As Physician Advisors, we should work closely with the Clinical Documentation Improvement (CDI) staff in our facilities. CDI staff work is frequently overseen by finance, as the traditional goals of CDI programs have been financial, i.e., mid-cycle revenue optimization. While this is the “meat and potatoes” of their work, there are other aspects of the CDI job that are often overlooked.
CDI documentation frequently supports not only coding validation denials, but medical necessity audits and denials as well. An appropriately answered query can go a long way in supporting the medical necessity or even an expectation of the duration of the hospital care. While this seems obvious, I believe that this aspect of CDI is not widely understood.
Similarly, many of us work closely with the Quality/Performance Improvement Departments and it is becoming increasingly clearer that CDI might be the link that bridges the Utilization Management and Quality Departments. The importance of documentation crosses from the UM/UR arena into multiple aspects of quality and even public reporting.
Earlier this year, a new position for Clinical Documentation and Quality Improvement (CDQI) was created in our hospital. For now, it is a pilot program, with a single FTE staff person who has experience in both fields as well as in coding. Based on early experience, it is becoming increasingly apparent that the usefulness of the position is justifying its continuation. To illustrate, let’s review the impact of this pilot on the publicly reportable mortality data.
Mortality data is reported as the O/E ratio, wherein “O” refers to observed mortality and “E” to expected mortality. While the retrospective review cannot change the number of “O” cases, it can certainly affect the “E” value of the fraction. While initially the focus was only on those cases that did not have the highest severity of illness (SOI/ROM 4/4), it became apparent that even those cases have documentation opportunities. In fact, the duties of the CDQI currently include a mandatory review of all mortality cases. The main purpose of the review is not financial (through DRG change) but rather to look for opportunities to improve the SOI/ROM score by complete highest specificity documentation and coding of the clinical condition of those patients that contributed to their demise. If any potential financial opportunity exists, it would obviously be reflected in eventual coding, though this is neither the purpose nor goal for the review.
Along with the new CDQI position, we carried out a thorough educational effort of the medical staff via department/section meetings where specialty-targeted presentations were presented utilizing actual chart data to demonstrate the impact of clinical documentation on SOI/ROM scoring.
We share service line and department-specific mortality data with the physicians and the support staff. For example, the hospitalist group gets a weekly email with the drilldown of their mortality cohort, including the results of the reviews and the CDQI comments on documentation opportunities.
As the scoring for the charts reviewed improved we realized that the impact would be even greater if the review were expanded to all patients with the reportable DRG (MI, CHF, etc). This would optimize the documentation for SOI/ROM to support the “E” denominator of the mortality ratio.
These efforts could be similarly replicated for any other of the numerous quality measures, such as catheter-associated urinary tract infections (CAUTI), central line-associated blood stream infections (CLABSI,) or any other patient safety indicators. And while most quality departments utilize nurse reviewers to abstract/aggregate data for their respective areas, the biggest difference between a traditional nurse reviewer and the CDQI is the ability to place a query and get appropriate documentation on the record which achieves improvement in both the quality and coding areas.
So, have you ever thought of your CDI staff as supporting quality metrics? We have found that it works.
Categories: | Tags: | Comments: (0)
| View Count: (1025) | Return