Hythiam (HYTM) Announces Publication Emphasizing that Predictive Models Can Be Used for Substance Dependence Solutions
Hythiam, Inc. (NASDAQ: HYTM) announced today the publication of a paper titled, "Application of Multivariate Probabilistic (Bayesian) Networks to Substance use Disorder Risk Stratification and Cost Estimation," Weinstein L., Radano T., et. al., Perspectives in Health Information Management, Volume 6, Fall 2009. The predictive modeling work that forms the basis for the publication was sponsored by Hythiam to develop advanced tools for use in its Catasys program. The paper, co-authored by Lawrence Weinstein, MD and Hythiam's Senior Vice President of Medical Affairs, features the creation of predictive models that draw on historic healthcare claims and diagnostic data to estimate the expected future costs of a population with a substance dependence diagnosis, and that can stratify the population based on the expected benefit of intervention. Thus, the predictive models can be utilized as the basis for Hythiam to develop new and innovative approaches for substance dependence solutions that result in improved member care and reduced costs for payors.
The predictive models and algorithms were developed in partnership by Hythiam and a leading data mining and analysis company, using its proprietary technology on composite commercial health plan data obtained from Thomson Reuters. The models identified the incidence, prevalence and cost of substance dependence in the population over a three year period. The models also utilized sophisticated relational and stratification data techniques to identify differential substance dependent subpopulations who were most likely to benefit from intervention on a validated prospective basis. A detailed analysis of cost differentials illustrated that intervention with specific combinations of members may lead to the greatest benefit for members and the most savings for payors.
Substance dependence in payor member populations has a significant impact on overall healthcare costs. However, there are a lack of effective tools to help identify, stratify and assess member populations most likely to benefit from intervention. The existing models highlighted in the publication demonstrate that member populations can be stratified into cost buckets, and further segmented by clinical patterns to help determine which members will likely benefit most from intervention. The predictive modeling tool discussed in this paper will provide Hythiam with the basis for segmenting members for intervention to maximize clinical outcomes and savings. Moreover, this work forms the basis to develop models that are also capable of identifying those members with a substance dependence diagnosis who will likely become high utilizers of healthcare in the future. Efficient, early intervention for certain substance dependent members can potentially prevent these members from becoming high cost, and result in even greater savings for payors.
"We are pleased the paper has been published, and it demonstrates that Hythiam continues to pioneer and seek innovative solutions for substance dependence," said Dr. Lawrence Weinstein. "The results from the modeling will be utilized to help our clients achieve efficient intervention with this population and demonstrate the need for more dynamic screening mechanisms. The models we designed are capable of identifying high cost substance populations that may be most appropriate for intervention, but we will continue to work toward creating interventions for substance dependent members prior to their developing and suffering from difficult diseases that also lead to significant health care costs."
The predictive models and algorithms were developed in partnership by Hythiam and a leading data mining and analysis company, using its proprietary technology on composite commercial health plan data obtained from Thomson Reuters. The models identified the incidence, prevalence and cost of substance dependence in the population over a three year period. The models also utilized sophisticated relational and stratification data techniques to identify differential substance dependent subpopulations who were most likely to benefit from intervention on a validated prospective basis. A detailed analysis of cost differentials illustrated that intervention with specific combinations of members may lead to the greatest benefit for members and the most savings for payors.
Substance dependence in payor member populations has a significant impact on overall healthcare costs. However, there are a lack of effective tools to help identify, stratify and assess member populations most likely to benefit from intervention. The existing models highlighted in the publication demonstrate that member populations can be stratified into cost buckets, and further segmented by clinical patterns to help determine which members will likely benefit most from intervention. The predictive modeling tool discussed in this paper will provide Hythiam with the basis for segmenting members for intervention to maximize clinical outcomes and savings. Moreover, this work forms the basis to develop models that are also capable of identifying those members with a substance dependence diagnosis who will likely become high utilizers of healthcare in the future. Efficient, early intervention for certain substance dependent members can potentially prevent these members from becoming high cost, and result in even greater savings for payors.
"We are pleased the paper has been published, and it demonstrates that Hythiam continues to pioneer and seek innovative solutions for substance dependence," said Dr. Lawrence Weinstein. "The results from the modeling will be utilized to help our clients achieve efficient intervention with this population and demonstrate the need for more dynamic screening mechanisms. The models we designed are capable of identifying high cost substance populations that may be most appropriate for intervention, but we will continue to work toward creating interventions for substance dependent members prior to their developing and suffering from difficult diseases that also lead to significant health care costs."
You May Also Be Interested In
- Auxilium (AUXL) Files Suit Against Watson (WPI) Over Testim 1% Gel
- Delphi Automotive (DLPH) in Talks with FCI to Buy Motorized Vehicles Unit
- Apple (AAPL) Amends RSU Award Agreements in Lieu of Upcoming Dividend; CEO Cook Defers
Create E-mail Alert Related Categories
Corporate NewsSign up for StreetInsider Free!
Receive full access to all new and archived articles, unlimited portfolio tracking, e-mail alerts, custom newswires and RSS feeds - and more!
