wart destruction or phototherapy for inflammatory diseases. By efficiently handling straightforward cases, these technologies allow human users to focus on complex scenarios requiring critical thinking and nuanced expertise.

Benefits of AI in dermatology

AI-driven coding provides significant benefits for dermatology practices, where accurate code selection and modifier application are critical to preventing claim denials, reimbursement delays, and unwarranted audits. By automating routine tasks such as code identification and compliance checks with billing guidelines, AI empowers dermatologists, NPCs, and their coding staff to adopt a proactive approach.

This shift enables a stronger focus on improving and streamlining medical record documentation, which enhances AI accuracy in code selection. The result is fewer coding errors, more timely claim submissions, and a smoother revenue cycle — ultimately contributing to greater operational efficiency and financial stability.

Challenges of AI in dermatology coding and documentation

While the integration of AI offers numerous benefits, it also presents several significant challenges that require careful consideration including:

• Cost and risk of implementation: Implementation of AI-driven systems entails costs associated with technology acquisition, integration, and staff training. Additionally, the potential risks of implementation — such as system errors, underperformance, or failure to meet expectations — can deter practices from adopting these technologies without clear evidence of return on investment (ROI).
  
  The dermatology practice must also invest not only financial resources but also time to ensure their teams are well-equipped to use these tools effectively. During the initial phases of adoption, these requirements may slow operational workflows, creating temporary barriers to efficiency gains.
  
  Smaller practices may struggle to justify the expense, especially if resources are limited.

• Data and documentation quality: Another challenge is the potential for AI systems to misinterpret nuanced or complex medical records. In dermatology, subtle distinctions in documentation — such as differentiating between benign and malignant lesions — can have significant coding and billing implications.
  
  AI systems rely heavily on the accuracy and completeness of medical documentation to perform the function of code selection and allocation effectively, inconsistent or poorly maintained records can lead to inaccurate code assignment, resulting in claim denials or compliance issues.
  
  Errors in these areas may necessitate human intervention to review and correct inaccuracies, highlighting the importance of maintaining a collaborative approach between users and AI systems. That said, ensuring high-quality data inputs requires continuous training for physicians, NPCs, and staff to adopt thorough documentation practices, which can be a time-intensive process.
  

  ChatGPT enters the chat

  ---

  Read about the possibilities and pitfalls of using ChatGPT in clinical practice.

• Complexity of claims: Dermatology encompasses a wide range of services, from routine procedures to highly specialized treatments. Many claims involve nuanced coding decisions, particularly when dealing with modifiers, payer-specific policies, or rare diagnoses.
  
  While AI systems offer powerful capabilities, they cannot fully replicate the expertise of professional coders, particularly in handling complex cases. For example, determining the appropriate procedure modifiers or accurately coding complex dermatologic surgeries and services often requires careful judgment and a thorough understanding of the context.
  
  These complexities can challenge AI systems, which may struggle to interpret unique scenarios or adapt to frequent changes in coding guidelines without significant customization. This underscores the importance of skilled coders working in tandem with AI tools, allowing automation to enhance rather than replace human expertise.
  
  

• Competing IT priorities: Implementing AI solutions often competes with other IT priorities within a practice, such as upgrading electronic health record (EHR) systems, enhancing cybersecurity, or managing telehealth platforms. Limited IT resources may force organizations to delay AI adoption or allocate insufficient support, compromising the effectiveness of the implementation.

Addressing these challenges requires strategic planning, including assessing financial feasibility, ensuring robust training, maintaining high-quality data, and fostering collaboration between coding experts, AI systems, and IT professionals.

By proactively tackling these obstacles, including phased implementation timelines, dermatology practices can unlock the full potential and maximize the full value of AI to enhance coding accuracy and streamline operations while minimizing risks.

AI impact on dermatology clinical documentation

The success of AI in medical coding hinges on the quality of clinical documentation. AI systems depend on accurate, detailed, and structured records to assign appropriate codes effectively. Ambiguous or incomplete notes can result in coding errors, leading to denied claims, delayed reimbursements, and potential compliance issues. For dermatology practices, where precise coding is critical, maintaining high-quality documentation is not just a best practice but a necessity.

Dermatologists and NPCs play a vital role in ensuring documentation supports AI-driven coding. Comprehensive and clear documentation helps AI tools maximize their potential, improving coding accuracy and streamlining billing processes. However, the integration of AI also raises important ethical and legal considerations. Issues such as data security, patient privacy, and algorithmic transparency must be managed to build trust and ensure compliance with regulations like HIPAA. Additionally, AI algorithms must be regularly assessed for biases that could impact coding fairness, especially given the diversity of patients and clinical scenarios encountered in dermatology.

Balancing automation with human expertise

Despite its promise, successful AI adoption requires addressing critical challenges. These include maintaining robust human oversight to handle complex or ambiguous cases, ensuring consistent data quality, and navigating ethical considerations. AI is not a replacement for human expertise but rather a powerful ally that complements it.

In a specialty where accurate coding is essential for reimbursement and high-quality care, AI stands as a transformative tool. Dermatology practices that embrace these innovations strategically can strike an optimal balance between automation and human judgment, creating a more efficient, precise, and patient-centered health care environment.

Does AI have a future in dermatology?

Yes.

As AI technology continues to evolve, its potential applications in dermatology are vast. Emerging innovations could include:

• Tailored AI tools: Customized solutions designed specifically for dermatology practices to address unique coding challenges.

• Real-time coding support: AI integration into electronic health records (EHRs) to suggest codes during patient consultations, ensuring immediate documentation accuracy.

• Teledermatology integration: Seamless alignment with telehealth platforms to enhance remote care while maintaining coding precision.

These advancements hold promise in reducing administrative burdens, allowing dermatologists and NPCs to focus more on patient care and clinical decision-making. By streamlining workflows and improving coding accuracy, AI can contribute to better financial performance and clinical outcomes for dermatology practices.

Additional DermWorld Resources