By: Marisa Colston, PhD, LAT, ATC
While working on the NATA Professional Responsibility in Athletic Training Committee on the research initiative for identifying athletic training shared professional values, we certainly did not envision at that time, their critical application to the use of Artificial Intelligence (AI). The potential for substantial advancements in healthcare through AI is exciting, yet it also brings forth numerous ethical considerations that must be addressed to make certain its use benefits all stakeholders.
We are truly at the proverbial tip of the iceberg in the discovery of various AI applications. In athletic training specifically, AI can be used to advance techniques in injury prevention and prediction, personalized training programs, performance analysis, rehabilitation, virtual reality simulations, and wearable technology integration, among many other uses. These AI tools can analyze vast amounts of data in the blink of an eye, providing a cost-efficient and time-saving way to use this technology to optimize an athlete’s potential while minimizing risks.
Like any other advancement, the user, not the tool, determines its ethical use. Implementing safeguards is critically important to ensure this technology improves care without compromising our professional values. The remainder of this blog will briefly outline some of the ethical issues to consider, along with the professional values (Caring and Compassion, Integrity, Respect, Competence, and Accountability) reflected within each issue.
First and foremost, privacy and security of all health data must be rigorously protected. There is a razor’s edge line between utilizing data for an athlete's benefit and infringing upon that individual’s privacy rights. Maintaining confidentiality is an expected competence in athletic training and competence, is one of our shared professional values.
Another issue to consider is that not all AI algorithms are fair and equitable for all people groups. You have likely heard or read that these algorithms are only as good as the data being fed. Biases contained in historical data may be perpetuated in the resulting AI predictions. Although it is the responsibility of the developers to create algorithms that are fair and equitable, athletic trainers, in supporting the professional values of caring, compassion and respect, should adequately review the information to promote judicious application.
The "black box" nature of many AI systems results in a lack of transparency in how decisions are made. Athletic trainers must understand why an AI system might recommend specific interventions or predict certain outcomes. A decision tree is an example of a machine-learning model where decision outcomes can be understood and explained. Transparency is essential to uphold the shared professional values of integrity and accountability.
Another ethical issue related to accountability relates to the ‘blame game.’ Who will claim responsibility if an AI-powered recommendation leads to an athlete’s injury? Clear lines of accountability must be established when using AI in all practice settings. Oversight measures should remain part of any decision-making process.
Finally, the use of AI should enhance, not suppress, a patient’s autonomy and ability to make informed healthcare decisions. This demonstrates the professional value of respect. Athletes must fully understand how their data will be used by AI systems, what decisions may be influenced by AI, and the potential risks involved to give proper informed consent.
The power of AI in athletic training can be harnessed responsibly by taking these ethical implications into consideration and implementing safeguards. We can use this technology to enhance the health care services we provide while upholding our shared professional values.
