Sometime soon, Greg Ahern, the director of corporate responsibility for data monitoring and analytics at Catholic Health Initiatives (CHI) in Englewood, Colorado, will get an email notification when claims for MS-DRGs with a secondary diagnosis code of Kwashiorkor are submitted by any of its 70 hospitals. Because Kwashiorkor—severe protein-calorie malnutrition—is a target of Medicare auditors, who consider it a third-world condition, CHI flags “any claims greater than zero for Kwashiorkor” and automatically puts them through DRG validation.
The email notifications will allow Ahern to act on Kwashiorkor faster. That saves the step of going into the system and requesting claims for review. Assuming Kwashiorkor was incorrectly assigned—not all hospitals agree that Americans are immune to this type of malnutrition—the MS-DRG will be reassigned. Other risk areas—diagnoses, DRGs and targets from the Program for Evaluating Payment Patterns Electronic Report (PEPPER), such as sepsis with a short length of stay and discharge disposition of home, could trigger email notifications when they cross a certain threshold. That would subject them to validation of the coding and diagnosis.
The notifications are part of the risk-based data analytics program at CHI, which recently merged with Dignity Health to form CommonSpirit Health. “We are doing risk-based data mining with formulas containing specific criteria at the code or DRG level and now progressing to thresholds with alerts—not randomly picking cases for review,” Ahern explains. “We set up our program to be an internal” recovery audit contractor (RAC).
The data mining is performed on both the hospital and physician sides. Although the method differs somewhat, the goal is the same: to detect and correct errors. “We want to capture them first, before the government does,” says Ahern, who used to work for a RAC. “We want to rebill or repay as necessary, educate, implement [corrections] and monitor so these things don’t happen again.”
