TruRisk, LLC

Website Links

Description

TruRisk is a limited liability company founded in early 1998 by Greg Binns, Ph.D. and Mark S. Blumberg, M.D. Their professional collaboration on risk adjusted studies using large medical claims data bases had begun a decade earlier. Today, TruRisk is fortunate to have recruited employees that each have 20 years experience working with Binns and similar databases. The major objective of TruRisk has been to develop and apply risk forecast models to improve premium rate setting for health insurers offering: -Life There is a growing realization that forecasts of health care risk can be improved by using computerized information from claims and enrollment files. However each of the most widely known claims based forecast models was developed to provide standardized costs for care as a basis for paying providers. Thus the DRGs, which were developed to reimburse providers for hospital care costs, served as a precedent for almost all subsequent health care cost risk models. These models understandably omitted all costs and most services as predictor variables since their inclusion would result in paying higher reimbursement to more costly or care- intensive providers. We recognized early that dollar costs of past claims are the “experience” used by health insurers and that this valuable information should not be omitted in models intended to forecast future costs as a basis for setting premiums. TruRisk also believes that models developed specifically for a given client's book of business will be superior to general purpose (off-the-shelf) models. Clients differ in their premium pricing cycle, their databases and benefit packages as well as in the health status of their enrollees. While developing a forecast model customized for a given client may initially require a little more time and effort than simply using a model off-the-shelf, there are number of features of the TruRisk approach which facilitate this modeling. Thus we have a large library of precoded candidate predictor variables which include groups of CPTs, ICD-9 Diagnoses and prescription drugs. They also include detailed demographic recodes which go considerably beyond age and sex. The selection of variables and their weights are entirely objective and fact based, thus avoiding the use of physician and other opinions. We have found conventional wisdom about risk factors to be frequently in error ( e.g., some undiagnosed symptoms may presage greater future costs than a related explicit diagnosis). Of course once a customized model has been developed its application in successive years is no more time consuming than using one off-the-shelf. We considered that our set of systematic procedures for developing risk models customized to a client

Products & services

Similar companies nearby