Machine learning: Use data to develop outcomes

A. Artificial intelligence (AI) and machine learning have been integrated in our daily lives from social media interactions, retail shopping and now healthcare. Unlike the retail industry, where ads are triggered based on a web user's interest, the healthcare industry uses predictive analysis to learn more about patients and improve their health outcomes, which in turn can lead to cost savings.

Machine learning platforms help identify potential issues or gaps in care. By doing this, we can then alert clinicians so they can develop a post-acute care plan that includes home health care that may prevent rehospitalization. For example, we can alert a healthcare provider of obstacles that could prevent a patient from following post-surgery instructions. Those may include non-clinical obstacles such as lack of access to proper transportation to keep a doctor's follow-up appointment, or even getting groceries to follow a recommended diet. With this information, clinicians can close gaps in care and develop a customized plan that will result in a better, and more cost-effective, patient outcome.

Additionally, this type of technology can also match a patient with the facility and health care provider that will best fit the patient's needs. For example, if a patient requires wound care, rehabilitation, or a facility that has high-quality outcomes in a specific medical specialty (i.e. cardiology, neurology etc.); the database will be able to suggest a provider who has been successful in caring for patients in similar circumstances. By properly matching a patient with the right provider at the start of care, we can improve outcomes and provide a cost-saving solution that avoids hospital readmission.

Steve Wogen is chief growth officer for CareCentrix. Reach him at Stephen.wogen@carecentrix.com.