Carlo N. De Cecco
MD, PhD profesor asociado de radiología e informática biomédica en la Facultad de Medicina de la Universidad de Emory, Atlanta, EEUU.
Artificial Intelligence (AI) is changing healthcare and medical practice around the world. It is estimated that global investments in AI, in this sector alone, will reach around USD 36 billion by 2025, a 50% growth from 2018. In combination with telemedicine, AI can play an pivotal role in improving and expanding people’s access to the public health system.
One of the main problems at the global level is the lack of access to health care. The World Bank and the World Health Organization (WHO) published a report in 2017 where they affirmed that 50% of the world’s population does not have access to health services, compounded by an aging population and a surge in chronic diseases. The population of Latin America and the Caribbean (LAC) is aging at a rapid rate. According to United Nations estimates, the population over the age of 60 in the region is projected to rise from today’s 11 to 25 percent within 35 years. They also pointed out that in the United States almost 45% of people have at least one chronic disease and that this number is expected to continue to grow. Patients with chronic and elderly diseases will therefore require more and more medical care and follow-up. According to the latest WHO data, 75% of deaths in the region are attributable to chronic diseases, nearly a 20-percent rise from 1990
This also raises the cost of health care and therefore impacts patients’ finances. Along with the increased burden on the health system, there is an upsurge in medical costs. Thus, controlling the levels of public health spending requires people to pay more out of pocket. Only six countries in the Americas have direct household spending below 20% of total health spending. According to the WHO, this figure protects their citizens against the risk of impoverishment or catastrophic health expenditures. According to data from 2014, Cuba is by far the largest health investor in the region, with more than 10% of GDP, followed by the United States, with just over 8%. Uruguay reaches the 6% target, followed closely by Panama, which nevertheless fails to comply with the WHO recommendation. Conversely, Latin American countries that invest the least in health are Haiti and Venezuela, with less than 2% of their GDP. Stress on primary health care could be greatly reduced by optimizing administrative processes related to preventive screening, immunization, referrals, and appointment systems, among other areas.
AI-driven health system improvement opportunities
The use of AI, in addition to task automation, helps with patient planning, diagnosis and prognosis, making health care more efficient, as it not only reduces costs, but also allows for remote results analysis, which leads to better distribution of health care services. Telemedicine allows patients to contact healthcare providers online and offers an alternative to face-to-face visits. Serving as the gateway to medicine, it offers an alternative to high-cost care environments or for patients who have to make long trips to health centers. Using AI capabilities in a telemedicine environment helps reduce the burden on clinical staff, which has a high impact on increased focus on patient care. An example of telemedicine advances is seen in Uruguay, where the Agency for Electronic Government and Information and Knowledge Society (Agesic) has promotes for more than a decade electronic medical records in the National Integrated Health System, which is interoperable between institutions and throughout the territory.
Successful development approach
One of the critical aspects for success in the digitalization of healthcare using AI and telemedicine is to have an adequate infrastructure, for which the initial investments are essential, preferably in combination with specialist partners to ensure compatibility across all systems. A common data infrastructure could help facilitate training and deployment.
In addition, for telemedicine and AI to be successful in clinical practice, regulations must allow for the optimal use of a model that covers this kind of health care, as well as medical licensing and liability.
There are two key financing-related factors that are important to ensure a sustainable approach. The first is related to the medical provider, for whom there must be a financial incentive to implement AI and telemedicine. The second has to do with patients, for whom telemedicine must be financially appealing, and ensure costs similar to traditional health care.
Training for AI use in telemedicine
Education should be the foundation of any approach to implementing telemedicine and AI, for both healthcare staff and patients. All medical staff should receive training on how to use AI tools and software, in addition to training on communicating with patients electronically. It is also essential that in rural areas local staff are trained to use and maintain the equipment. At the same time, patient training is also necessary to overcome generational, cultural and educational barriers, particularly for the elderly to make more efficient use of these tools.
Lastly, data protection is essential for healthcare in general and becomes more important with the growing risks associated with digital care. A standardized approach to assessing platform risks improves use of and trust in these systems. To that end, regulations should be geared toward minimizing risks and governments should promote data transparency to avoid discriminatory biases when using AI algorithms.
AI and telemedicine will play a pivotal role in healthcare, given that the COVID-19 pandemic has accelerated the implementation of telemedicine, a trend that is expected to grow continuously over the coming years.
This study was produced within the framework of the State Digital Innovation Board regional initiative, under the Vice Presidency of Knowledge of CAF—development bank of Latin America—, which will be presented at the ExperiencIA, data and artificial intelligence in the public sector forum<> on September 15.
