Increasing health literacy with AI drives better health outcomes and lower costs

In this article, I’ll provide a few ideas on how we can use recent developments in learning science and artificial intelligence to improve health literacy at scale, driving better health outcomes and reduced costs for patients, pharma, providers, and payers.

Challenges in improving the health of people and delivering affordable healthcare

The only national data published so far on health literacy skills in the United States of America, released in 2006 by the U.S. Department of Education, found that limited health literacy coincided with adults self-reporting the worst health. While 53% of American adults were found to have Intermediate health literacy, 36% were found to have Basic to Below Basic health literacy. Similarly in Europe, findings from the European Health Literacy Survey (HLS-EU) found that at least 1 in 10 respondents showed insufficient health literacy, and almost 1 in 2 had limited (insufficient or problematic) health literacy.

A systematic review of peer-reviewed studies found that “patients with low literacy had poorer health outcomes, including knowledge, intermediate disease markers, measures of morbidity, general health status, and use of health resources.” Even after adjusting for age, gender, race, health status, and socioeconomic status, the odds of hospitalization at a public hospital over 1 year were 1.3x to 1.7x higher for patients with lower literacy than for patients with higher literacy.

The findings shed light on some of the challenges we face in delivering affordable healthcare given that limited health literacy has been linked to increased cost for all parties in the healthcare system.

Breaking down the concept of “Health Literacy”

The U.S. Department of Health and Human Services (HHS) defines health literacy as:

“the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.”

If we break down the definition of Health Literacy from the HHS, we can come up with achievable solutions that build towards better health outcomes.

  1. “To obtain basic health information” requires that the right health information is easily and readily accessible by the patient. Somewhat challenging given that there is a breadth of health information readily available online, and patient care education from the hospital can sometimes come in an analog format, making it hard for the patient to easily access the right health information.
  2. “To process basic health information” requires that the right health information be accessible through an easily communicative channel. This is more challenging given that health information is often described using complex and jargony words, and at times, not in the native tongue of the patient.
  3. “To understand basic health information” requires that the right health information be made applicable to the patient’s unique circumstances. If the patient identifies and relates to health information that is applicable to their lifestyle, then they can start to build a knowledge base and a longer-term understanding. This can be very challenging given that the first two steps above (Obtaining and Processing) must be completed successfully, then the patient must proactively translate what they’ve obtained and processed into knowledge that is applicable to their own circumstances.

Improving health literacy through personalized, adaptive learning

3 applications of the latest in learning science and artificial intelligence can improve health literacy at scale:

  1. A personalized learning platform can serve the right health information to any given patient, with confidence that the information is up-to-date, relevant, and accessible wherever the patient is.
  2. A personalized learning platform can help present the right health information in a format and language that is easily understood.
  3. A personalized learning platform can adapt content to address patients’ specific knowledge gaps, and over time enable the patient to build a robust understanding of how to live a healthier life.

Using a personalized learning platform like Sana enables knowledge mastery and reinforcement of topics using spaced repetition. Capturing relevant data from the patient’s learning progress also unlocks powerful analytics that can help healthcare professionals to understand what information is not easily processed by the patient, and to use this feedback to improve the educational content.

The platform can also provide insights into specific knowledge gaps, and allow the healthcare professional to systematically nudge the patient to continue addressing knowledge gaps and to reward them as they progress through their education. This encouragement and reward journey builds engaging learning experiences that translate into strong learning habits.

Use cases for improving health literacy with a personalized learning platform like Sana range from post-discharge patient care education (e.g. self-management, symptom awareness, nutrition, exercise, and medical follow-ups) to drug education (e.g. medication management and reconciliation, drug side-effects, proper dosage).

Delivering value to by improving knowledge to reduce costs

Personalized learning platforms can reduce the time for patients to acquire mastery of health information topics by 50% while ensuring knowledge is retained by up to 3x longer than traditional learning methods. These platforms can enable methods that are proven to improve health literacy, such as teach-back, jargon-free and slowed-down communication, and understandable written communication. This can translate into direct cost savings for patients, pharma, providers, and payers. For example, improving health literacy has been found to reduce hospital readmission rates, potentially driving significant savings given that readmission costs amount to nearly one-third of the United States’ total health expenditures. The Center for Medicare and Medicaid Services (CMS) found that an estimated 20% of patients are readmitted within 30 days after discharge and, of the $17.5 billion in Medicare spending on readmissions, $12 billion is potentially preventable.

Combining these benefits with the ability to deliver through a mobile experience, anytime anywhere, with multi-language support, actionable insights, and gamification to build strong learning habits, we can improve health literacy to drive better health outcomes and reduced costs.

About the author

Jon Lexa is the Director of Operations at Sana Labs, based in Stockholm, Sweden. Jon’s past experience includes working at Boston Consulting Group and a data science startup in NYC and London.

Jon studied Perceptual and Adaptive Learning and Causal Inferences as part of a BSc in Cognitive Science at University California Los Angeles. Jon also received an MBA from INSEAD.

If you want to learn more about how you can leverage personalized, adaptive learning in your organization, please send a note to


CDC Understanding Health Literacy

Literacy and Health Outcomes: A Systematic Review of the Literature

Reducing 30-Day Readmissions: Health Literacy Strategies