Understanding The Impact of Artificial Intelligence On Professional CASE Management Practice




Artificial intelligence (AI) can be defined in many ways, but it generally refers to the ability to enable computers and robots to emulate human thoughts, behaviors and tasks. Data scientists use a variety of tools to analyze data and solve problems. Machine learning (ML) is one form of AI that uses technologies and algorithms to identify patterns and insights from data by applying that learning to continuously improve decision making. ML systems analyze numerous case scenarios while continuing to learn and adapt in order to achieve the most accurate prediction for a clinical outcome. The whole idea is translating data into insights for decision making, allowing for more efficient processing by mimicking human cognition. Natural language processing (NLP) is another form of AI which uses algorithms to analyze the spoken or written word and communicate back. The pervasive use of virtual assistants, such as Siri or Alexa, are examples of the use of NLP in our everyday lives. In healthcare, AI incorporates and analyzes medical and related data with the goal of predicting the likely outcome. AI will likely result in moving case managers and other healthcare professionals into the realm of algorithm-based healthcare.


There are a number of ways that AI can enhance professional case management practice with human oversight of AI outputs, including:

  • Clinical pathways: AI could assist in identifying clinical pathways based on patient data, including recommendations for interventions. Depending on the quality and quantity of data, this could assist case managers as the patient’s condition changes or when the patient is facing roadblocks to recovery. This could also be helpful when there is a broad constellation of physical health, behavioral health, oral health and social determinants of health that are all impacting the case management plan of care.
  • Content creation: AI could provide recommendations for culturally and linguistically appropriate information and education for clients. This presumes the AI platform has information about the demographics of the population and the individuals, such as race, ethnicity, sexual orientation, gender identity, etc. The content could be adjusted to meet the literacy and health literacy needs of populations and individuals.
  • Predictive capabilities: AI has already proven that algorithms, when properly applied, can fairly accurately predict the length of a hospitalization and the anticipated discharge status, for example. This could prove very helpful for case managers who are involved in discharge planning activities, so that the case manager can conduct anticipatory planning activities based on the predicted course of the hospital stay.


There are some limitations and cautions in using AI in professional case management practice, including:

  • Need for individual assessments to confirm needs: Although AI can predict risks for a population and the potential risk for a client, the actual presence of the need for services should occur within the context of a case management assessment. For example, a client may be at risk for social determinants of health based on the zip code of their residence, but that does not necessarily mean that a specific individual client actually has that need until the case manager conducts the assessment.
  • Discriminatory algorithms: There is concern that some AI algorithms have been created using only data from patients with certain demographics, such as only white men. This limited usage could result in incorrect or biased outputs for communities of individuals who were not included in the data set.
  • Limitations in data: The data that feeds the AI algorithms could be limited, thus leading to incorrect conclusions regarding the needs of populations and individuals. For example, if the data being used is only medical claims data, this could result in wrong conclusions.
  • Age of the data: If the data being used is old, then the AI algorithms are not using the most up-to-date information when forming recommendations. There may also be issues with outdated coding schemas that may not be aligned with more current coding frameworks.
  • Lack of alignment with evidence-based care: If the data set does not have the most current evidence-based care guidelines, the AI algorithms may not reach the best conclusion based on these current practices.
  • Limitations of incorporating sociocultural factors: At this time, AI is not able to incorporate social, cultural, religious/spiritual and other related factors of a client, such as preferences for complementary and alternative medicine practitioners. The case manager will need to continue to assess the sociocultural aspects of the client in order to develop a responsive and respectful case management plan of care.


There are a number of actions case managers can take in preparing to incorporate AI into professional case management practice, including:

  • Continuous learning and development: It is our duty to our clients, our organization and ourselves to continuously learn and develop our knowledge, skills, and abilities in the provision of professional case management services. With all of the opportunities and challenges with AI that are rapidly evolving, case managers need to stay abreast of these AI developments.
  • Involvement in the organizational AI Initiatives: Case managers can become aware and involved in the AI initiatives that are being developed at your organization. Case managers can give a voice to what does and does not work with AI in the practice of professional case management services.
  • Being involved with the design and implementation of AI tools: Since case managers could use AI tools that are valid and reliable, there may be opportunities to become involved in the creation and deployment of AI tools that can enhance the practice of professional case management services.
  • Determining how to use AI in client-centric ways: The key value of professional case management practice is being client-centered, and AI has the potential for further enhancing the focus on the client. There is also the possibility that AI tools are not client-centered if they are not designed with the client in mind. Case managers have the ability to keep the client front and center in AI projects.
  • Developing and implementing best practices and safeguards to prevent AI-related harms: Although AI offers many opportunities for enhancing professional case management practice, there are also risks, especially within the context of healthcare. Case managers may want to consider learning and development requirements before deploying AI, identifying use case limitations, transparency requirements with clients when using AI tools, and other best practices. These kinds of safeguards should be put in place before being deployed with clients.


Technology has been and continues to be helpful in enhancing professional case management practice. AI has the potential to further optimize effectiveness and efficiencies with case management services. This could include recommendations on clinical pathways, creating content that meets the cultural and linguistic (including literacy and health literacy) needs of the clients served, and predicting risks for individuals and populations. However, there are some valid concerns about how algorithms are developed that are not inclusive, resulting in incorrect recommendations and decisions. AI tools are limited by the quality, quantity and types of data elements that power the calculations of a system. This all means that we must live in cognitive dissonance, knowing that there is optimism for how AI can enhance practice while acknowledging and recognizing the limitations.

Professional case managers need to be aware of the opportunities and challenges with AI, so that AI could become one of the useful tools in the case management toolbox!


Bonavitacola, J.R. (2024). AI capable of detecting PAH associated with congenital heart disease. American Journal of Managed Care. January 9, 2024. https://www.ajmc.com/view/ai-capable-of-detecting-pah-associated-with-congenital-heart-disease. Retrieved 1/30/2024.

Columbia Engineering. Artificial intelligence (AI) vs. Machine Learning https://ai.engineering.columbia.edu/ai-vs-machine-learning/. Retrieved 1/26/2024.

Davenport T, Kalakota R. The potential for artificial intelligence in healthcare. Future Healthc J. 2019 Jun;6(2):94-98. doi: 10.7861/futurehosp.6-2-94. PMID: 31363513; PMCID: PMC6616181. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6616181/. Retrieved 1/26/2024.

Emerson, J. (2024). CMS issues AI guidance for Medicare Advantage Plans. Becker’s Payer Issues. February 8, 2024. https://www.beckerspayer.com/policy-updates/cms-issues-ai-guidance-for-medicare-advantage-plans.html. Retrieved 2/9/2024.

Kantor, G.S.; Renton, M.; Rakover, J.; & Barker, P.M. 5 takeaways from a discussion on generative AI QI. Institute for Healthcare Improvement (IHI). https://www.ihi.org/insights/5-takeaways-discussion-generative-ai-and-qi. Retrieved 1/26/2024.


Comments are closed.