Big Data Analytics Finding Gaps in Chronic Disease Management Care

Using Big Data Analytics for Risk Assessment

According to research by Marshall University, the success of chronic disease management and population health management is dependent upon a provider’s ability to identify high-risk patients. By implementing big data analytics, practitioners can identify illnesses and diseases in a timely manner, allowing for a quick treatment turnaround and reducing costs.

Currently, there are five chronic conditions that account for 75 percent of healthcare spending. These include cancer, cardiovascular disease, diabetes, obesity, and kidney disease.

The incorporation of big data analytics will both improve the quality of care and reduce the cost associated with chronic conditions. By identifying high-risk potentials in patients through big data analytics, medical professionals can engage in preventative care methods to hopefully minimalize the risk.

In order to reduce chronic disease occurrence, big data analytics can offer healthcare providers insight into high-risk patients, allowing the opportunity for providers to intervene even if patients are asymptomatic. Providers should have a coordinated prevention approach that promotes healthy behaviors, increases early detection and diagnosis, supports all demographics, and eliminates health disparities.

Artificial Intelligence Guides Treatment Process

Artificial intelligence can also assist in the predictive decision-making process regarding chronic conditions. In a study examining the effectiveness of big data analytics in care management of Type 2 Diabetes Mellitus (T2DM), researchers discovered massive amounts of data generated by treating the illness.

The different structured data categories included demographic data, semi-structured data such as reports, prescriptions, symptoms, diagnosis, etc. The unstructured data included audio, video, and other multimedia data. In order to have successful chronic disease management care, big data analytics can play a crucial role in the organization and quick decision-making process.

For big data analytics tools and artificial intelligence to detect potential gaps in chronic disease management care, they need to be trained on quality information. If an algorithm is trained on inaccurate data, providers cannot expect high-quality results. This mistake will lead to poor care delivery and potential health disparities.

Picking the Right AI, Analytics Tools to Avoid Bias

To ensure high quality and accurate data, algorithms need to avoid biases, something that happens when training the algorithm. Developers should make sure the algorithm is trained on data from diverse populations to ensure equal care access and the elimination of health disparities.

According to the Center of Applied AI at Chicago Booth’s playbook, there are four steps to eliminating algorithm bias. These steps include assessment of the algorithm, understanding of the target, updating or eliminating the algorithm if it is not efficient anymore, and continuing to monitor and conduct audits on the algorithm.

With artificial intelligence and big data analytics becoming more prominent in the healthcare industry, the World Health Organization (WHO) released a report discussing the use of AI and how to avoid creating health disparities.

“Like all new technology, artificial intelligence holds enormous potential for improving the health of millions of people around the world, but like all technology it can also be misused and cause harm,” Tedros Adhanom Ghebreyesus, PhD, MSc, WHO director-general, said in a press release. “This important new report provides a valuable guide for countries on how to maximize the benefits of AI, while minimizing its risks and avoiding its pitfalls.”

Along with the report, WHO put out its guidance for how to avoid health disparities in AI explained in six principles. The first of the principles was to protect human autonomy. This means that humans should remain in control of the healthcare system as well as medical decisions. This also includes protection of privacy and data.

Secondly, artificial intelligence should promote human well-being, safety, and the public interest. Those designing the AI technology should satisfy safety requirements and ensure technology is performing accurately and efficiently. Additionally, measures of quality control and improvements must be available and in use.

Continuing, WHO highlights the importance of ensuring transparency, explainability, and intelligibility. This means publishing and documenting information regarding the design and deployment of AI technology. Information should be easily accessible and promote debate for how artificial intelligence technology should be used in healthcare.

Next, AI technology and big data analytics should foster responsibility and accountability. While the AI machines are performing the tasks, it is the responsibility of the stakeholder to make sure the AI technology is being used correctly and in the appropriate conditions.

According to WHO, the next principle to follow is ensuring inclusiveness and equity. Artificial intelligence technologies are to be trained on a wide variety of demographic data and to promote equitable access and use. Types of information received in demographic data could include age, sex, gender, income, race, ethnicity, and sexual orientation.

Lastly, WHO’s final principle states that providers should be promoting AI that is responsive and sustainable. This includes ensuring transparency and minimizing an environmental impact. 

“Designers, developers and users should continuously and transparently assess AI applications during actual use to determine whether AI responds adequately and appropriately to expectations and requirements,” WHO continued. “Governments and companies should address anticipated disruptions in the workplace, including training for health-care workers to adapt to the use of AI systems, and potential job losses due to use of automated systems.”

In order for big data analytics to detect gaps in chronic disease management care, algorithms and artificial intelligence need to be trained properly to ensure the systems are preventing health disparities rather than causing them.