August 23, 2018
Medication adherence in elderly patients
Medication adherence is a very critical part of the treatment in patients with chronic diseases. Elderly patients aged ≥ 65 years are prone to non-adherence of medication due to their underlying multiple chronic conditions and higher risks of polypharmacy. Polypharmacy is defined as the simultaneous use of multiple drugs to treat a single ailment or condition, or the simultaneous use of multiple drugs by a single patient, for one or more conditions.
Elderly people with chronic disease like diabetes mellitus, hypertension, arthritis, Alzheimer’s disease, dementia and respiratory disorders take multiple medications at a time. Poor drug compliance in these patients increases the costs for health care system due to the medication wastage.
A review published by the Annals of Internal Medicine found that non-adherence of medications causes nearly 125,000 deaths in a year. Medication adherence incurs between $100–$289 billion a year.
AI for Health Care Agents
Artificially intelligent Health Care Agents are increasingly being used to improve adherence and outcomes in elderly patients.
Agent is a person or a tool that can make decision or act for a person when that person is not able to do so. Health Care Agent makes health care related decision for a person when that person is not able to make a decision or perform a specific task on his/her own. Within the context of Artificial Intelligence, agents are autonomous entities that observe through sensors, act upon the surroundings and performs an activity to achieve the specific goal.
I this article, I present a the state-of-the-art AI tool that was developed by a group of researchers at Umeå University in Sweden. The tool can be used to assist elderly people in their daily activities.
Smart Augmented Reality mHealth
Researchers at the Department of Computing Science and Department of Community Medicine and Rehabilitation at Umeå University in Sweden have developed a smart mHealth application called “Medication Coach Intelligent Agent (MCIA)”. The application is based on augmented reality. The MCIA manages various information such as medication plan of patient, medication restrictions, and the patient’s preferences. It also manages sensor input data from an augmented reality enabled eyeglasses/headset.
After the application is fed with all the required information, the MCIA produces the holistic decision. The decision provides instructions to the patient about medication intake based on individual patient’s condition (individually tailored).
The application can help elderly patients with the chronic diseases in their drug adherence issues. One of the innovative and interesting things about this applications is that it has persuasive techniques that can engage the patient in the medication intake.
When an elderly patient gets a prescription from a physicians, he/she gets instructions about medication adherence and information about the dose and timings of medications. These medications should either be taken on regular basis or taken on demand. The main purpose of MCIA is twofold. Firstly, MCIA makes sure that the patients take their regular medications on the designated times. Secondly, MCIA makes sure that the patients do not exceed the maximum dosage per day of their medications that needs to be taken on demand.
The architecture for MCIA consists of three major compenents; the MCIA, external agents and databases. The external agents and databases provide MCIA the input it needs to supply the services to the patients. The application can be mounted on a camera in eye glasses and gives visual information about the medication adherence.
Based on the information fed into the application, it displays images to the patients whether they should take the medication or not. The augmented reality superimposed images visualize instructions to the patient. It gives them visualization of the medications’ dosages and timings.
Proof-of-concept prototype of MCIA
The researchers conducted proof-of-concept prototype of the MCIA application. They tested the prototype on individuals to see if there was any difference in their willingness to use augmented reality headset for the medication use with respect to their ages and experience with the smart technology.
The researchers tested the prototype on 15 individuals. Their results showed that 20% of the participants perceived the application as hard to use while 13% thought that they would need a lot of training to use the technology. Of all the participants, 100% were willing to use the technology in the future.
Future of AI in elderly care
Elderly care assistance is one key area where AI agents can be used to help elderly people with certain chronic diseases or physical disability related conditions. Population of elderly people is increasing with the demographic shift. This demographic shift is in constantly increasing the incidence of chronic diseases in elderly people. More and more elderly people need assistance in their day to day activities.
AI has the potential to alleviate the burden of care giving for elderly people and has recently been the focus of many researchers in this field. The innovative tool developed by the Swedish researchers shows that AI has the potential to become a game changer in the medication non-adherence for elderly patients with multiple chronic conditions.
Rafiq Muhammad, MD, MIHMEP, PhD candidate is Medical Doctor and Health Care Manager, currently enrolled as a PhD Candidate at the Department of Learning, Informatics, Management and Ethics (LIME), Karolinska Institutet, Stockholm, Sweden. His PhD project explores the application of Artificial Intelligence in the development of Decision Support System to patients with Multiple Chronic Conditions.