Is the patient the cure to AI healthcare ills?

A.I. Doctor

A.I. Doctor

Credit: Thinkstock

The expectations of big data and artificial intelligence disrupting the medical industry has been less then impressive to date. Why is that?

Machine learning and artificial intelligence (AI) works best on large volumes of data. One would think that with all its complexity and its mountainous volumes of data, the medical industry would be the perfect place for AI to be a disruptive force.

The problem isn’t that AI isn't acquiring enough data to be a disruptive force. The problem is that it’s not acquiring the right data to solve some of the most egregious cost increases in the history of health care industry.

In 2014, the U.S. Government Accountability Office reported $77.4 billion in improper payments of Medicare and Medicaid collectively. These payments, that were made in an incorrect amount or should not have been made at all, are contributing to excessive health care costs.

Where's the competition?

Which brings to light the real issue within the medical industry, that there is a silo of information not being shared between the providers and the payers. What incentives are there in place for each of these parties to work together? Competition is focused on the wrong things. Providers are maximizing the number of procedures, whether they are medically necessary or not. Likewise, payers are insisting on lower reimbursements with no consideration of the impact on the quality of care.

This issue is one of the main goals of President Obama’s Patient Protection and Affordable Care Act, to move from a fee-for-service model to a value based model. Providers under the fee-for-service model are paid based on service rendered with no real incentives to affect the outcome of the patient. The value based model addresses this issue by incentivizing providers to cure the patient more effectively through bonuses

The Boston Consulting Group in its white paper “Competing on Outcomes Winning Strategies for Value-Based Health Care” articulates the point that “transparency of patient results can align incentives so that payers, providers, suppliers, and patients all work towards the same goal.”

Need for standardized outcomes

Having transparency of patient outcomes is a monumental step in providing AI with the right data. The International Consortium for Health Outcomes Measurement (ICHOM) is a non-profit organization founded by the Boston Consulting Group, Michael Porter’s Institute for Strategy and Competitiveness at Harvard Business School, and the Karolinska Institutet with the purpose of transforming the health care systems worldwide by measuring and reporting patient outcomes in a standardized way.

Additionally, the U.S. Department of Health & Human Services Agency for Healthcare Research and Quality (AHRQ) has created an Outcome Measures Framework (OMF). The framework is to serve as a conceptual model for developing standard outcome measures for evaluating the safety, effectiveness, or quality of medical treatments.

Having comprehensive patient outcomes is critical for building A.I systems that provide value at all levels of the health care ecosystem. Pooling data outcomes at a national or international level will allow for identifying fraud, waste and abuse against such standards.

Capturing patient outcomes

Collecting patient-reported outcomes, however, can be challenging. Capturing certain outcomes directly after a procedure is relatively straightforward, but following a patient over time to track the evolution of his or her condition can be far more difficult.

Technology promises to help overcome this challenge. Software tools developed to track patients and their conditions over time are becoming more prevalent and more powerful. As an example, Apple just recently announced CareKit, an open source software framework enabling developers to build applications that help users track their medical conditions. These applications should provide the medical industry with the capability to validate the data through a verification process. This approach will lead to better data, and the ability to reduce the fraud, waste and abuse that now occurs.

Application developers should become ICHOM certified to assure data compliance with ICHOM Global Health Outcomes Benchmarking (GLOBE) program. The GLOBE program will create a central place where data, collected in accordance with the ICHOM standard sets, are securely compiled and stored. The GLOBE program will begin in Q2 of 2016.

AI solutions will benefit from having access to the best data and information. The greatest possible competitive advantages stem from having knowledge of customers that your competitors lack. Companies can’t be competitive if they can’t stay ahead of the disruptive forces that are changing customer experience, which is changing customer expectations.

This article is published as part of the IDG Contributor Network. Want to Join?

Drexel and announce Analytics 50 award winners
View Comments
Join the discussion
Be the first to comment on this article. Our Commenting Policies