Prediction of medication adherence using artificial intelligence.

Artificial intelligence (AI) is increasingly present in the health system fulfilling various functions such as helping patients comply with medication regimens (1). The World Health Organization states that up to 50% of prescribed medications are not taken as intended, which has been called "the problem of medication adherence" (1). This problem is responsible for a series of consequences, such as low therapeutic efficacy, risk of harm to patients, and the creation of unnecessary costs for the different parties involved.

But let's go by parts; first, it is essential to analyze the factors that affect adherence. There are multiple reasons, which can be forgetfulness by the patient, complexity of the regimen, complications with medical prescriptions, adverse effects, or lack of coverage by the health system. It is also important to mention that many patients do not have the possibility of accessing the necessary care to mitigate the factors that affect medication adherence (2). Similarly, this problem directly affects the work of doctors and health personnel. They need simple but straightforward solutions that allow them to identify the problem early and intervene to provide timely help to patients without ultimately affecting treatment.

AI can solve all these problems by improving clinical outcomes and improving patients' quality of life. In various studies, AI has shown relatively good results in measuring medication adherence levels (1). Likewise, higher patient compliance rates and better knowledge of these are offered when taking the medication. Similarly, several improvements have been observed in optimizing the doctor's time, helping them provide better quality patient care. Reference is also made to the developing potential of remote monitoring and real-time communication (1).

However, these named benefits do not mean there are no concerns when using AI to solve this medication adherence problem. One of these concerns is the patients' characteristics since not all have the exact needs or the same description. Likewise, there is a significant concern among family members and health professionals since it can lead to depersonalization in medical care. This has led to a growing skepticism toward automation, and some minority groups are left out of the generalization. However, all these problems are covered since AI solutions, on the contrary, provide a humanization to medicine and allow personalized care to be generated in real-time.

References:

1. Babel A, Taneja R, Mondello Malvestiti F, Monaco A, Donde S. Artificial Intelligence Solutions to Increase Medication Adherence in Patients With Non-communicable Diseases. Front Digit Health. 2021 Jun 29;3:669869. doi: 10.3389/fdgth.2021.669869.
2. Vidya Koesmahargyo, Anzar Abbas, Li Zhang, Lei Guan, Shaolei Feng, Vijay Yadav, Isaac R. Galatzer-Levy, Accuracy of machine learning-based prediction of medication adherence in clinical research, Psychiatry Research, Volume 294,2020,113558,ISSN 0165-1781, https://doi.org/10.1016/j.psychres.2020.113558.

If you want to know more about Arkangel Ai contact us here and one of our team members will contact you for a one-on-one session.

‍