The Importance of Predictive Modeling in the Healthcare Industry

Prediction is a word that’s often used in scientific discourse and also in social science discourse. The process of prediction is a rigorous, often quantitative statement or forecasting that gives various scenarios of what will happen under specific conditions. Predictive modeling uses various statistical data to predict outcomes. Most often the event one wants to predict is in the future, but predictive modeling can be applied to any type of unknown event, regardless of when it occurred. Healthcare Predictive Modeling is good because knowledge can be transferred into action.

Healthcare predictive modeling is used in various areas of the healthcare industry. One of those areas is in hospital readmissions.

Academically speaking, predicting hospital readmissions is a very active topic. Thus far in 2013, 36 peer-reviewed journal articles have been published on the subject along with three additional review articles. Highlighting this rapidly growing interest are recent papers focused on simplified readmission scoring for elderly patients, the relationship between readmission and mortality rates, and a systematic review of tools for predicting severe adverse events. Prediction discussions associated with specific areas such as heart failure or within pediatric populations are also very active. In the financial year of 2012 IPPS final rule, CMS finalized the following policies with regard to the readmission measures under the Hospital Readmissions Reduction Program:

 

  • They define readmission as an admission to a subsection hospital within 30 days of a discharge from the same or another subsection (d) hospital

 

  • Adopted readmission measures for the applicable conditions of acute myocardial infarction (AMI), heart failure (HF), and pneumonia (PN)

 

  • Established a methodology to calculate the excess readmission ratio for each applicable condition, which is used, in part, to calculate the readmission payment adjustment.

 

  • A hospital’s excess readmission ratio is a measure of a hospital’s readmission performance compared to the national average for the hospital’s set of patients with that applicable condition.

 

  • Established a policy of using the risk adjustment methodology endorsed by the National Quality Forum (NQF) for the readmissions measures to calculate the excess readmission ratios, which includes an adjustment for factors that are clinically relevant including certain patient demographic characteristics, comorbidities, and patient frailty.

 

  • Established an applicable period of three years of discharge data and the use of a minimum of 25 cases to calculate a hospital’s excess readmission ratio for each applicable condition. 

 

For the healthcare industry, like other industries, predictors will always be more useful in the framework of a complete set of data, where the knowledge can be fully leveraged to action. Furthermore, the full clinical utility of prediction or risk stratification is only possible in a data-rich enterprise warehouse environment. But perhaps most importantly, these predictor-intervention sets can best be monitored and measured within that same data warehouse environment.  If the predictor is used standalone or housed elsewhere (siloed), this important evaluation step may not be possible.

 

Another way predictive modeling can be used in healthcare is employers providing healthcare benefits for employees can input characteristics of their workforce into a predictive analytic algorithm to obtain predictions of future medical costs. Predictions can be based on the company's own data or the company may work with insurance providers who also have their own databases in order to generate the prediction algorithms. Companies and hospitals, working with insurance providers, can synchronize databases and actuarial tables to build models and subsequent health plans. Employers might also use predictive analytics to determine which providers may give them the most effective products for their particular needs. Built into the models would be the specific business characteristics. For example, if it is discovered that the average employee visits a primary care physician six times a year, those metrics can be included in the model.