Improving Outcomes with Medication Adherence

The following is a guest article by Erin Benson, senior director, market planning, LexisNexis Risk Solutions, Health Care.

Healthcare stakeholders agree: the solution to medication non-adherence is a “$100 billion opportunity.” About three in five of us have a daily recommended regimen, so taking medications as prescribed is vital for successful health outcomes.

The burden of chronic illness, in particular, becomes much heavier as treatments lose their effectiveness due to non-adherence. CMS’ 5-Star Quality Rating System makes it clear that we should do better to help patients manage diseases like hypertension, diabetes, and high cholesterol with improved adherence strategies. However, our support solutions rely almost entirely on clinical protocols. Care managers, providers and data professionals should work together to understand reasons behind regimen non-compliance and take proactive steps to help patients overcome adherence obstacles. Insights into social determinants provide an innovative means to drive medication adherence for optimal health outcomes, and, for healthcare organizations to earn pay-for-performance bonuses as a result.

Predictive and proactive

Research has shown that medical care alone has a very limited effect on overall population health. It’s time we branch out beyond the one-dimensional approach. Clinical information that healthcare professionals collect can be paired with social determinants of health (SDOH) data to determine the best care plan possible. Health plans and providers can identify factors that get in the way of medication adherence by analyzing environmental, economic and community attributes of each patient.

This is possible with technology solutions automatically generating insights to identify patients’ degrees of need. SDOH data can generate medication non-adherence risk levels for patients so care managers and providers can better allocate their resources. For example, a socioeconomic health score can reveal likelihood of non-adherence as well as what determinants drive this prediction. To cite a few examples, with that information the provider can tell that the patient may benefit from additional education, financial assistance alternatives, or assistance with transportation to follow the regimen more closely. By connecting patients with resources to address socioeconomic needs, health plans and providers are treating the holistic patient, not just the clinical patient. This is an important distinction when it comes to medication adherence. Together, these components tell the story of the patient condition, and how patient needs can be met within the greater care management ecosystem.

Predictive modeling frameworks can utilize hundreds of socioeconomic attributes to predict how likely an individual is to be adherent to their medications and which social determinants are likely driving their need for additional care. While this can and does allow for better communication between the patient and provider during a clinical visit, intervention does not fall on the physician alone. Discharge planners, social workers, health coaches, health plan care managers, pharmacists, or patient experience officers use non-clinical data to identify gaps for better patient care, building bridges to public health and community resources.

Since this collaborative and data-driven approach to care management is a relatively new strategy, stakeholders can read eHealth Initiative’s “The Guiding Principles for Ethical Use of Social Determinants of Health Data.” With this as a guide, care teams can work together to eliminate barriers to achieving medication adherence and reduce access challenges for in-need individuals.

Insights and understanding

Using social determinant insights to drive medication adherence isn’t an easy task. Manually combing through patient socioeconomic data can be challenging since not all data is created equal: zip codes are useful, but not holistic in providing a picture of the individual needed to make predictions about health outcomes. Healthcare organizations looking to evaluate SDOH data to use in predictive models should seek out non-clinical data sources that are current, comprehensive, and longitudinal. For ease of use, social determinant data should be easily standardized and integrated, regularly updated, and consistently linked to individual patients so care managers and providers can access actionable health insights.

It’s also important to ensure SDOH attributes have been clinically validated against actual health outcomes for their predictive power. For example, LexisNexis Risk Solutions has conducted the SDOH data analysis and validated the following insights:

  • College attendance is positively correlated with medication adherence.
  • People with relatives or associates who live within 25 miles are more likely to be adherent.
  • Those individuals who are predicted to be “most adherent” are twice as likely to follow up with office visits within two weeks of an acute event; a practice which is known to reduce readmissions.
  • In the 12-month period analyzed there was a three-fold increase in Emergency Room (ER) visits for patients with the lowest predicted medication adherence score.

By noting the top clinically validated socioeconomic risk factors that drive a patient’s non-adherence risk, care managers and providers can decide on the best course of action for implementing a successful care management plan. As healthcare continues to evolve from a volume-based to value-based service, understanding patient needs is a top priority. To thrive in this environment, healthcare organizations need to deliver the right intervention at the right time to the right patient. SDOH data will help them do just that.