Data Analytics and Population Health – Keys to the Transition to Payment for Quality and Cost Effectiveness
The following article was published in the June 2016 edition of South Florida Hospital News and Healthcare Report.
In 1854, John Snow helped solve a public health problem by concluding that the cause of cholera in London was primarily from people drinking water from the pump on Broad Street which had the bacterium that caused cholera. He interviewed people, drew a map with black marks for each death and concluded that the center of the affected area was the Broad Street pump. The brewery workers across the street from the pump were by and large fine as they drank the liquor provided by their employer and generally did not drink water from the pump.
What does John Snow have to do with Population Health, particularly today? Well, his finding of epidemiology as a field of modern sciences used data analytics. Dr. Snow did not know about Big Data, Data Analytics, Precision or Personalized Medicine, and Predictive Analytics. He did not have a personal computer, a laptop, or even a smart phone. He painstakingly interviewed people, drew a map, and tested his theories.
Much of Dr. Snow’s work is relevant to the Population Health of today, Data Analytics, and Predictive Analytics. Dr. Snow was concerned with the health of the population. He used a form of data analytics to determine the cause of cholera for the individuals who contracted it. He was able to predict who would get cholera.
Although Population Health today may be much more sophisticated because we have access to much more data and we have sophisticated computing power, it has many roots in Dr. Snow’s work and Public Health. Population Health is moving to center stage today as we see the shift from fee-for-service medicine to payment for quality and cost-effectiveness.
Entities need to understand their populations by studying the data about their health. They should focus on patients and the management of their care, also on preventive care. They should use predictive analytics to determine which interventions should be implemented to reduce the chances of an individual getting a particular malady or its severity. They should employ precision and personalized medicine techniques to focus on the health of specific individuals.
There needs to be a focus on pharmacy, comparative data sets for drugs administered, types of persons and dosages, medication adherence, severity of malady, co-morbidities, varying types of drugs an individual might be taking. Many of these concepts are interrelated and can make it quite difficult to develop a personalized plan for an individual.
Population Health today can still involve qualitative research as in the days of Dr. Snow where extensive interviews were conducted. However, today, we have electronic health records (EHRs) and personal health records (PHRs), coupled with clinical decision support systems (CDSS) and computerized physician order entry (CPOE) which make it possible to have significant amounts of data about a patient available and to aggregate the data from numerous individuals. Thus, care of a patient can be better personalized.
Technology may make it much easier to sift through the data we have, but there is much more data to consider and ensuring that it is clean and comparable is no easy task. Data Analytics and the concepts of Population Health will be key to this transition to payment for quality and cost-effectiveness.