Resources - Knowledge Management

Analytics in Depth

Prescriptive Analytics

Prescriptive Analytics has been hailed as the final piece of the Business Analytics (BA) puzzle. It starts with descriptive analytics - describing what HAS happened and why; next is predictive analytics - forecasting what MIGHT happen; Prescriptive Analytics aims to offer insight into what ACTIONS to take next, suggesting the best course of action among available choices.

If your organization has yet to formulate a plan to utilize Prescriptive Analytics in your business intelligence systems, don't feel left out - yet. According to a recent Gartner study, only 3% of organizations are using it currently. However, that percentage is expected to grow rapidly. Organizations taking advantage of the impacts that Prescriptive Analytics offers will enjoy a distinct advantage in their industry.

Building a Prescriptive Analytics Model

Prescriptive Analytics is very process-intensive. Historical data, business rule algorithms, computational modeling techniques, constraints and variables, and machine learning algorithms are combined to prescribe optimum business outcomes. For best results, a combination of structured and unstructured data sets should be used.

Advancements in computing power, along with the development of our sophisticated mathematical algorithms, make Prescriptive Analytics available today. Some of the methods used in Prescriptive Analytics include optimization, simulation, game theory and decision-analysis techniques.

In situations where decisions need to be based on too many conditions - variables, options, data points and constraints - where the human mind is incapable of evaluating it all without the use of technology - Prescriptive Analytics is the answer.

Where can Prescriptive Analytics be used?

Prescriptive Analytics are ideal for situations where experimentation in business operations would be overly risky, expensive, or time consuming. Analytical models and simulations are run using complex, randomized variables to discover and optimize the range of potential outcomes.

Prescriptive Analytics can also pinpoint decision options that take advantage of future opportunities, or limit future risks, illustrating the implications of each available decision. Prescriptive Analytics can continually take in new data to revise predictions and prescriptions automatically, thereby improving the accuracy of predictions, and prescribing continually more insightful and impactful decision options.

Prescriptive Analytics Examples

An example of a completely automated Prescriptive Analytics system is Google's self-driving car. It has elaborate data input systems (to "see" streets, other cars, pedestrians, etc), built-in process controls (hardware, software, firmware and more) to prescribe the action (moving, turning, and stopping).

In more conventional business applications, typical examples of Prescriptive Analytics applications include inventory management, production planning, operational resource allocation, supply chain optimization, utility management, transportation and distribution planning, marketing mix optimization, sales lead assignment, pricing and financial planning.

Healthcare Providers use of Prescriptive Analytics

Healthcare is moving rapidly to incorporate measures of value into payment models, with more than two-thirds of payments expected to be based on value measurements in five years, up from just one third today. That’s one of the findings from The 2014 State of Value-Based Reimbursement, an independent research study of 464 payers and providers conducted by ORC International and released at the AHIP Institute 2014 Conference in Seattle. The report was sponsored by McKesson Health. The need for analysis such as Prescriptive Analytics is now more than ever because providers will be able to enhance their clinical care delivery to the population they manage with analytics, and get more reimbursement from value-based care.

Pharmaceutical companies can expedite their drug development by identifying optimum patient cohorts - analytics can predict the time and expense of choosing one patient cohort in a specific country vs. another, by identifying patients who are anticipated to be compliant and won't drop out of the trial due to a variety of complications.

Are there any limitations?

Prescriptive Analytics is not infallible. It is susceptible to the same distortions that can defeat descriptive and predictive analytics efforts, including limitations of the data sets, and unforeseen external forces. The effectiveness or impact of Prescriptive Analytics is also dependent on how well the decision model captures the impact of those decisions being analyzed. The biggest limitation is not in analysis but possibility of “lost in translation”. Hence, we provide data exploration methods that yield detailed, specific, actionable information that is specific to YOUR Company’s business needs, in a method of communication that is easy for you to understand.

How can FluentAnalytics benefit you?

Provider Benefits: Given the reduced payments to hospitals by healthcare law since 2009, Supply Chain Analytics is THE need of those providers and nurses. But the daily usage of bandages and medicines depend upon hospital bed numbers. FluentAnalytics’ implementation of Prescriptive Analytics is big here! We can also benefit providers in their capacity planning. FluentAnalytics can utilize operational and usage data, in combination with external data - economic data, and trends in population demographics and population health - to more effectively plan for your future capital investments. Now that the emphasis is on outpatient treatments, hospitals are looking into what parts of outpatient can be included, where reimbursements are strong and penalties for return visits are not as severe. Should they include radiology also? What will be the patient flow then? Such questions are answered by our expertise in Prescriptive Analytics.

Member Care Benefits: What is the right intervention, for the right member, in the right venue, at the right time?  Interventions are implemented systematically and strategically, focusing on diagnoses that impact on risk score accuracy and member care. So what are the diagnosis identifications that need to be made for patients who have taken appointment for hospital stay? FluentAnalytics has the answer. Our Prescriptive Analytics facilitates significant financial value capture and gap closure by coordinating the most appropriate interventions for identified members throughout the year. In addition, we empower providers in multiple venues with point-of-care decision support via interoperable and patient-specific analytics embedded in the provider's EHR to efficiently and conveniently resolve gaps.

Satisfaction Score Benefits: Bundled payments are reality in Healthcare; several procedures previously reimbursed individually are now being bundled as one service. If you’re Johns Hopkins Hospital, then you’re recognized for being exceptional in ENT, or Cleveland Clinic for complex kidney diseases. The patient base is available to them based upon their brand value. So, how does a local hospital in Anaheim or Sacramento create a profitable patient base in the era of bundled payments? The answer is in patient satisfaction scores; patients that experience exceptional care will spread the word by various channels of communications (social media).

Unfortunately, the healthcare industry too often looks at patient satisfaction survey results after the effect – reviewing poor scores from disgruntled patients. What if, prior to a patient’s care and subsequent survey, an analysis of which services and combinations thereof will most likely result in a favorable patient score? FluentAnalytics can create such “what if” situations, mathematically modeled through Prescriptive Analytics. It involves the optimization of multiple variables (e.g. patient mix, treatment plans, resources, physical space, payment structure, in-house vs. affiliated, etc.) and trade-offs (e.g. cost vs. quality). This analysis predicts the possible consequences based on different choices of action, and recommends the best course of action for any pre-specified outcome. FluentAnalytics can help you earn positive patient survey scores resulting in increased brand value for value-based disease management.

According to Gartner, more than 30% of analytics projects by 2015 will provide insights based on structured and unstructured data. The need for prescriptive tools to recommend specific courses of action and show the likely outcome of each decision is becoming increasingly critical. FluentAnalytics is providing the next business analytics evolution through Prescriptive Analytics, an automated system that combines big data, business rules, mathematical models and machine learning to deliver “what if” scenarios in a timely fashion.