Regional Business Director, Elsevier/MEDai
After graduating with a Master’s in Health Administration, I followed a career path into the hospital setting and was fortunate to build a telehealth /telemedicine program, from the ground up, for a large IDN in the Midwest. This foray into Health Information Technology (HIT) was a great experience learning how technology applications can improve access to specialized healthcare and substantially improve the quality of life for patients in rural areas.
The technology of ‘Informatics’, however, was very new to me. A basic understanding of the field and recognition of the US health delivery systems need for more targeted, tailored healthcare led me on a new career path with an analytics vendor. What was surprising, initially, is the lack of understanding of this field and the ambiguous terms and definitions. Using data to help make better strategic decisions has generally been understood as a good approach, however most adopted decision support techniques simply mine and re-organize existing data elements to make it more consumable.
The difference between data queries (i.e. pulling your own reports) versus having clinically relevant data automatically pushed to decision makers is what is going to make a difference in the future. Artificial intelligence and advanced predictive models have been developed based on their ability to map patterns between relevant data points. Vast data repositories with millions of patient records provide a robust foundation for the development of such models and will only continue to grow during the adoption of electronic health records. The intelligence of the models is such that is allows unique characteristics of data sets (or populations) to automatically adjust the models as required to produce the most accurate outputs. The science and math, however, do not negate the need to integrate clinical expertise into the data mapping process.
Many in the clinical community are going to be required to adopt a new mindset around the use of data concerning their patients. The ability of computers and their vast processing ability should not be ignored but rather used to augment the physician’s expertise and wisdom based on his/her past experiences. This science can then allow clinicians to maximize their time with patients and focus on the individual characteristics of each patient that may not show up in the numbers…
This type of predictive science has been slow to come to healthcare even though it’s been used in other markets for several of years. The perverse incentive structure in healthcare today is partially to blame and has become the focus of several recent government incentives and legislation. Physicians, hospitals, and other providers will now be accountable for the care they provide and the outcomes of their patients, and will no longer be paid based on the amount of procedures. The consumer is also becoming accountable for their own physical health and for selecting the highest quality providers to provide their care. Accurate, credible data is the key to empowering these types of decisions to be made.
The next decade promises to hold unfathomable advances in the field of informatics as it relates to healthcare. Arguably the most deserving field for such technology applications will begin to catch up to other more retail oriented solutions. As we learn more about our own health, and the data that supports our insurance ratings, we can begin to play a more active role in improving our own health status, and ultimately the quality of the care that we receive.