Preventable Readmission Software Not Yet Adequate, Study Finds
April 8, 2014 in News
Automated classification software is not yet ready to replace manual reviews in identifying potentially preventable hospital readmissions, according to a study published in BMC Medical Informatics and Decision Making, Clinical Innovation Technology reports (Pedulli, Clinical Innovation Technology, 4/7).
For the study, researchers looked at 459 all-cause readmissions within 30 days at 18 hospitals operated by Kaiser Permanente Northern California (Slabodkin, Health Data Management, 4/7).
The manual review included:
- A chart review tool;
- Interviews with patients, families and treating health care providers; and
- A nurse and physician team evaluation of preventability on a five-point scale.
Researchers then analyzed the readmissions data with 3M’s Potentially Preventable Readmission software and compared the sensitivity and specificity of the two review methods (Hall, FierceHealthIT, 4/7).
The study found that:
- The manual review identified 47%, or 227, of readmissions as potentially preventable;
- PPR software identified 78%, or 358, of readmissions as potentially preventable; and
- The two methods agreed upon 56%, or 258, of the cases (Clinical Innovation Technology, 4/7).
The authors noted, “Substantial differences existed between manual review and automated classification methods … [that] may have occurred because PPR [software] uses a sole criterion to identify potential preventability: clinical relatedness to the index admission” (Health Data Management, 4/7).
According to FierceHealthIT, hospitals increasingly are seeking automated solutions to identify which patients need better care coordination (FierceHealthIT, 4/7).
However, the authors concluded, “Not enough concordance currently exists between methods” for hospitals to efficiently “use automated classification to replace manual review for quality improvement initiatives” (Clinical Innovation Technology, 4/7).