top of page

PreVista Features
Click this icon for a list of key pages >

Features

PreVista is fundamentally different from standard 'risk management'and 'predictive modeling' products, which have very poor accuracy (typically R2 = ~.3 or 30%). Using a mix of artificial intelligence technologies and a new analytic approach, PreVista improves accuracy by 200% to 300% on average, and adds the ability to tune forecasts for specific members, groups, conditions and customer needs.

This PreVista report shows predicted healthcare costs with a graph that highlights outliers -- and the ability to search for and drill down on a specific member.  These results are available as data that you can load into your data warehouse or application.

This PreVista report shows a different type of analysis:  identification of frequent prescribers of addictive drugs linked to members who purchased unusual quantities of the these drugs. PreVista uncovered an underground network of pill mills and addicts, previously unknown to the health plan and the state.

  • Significant improvements in accuracy: PreVista R2 scores typically range from .7 (70%) to .98 (98%), even with the most complex targets and data.

  • Difficult goals:  PreVista can accurately predict very complex conditions and goals.  What challenging risk would you like to predict?

  • All of your data:  PreVista is not limited to health care claims, which are often 60+ days old.  You can include new EMR, HIS, HIE, lab results and 'Big' data sources, etc. This data is often more current, to support near real-time prediction and intervention.

  • All of your member groups:   You can predict risk, cost, readmissions and other factors for each age range, gender, plan group, geographic region, etc.  This also improves predictive accuracy.

  • Individual members:  With 70% to 90% accuracy, we can predict results for each of your members, instead of low-quality 'catch all' forecasts for an entire population.  

  • Complex conditions:  PreVista can solve the toughest predictive problems with the most complex data.  We can accurately predict risk / cost for people with diabetes, for example, in ages ranging of 2 to 98, with the typical host of co-morbid conditions, and a mix of standard claims plus new CM/DM data.  What is your most difficult predictive challenge?

  • Every chronic condition:  You can focus on each chronic condition in your membership -- to target interventions.  This ability to focus also improves accuracy.

  • Rapid results:  PreVista can deliver new forecasts in near real-time -- every time data is refreshed for a patient or group, e.g. based on updates from your EMR system.  There is no time-consuming 'batch' process. You can respond quickly.

  • All of your goals:  PreVista can identify any pattern in your data.  The standard examples noted above just scratch the surface.  If the answer is hidden in your data, PreVista can find it.​

Prediction Predictive modeling forecast health risk redmission EMR HIS HIE claims reports  acquisition custom accuracy 3M™ McKesson™
Ingenix™ Optum™ Milliman™ MedAI™ ISO™ DxCG™ DRG™ CMS ACG™ Verisk™ pharmacy fraud abuse actuary payer provider STARS healthcare

bottom of page