Administrative Memos

200004

FROM: Theodore O. Will, Chief Executive Officer
DATE: Feb 09, 2000
SUBJECT: IPRO's Payment Error Prevention Program (PEPP): DRG 475 Aggregate Report
IPRO CONTACTS:

Kathy Terry, Ph.D., Director, Data Analysis & Evaluation, Extension 261 and Renato Estrella, R.H.I.A., Administrator, Hospital Information Management, Ext. 261

Enclosed please find a copy of IPRO's DRG 475 Aggregate Data Report (Attachment I). This Report is based on our analysis of the New York State Medicare inpatient claims data and validation of a sample of medical records obtained from forty (40) hospitals.

IPRO's analysis showed DRG 475 (Respiratory System Diagnosis with Ventilator Support) to be among the top twenty-five (25) DRGs billed in the 1998 NYS Medicare claims data. Further analyses identified hospitals that potentially overutilized DRG 475. We used the ratio of DRG 475 claims compared to DRG 127 (Heart Failure and Shock) claims as an indicator of potential overuse.

In addition, The Department of Health and Human Services, Office of the Inspector General (OIG) identifies DRG 475 (Respiratory System Diagnosis with Ventilator Support) as a DRG at great risk of upcoding. A 1998 OIG report analyzed the use of DRG 475 from 1993 to 1996. The report showed an increase in the use of this code nationally.

IPRO has selected forty (40) hospitals which had billing proportions of greater than twenty percent for DRG 475 when compared to DRG 127 for further study. The enclosed report provides our aggregate findings in regard to the coding/validation of medical records billed to DRG 475 obtained from the hospitals under study.

If your hospital was one of the forty (40) hospitals selected for IPRO assessment, we have also enclosed a copy of your hospital-specific findings along with this mailing. Within the next week a member of the PEPP intervention team will be contacting those hospitals with failed "DRG-change" indicators to schedule a meeting to discuss hospital interventions. The remaining twenty-two hospitals with failed "education" indicators may wish to contact IPRO for assistance in formulating follow-up. IPRO will be offering regional training sessions on coding for DRG 475 in March, 2000. Further information in regard to the dates and locations will follow.

For all other hospitals, while your proportion of DRG 475 to DRG 127 Medicare claims for 1998 may not have triggered IPRO review for this PEPP project, we encourage additional hospital analyses to confirm that your DRG 475 coding/billing patterns are consistent with Coding Clinic guidelines related to mechanical ventilation usage, that they accurately reflect the principal diagnosis and have correct sequencing of the principal diagnoses based on adequate physician documentation. To assist you, in addition to the planned coding seminars mentioned above, we have included a list of suggested actions for preventing coding errors in DRG 475 (see Attachment II).

Should you have any questions in regard to IPRO?s PEPP activities or this report, please feel free to contact Andrea Goldstein, Vice President, Health Care Assessment, Dr. Kathy Terry (data questions) or Renato Estrella (coding/DRG issues).

DRG 475 - Aggregate Data Report

I.Background

The Office of the Inspector General (OIG) lists DRG 475 (Respiratory System Diagnosis with Ventilator Support) as a DRG at great risk of upcoding. Furthermore, IPRO analysis shows that the volume of Medicare claims billed to DRG 475 ranks it among the top 25 DRGs in the 1998 NYS Medicare claims data. According to both 1997 and 1998 MEDPAR data, when comparing DRG "pairs", the national benchmark billing proportion for DRG 475 was 13.2% vs. 86.8% for DRG 127 (Heart Failure and Shock). Analysis of 1997 NYS Medicare claims data revealed a proportion of 14.5% for DRG 475 when paired with DRG 127. Subsequent analysis of 1998 NYS Medicare claims data revealed a proportion of 15.0% for DRG 475, indicating higher proportions than the national benchmark for each of these years.

As part of our Year One PEPP activities, IPRO conducted hospital specific analysis of calendar year 1998 NYS Medicare claims for DRG 475 which identified a number of hospitals with claim proportions above the NY State average of 15.0%. Hospitals with the highest proportions were identified as outliers and selected for inclusion in this project. Subsequently, Medicare inpatient medical records were requested for data abstraction to identify possible patterns in miscoding for these hospitals. Forty (40) acute care hospitals were identified. Twenty-five (25), or the maximum if less than twenty-five (25), medical records were randomly selected for review from each of these facilities. In total, 769 charts were requested.

II.Findings

In general, billing to DRG 475 requires that the patient?s medical record document the duration of time that a ventilator was used (exclusive of the time a patient is placed on ventilator support during surgery) and, include evidence of a principal diagnosis of respiratory failure or respiratory distress. Further, correct sequencing of diagnosis codes is critical for accurate DRG 475 billing. As shown in Figure 1, analysis of chart-abstracted data revealed that three (3) percent of the charts reviewed by IPRO either did not have evidence that the patient was on a ventilator during the hospitalization or indicated that the ventilator was only utilized during a surgical procedure (Indicator 1). Six (6) percent of the medical records abstracted (less those meeting Indicator 1) did not have documentation to support a principal diagnosis of respiratory failure or respiratory distress (Indicator 2). Sixteen (16) percent of the abstracted charts (less those meeting Indicator 1) demonstrated inappropriate billing to DRG 475 based on incorrect sequencing of diagnosis codes (Indicator 3).

Figure 1: DRG 475 Aggregate Report - DRG Change Indicators

N=(713) N=(694) N=(694)

C.I.(1-4%) C.I. (4-8%) C.I. (13-19%)

Additional data analyses identified other areas of concern that could contribute to inappropriate billing practices. The following data represents billing errors that would not result in a DRG change but rather relate to inappropriate usage of procedure and/or diagnosis codes. As shown in Figure 2, thirteen (13) percent of the medical records abstracted with documentation of ventilator usage for less than four (4) days did not bill to the appropriate procedure code (i.e., 96.71, - continuous mechanical ventilation for less than 96 consecutive hours). Eighteen (18) percent of the abstracted charts with documentation of ventilator usage for more than four (4) days did not bill to the appropriate procedure code (i.e., 96.72 - continuous mechanical ventilation for 96 consecutive hours or more). Most significantly, seventy-eight (78%) percent of the medical records reviewed by IPRO with respiratory failure due to an acute or chronic respiratory condition did not bill to the appropriate principal diagnosis code (i.e., 518.81- acute respiratory failure).

Figure 2: DRG 475 Aggregate Report - Educational Indicators


Educational Indicator	Num	Den	Percent	C.I.
Ventilation support for less than 4 days without procedure code 96.71 *	61	481	12.68%	10-16%
Ventilation support for 4 or more days without procedure code 96.72 *	37	209	17.70%	13-23%
Resp. failure associated with an acute/chronic resp. condition without principal diag. code of 518.81	417	535	77.94%	74-81%

* Note: There are 4 charts for which this information was not abstracted.

III. Conclusion

Combined, the results for both the DRG change indicators and the educational indicators clearly indicate that there are both hospital coding and billing practices for DRG 475, in select hospitals in New York State, which are in need of improvement.

To assist you, in addition to the planned coding seminars mentioned above, we have included a list of suggested actions for preventing coding errors in DRG 475 (see Attachment II).

PREVENTING CODING ERRORS IN DRG 475

Suggested Actions

Verify that your current coding and billing processes include:

Continuing education sessions with coders on ICD-9-CM guidelines in the following areas:

Respiratory failure as a principal diagnosis;
Correct sequencing of respiratory failure diagnoses;
Assignment and duration of mechanical ventilation usage.

Physician education on appropriate medical record documentation to support assignment of ICD-9-CM codes.
Physician query process for answering documentation questions from the coding staff.
Ongoing internal quality control to monitor the accuracy of coding assignment for DRG 475 and the completeness of documentation to support coding accuracy.