KenSci
KenSci Menu KenSci
KenSci Close KenSci

Financial analytics

Predict risk to improve care costs

Financial analytics
Predicting financial risk to reduce costs and minimize variance
Delivering ROI in 12 weeks

Predicting financial risk to reduce costs and minimize variance

Finance and Acturial teams in ACOs and Payers leverage KenSci to identify cost risks, medical fraud & abuse, and levers to optimize care delivery cost. Healthcare organization, use the KenSci solution to track key metrics and look out for early warnings that could impact fiscal performance. KenSci’s risk prediction platform for healthcare is engineered to ingest, transform and integrate disparate sources of healthcare data, including EHR, Claims, Admin/Finance, streaming and other sources. KenSci uses machine learning to recognize patterns in large volumes of data, helping health systems view the granular details of the patient’s history and predict future risks for optimal care.

Use cases we solve for

Claims Denial Prediction
Bundled Payments
Fraud, Waste and Abuse Prediction
Population Cost Prediction Solution
RX Analytics Solution
Claims Denial Prediction
Claims Denial Prediction
Bundled Payments
Fraud, Waste and Abuse Prediction
Population Cost Prediction Solution
RX Analytics Solution
Claims Denial Prediction

Claims Denial Prediction

The financial viability of a healthcare organization is dependent on medical claims submission and payment. This review process has become very complicated and adversarial. The post-adjudication claims appeal process requires the involvement of the clinical team, treating physicians may need to amplify on ...

+ read more

The financial viability of a healthcare organization is dependent on medical claims submission and payment. This review process has become very complicated and adversarial. The post-adjudication claims appeal process requires the involvement of the clinical team, treating physicians may need to amplify on clinical documentation and rationale in order get reimbursement approved. Denial of claims is commonplace.

 

The deficit

Data on medical expenditures or costs of treatment typically feature a strongly skewed distribution with a heavy right-hand tail, including a few intermediate spikes depending on underlying population. Leveraging existing large and varied claims, clinical, and survey data to identify claims that are likely to be protested, as well as collect feedback and lessons learned to take measures for process improvement, and improved data collection from 3rd party payers.

Predictive cost estimate

Predictive cost estimate

Identifies the patients’ structural characteristics that are influencing costs to obtain an estimate of expected costs for future years, maximizing reimbursement and simplifying patient collections.

Prediction and risk stratification

Prediction and risk stratification

Classifies new claims with predictions and risk stratification so measures can be taken to optimize resource utilization and billing process workflows.

Optimize resource utilization

Optimize resource utilization

Improves accountability in care, and optimizes resource utilization and workflows in the context of healthcare delivery.


End to end support

End to end support

Supports end-to-end revenue cycle management workflow and analytics, and assists providers to scale capabilities and improve margin performance.

Bundled Payments

Bundled Payments

Episode-based bundles of payment provide a single reimbursement to providers for each clinically defined episode of care. When bundling is used in an affordable care organization (ACO) model of delivery, there is potential to improve quality of care and reduce economic burden). Under a system of bundled ...

+ read more

Episode-based bundles of payment provide a single reimbursement to providers for each clinically defined episode of care. When bundling is used in an affordable care organization (ACO) model of delivery, there is potential to improve quality of care and reduce economic burden). Under a system of bundled payment, or episode-based payment, reimbursement for multiple providers is bundled into a single, comprehensive payment that covers all associated services. By structuring a payment around a patient’s total experience of care, bundled payments aim to support better care coordination and, ultimately, better outcomes for patients

 

The deficit

Reducing costs associated with this high-risk and high-cost patient population is necessary for the bundle to be profitable. Additionally, providers may have to assume financial risk when a patient receives subsequent care at another facility, such as a costly visit to the ER at another hospital. Providers need the ability to identify who is high risk to target resources towards those individual to avoid modifiable cost drivers such as readmissions and variation in post-acute care. The ability to target specific patients to tailor discharge plans and enhance care coordination is a vital component of cost containment

Data driven insights

Data driven insights

KenSci bundled-payment solutions will help healthcare organizations gain insight into the actual costs of care to price bundles accurately and appropriately be compensated for the total cost of care for disease or procedure.

Reduce overall costs

Reduce overall costs

Payment solutions bring to light the actions needed to reduce cost variation so that organizations can share in the savings of the bundle by proactively reducing costs to below benchmark pricing.

Identify high cost outliers

Identify high cost outliers

KenSci machine learning models enable identification of outliers that result in the highest concentration of costs, especially those propelled by irrational variation.

Target high risk patients

Target high risk patients

By targeting these high-risk patients and practices with effective prevention, the tools improve patient care while simultaneously ensuring the profitability of the bundle.

Fraud, Waste and Abuse Prediction

Fraud, Waste and Abuse Prediction

In the United States, roughly one-third of all healthcare expenses are caused by fraud, waste, and abuse (FWA). As of 2009, healthcare FWA accounted for 22% of healthcare waste, or up to $200 billion a year in fraudulent Medicare and other claims. In 2009, A staggering 70% of US Healthcare Expenses accounted ...

+ read more

In the United States, roughly one-third of all healthcare expenses are caused by fraud, waste, and abuse (FWA). As of 2009, healthcare FWA accounted for 22% of healthcare waste, or up to $200 billion a year in fraudulent Medicare and other claims. In 2009, A staggering 70% of US Healthcare Expenses accounted for Fraud, waste and abuse. 80% of healthcare FWA is committed by medical providers, 10% by consumers, and the remaining by others

(insurers themselves and their employees)

 

The deficit

Existing rules-based filters and data, and computational infrastructure are often inadequate to stay ahead of the FWA capability gap and business problem. The reliance on manual predictive approaches lead to operational constraints as with a limited set of known parameters, based on heuristic knowledge and not as robust as machine learning (ML) strategies. The lack of training data (or marked fraudulent claims) today complicates the application of supervised techniques for successful modeling

Stay ahead of the market

Stay ahead of the market

KenSci’s systems and infrastructures for data ingestion, extraction, and analysis tackle fraudulent claims, and have evolved to stay current with business expansion and FWA tactics.

ML based fraud detection

ML based fraud detection

KenSci has developed a comprehensive solution that engages hospital systems with a machine learning-based software to detect and help reduce FWA in their claims transaction system.

Extendable and expandable

Extendable and expandable

The solution can be extended for multiple payers, multiple geographies and add additional geographies. It analyzes a range of factors such as improper coding, phantom billing and unnecessary provision of care to determine the total predicted fraud as well as provide an average per member predicted cost.

Insights at granular levels

Insights at granular levels

It provides analysis at an individual patient level determined by the total claimed amount, number of admission, the total number of services and other relevant factors. 


Population Cost Prediction Solution

Population Cost Prediction Solution

Spending for health care services is highly concentrated among a small proportion of people with very high use. In 2009, 20% of all personal health care spending, or $275 billion, was on behalf of just 1 percent of the population. The 5% of the population with the highest spending was responsible for nearly ...

+ read more

Spending for health care services is highly concentrated among a small proportion of people with very high use. In 2009, 20% of all personal health care spending, or $275 billion, was on behalf of just 1 percent of the population. The 5% of the population with the highest spending was responsible for nearly half of all spending

 

The deficit

This emphasizes the need for population health cost prediction for quality improvement, cost containment and supporting value-based care. Much of the total cost of caring for a patient involves shared resources, such as physicians, staff, facilities, and equipment. The failure to prioritize value improvement in health care delivery has led to ill-advised and variable cost containment. Accurate prediction is necessary to identify the internal costs structures, how to price them, and determine whether they will be profitable for the provider to add value

Ties isolated data sources

Ties isolated data sources

Plumbs together isolated data stores and helps users understand and control costs using descriptive analytics.

Actionable insights delivered

Actionable insights delivered

Applies algorithms via the KenSci Machine Learning Platform to fully leverage the richness and latent patterns embedded within the claims data to derive accurate and actionable insights.

Learns from existing data

Learns from existing data

Learns from a pool of historical clinical data to predict each population Per Member Per Month cost by medical condition. It stratifies patients into risk categories for administrators and providers to identify who is at highest risk for increased length of stay and readmission so that providers can develop a precise plan of care.

Predicted future alerts

Predicted future alerts

Can alert an administrator and provider that a population will have an elevated risk of heart disease in the next three years, giving them the chance to prevent the onset of illness and apply disease risk management in a timely fashion.

RX Analytics Solution

RX Analytics Solution

For many hospitals and health systems, unwarranted variation in care can lead to suboptimal patient outcomes and increased unnecessary costs. This variation exists when there is a gap between best practice and the current practice. Variation among medical prescriptions is one primary source of care and cost ...

+ read more

For many hospitals and health systems, unwarranted variation in care can lead to suboptimal patient outcomes and increased unnecessary costs. This variation exists when there is a gap between best practice and the current practice. Variation among medical prescriptions is one primary source of care and cost disparity. The cost associated with prescription medication is one of the fastest growing elements of medical care. Analyses have shown that prescription drug prices have increased at double-digit rates which is more than other components of medical care.

 

The deficit

Prescriber behavior varies widely leading to irrational, wasteful, and potentially abusive medication allocation. Inappropriate and over-prescribing of medications increases cost and exposes patients to adverse reactions and drug interactions. For instance, Adverse Drug Events (ADEs) account for nearly 700,000 emergency department visits and 100,000 hospitalizations each year in the United States

Manage pricing better

Manage pricing better

By leveraging ML and analytics, healthcare organizations can utilize data to uncover insights related to pharmacy variation that will improve standardized prescription practices such as enabling the establish of a drug formulary, arrangement of wholesale pricing, and encourage the use of drug substitutions while reducing costs.

Reduce overall drug costs

Reduce overall drug costs

The Rx Analytics Solution is designed to address the issue of variation in cost, volume and prediction to significantly reduce costs and investigate drug price adjustments across patient episodes of care.

Data driven cost analysis

Data driven cost analysis

Analytics, powered by machine learning, result in actionable insights that help reduce potentially 25% or more of the global spend on prescription costs to the healthcare organization.

Improve organizational profitability

Improve organizational profitability

Rx cost savings will eventually result in savings for patients as well as increased overall revenue for the organization.

Stay one step ahead

  • Optimize PM/PM margins

    Optimize PM/PM margins

    Review predicted costs and PMPMs for future months using KenSci’s advanced ML models.

  • Predict population cost

    Predict population cost

    Predict population and costs by identifying patterns of clinical and cost outcomes.

  • Identify future high utilizers

    Identify future high utilizers

    Investigate care pathways that could be cost drivers and proactively help stay within budget.

Get in touch