My name is Kyle Rybarczyk, and I recently joined Forrester as an analyst covering data analytics, AI/ML, enterprise health clouds, electronic health records, and the digital front door in the healthcare sector. I have a passion for healthcare, the ever-changing digital world, and how the two will combine to streamline workflows and improve patient outcomes while making healthcare more accessible and affordable for all. After all, as Gandhi said, “It is health that is real wealth and not pieces of gold and silver.”
Prior to joining the healthcare team at Forrester, I served 10 years in the US Army’s Medical Service Corps, where I led and managed healthcare data analysis, biomechanical research, medical logistics, and medical administration in dynamic clinical environments.
The Data Dilemma
Over the past few years, technology has moved forward at a sprint catalyzed by the coronavirus pandemic. The healthcare sector is not immune to the complexity and challenges that come with the rapidly changing landscape. With over $4.1 trillion spent on healthcare annually in the US alone, organizations will need to move quickly to leverage the digital tools available to increase profits and improve patient health. Unfortunately, there are numerous hurdles that must be scaled in order to achieve workflow efficiency and data-driven healthcare that even the $36 billion already spent by the federal government cannot solve.
Healthcare organizations (HCO) have been attempting to leverage massive quantities of data to their advantage for years. The problem: The current algorithms are ineffective, the data is siloed, missing, or inefficient, and there is significant bias. The true promise of AI doesn’t stand a chance in this arduous situation. As the saying goes, “Garbage in, garbage out.” The repercussions in this industry result in incorrect or missed diagnoses and a lack of patient-focused care that can lead to morbidity or even death. In fact, the average lifespan in the US has fallen since 2019 by almost a full year. The impact is even larger for minority groups in the US, including the African American population that experienced a drop in life expectancy of three years.
Make The Data Work For You
Front and center is the healthcare interoperability imperative. Healthcare and life sciences organizations need to have a robust data team at their disposal. Although healthcare data has its own nuances and quirks, data analysts and scientists should be recruited from the financial or business intelligence sectors to help kick-start a data evolution. With a proficient team, the following challenges can be tackled in earnest:
- Centralize your scattered data. About 30% of the world’s data generation is from healthcare — that’s more than any other industry. By the year 2025, humans will have approximately 5,000 digital device interactions per day. That is a magnanimous amount of data, and it will need a remarkable amount of effort to put it to effective use. A solution that helps ameliorate the fragmentation of data is the cloud. Moving to a cloud-based storage and computing system can help securely collect, store, and aggregate data — whether in the form of electronic medical records (EMR) or unstructured data — and leverage it to ultimately reduce errors and drive interoperability and provider collaboration. The flexibility, scalability, and cost-effectiveness of the cloud is the enabler of population health management and precision medicine.
- Achieve the Quadruple Aim with AI and ML. A healthcare data team should be able to set up ML models to extract meaning from all of the data. Natural language processing, pattern detection, and deep learning will be a business necessity. Leaders within the HCO should start by learning as much as possible about AI and ML so they can leverage it to the best of their organization’s ability. Some of the giants in the tech industry are already making software to assist in an organization’s ML. For example, leveraging AI and ML has been proven to detect heart arrhythmias from electrocardiograms better than a trained expert with a 95% success rate. This technology also allows for process automation of normal tasks such as updating patient records or medical billing with reduced error. All these benefits increase efficiency and reduce workload on clinicians, which in turn reduces burnout.