Concurrent Surgical Patient Safety Indicator Evaluation Results in More Accurate Reporting and Reimbursement

We sought to determine whether concurrent (before discharge) Agency for Healthcare Research and Quality patient safety indicator evaluation would result in a more expeditious review, accurate reporting, and improved reimbursement. We compared the period of preconcurrent (preC) coding (January 2012 to June 2012) with the period after concurrent coding (postC) began (July 2012 to December 2012) for total billing errors. There were 276 records reviewed in the preC versus 424 in the postC time periods. Overall coding errors were 225 (81.5%) preC versus 365 (86.1%) postC (P = nonsignificant), whereas documentation errors were present in 26 (9.4%) preC versus 40 (9.4%) postC (P = nonsignificant). Total charges were $3,782,024 preC and $2,011,144 postC. Recodes requiring rebilling were 21 (7.6%) preC for a total of $213,723 rebilled versus four (0.9%) postC for a total of $31,327 rebilled (P < 0.0001). Time from service to review was 98.7 preC versus 52.3 postC days (P < 0.0001). Time from service until rebill submitted averaged 100.8 preC versus 54.0 postC days (P = 0.06). Concurrent review allows for more accurate reporting because recodes are completed before discharge. Billing delays prolong time to reimbursement and results in loss of revenue.

THE AGENCY FOR HEALTHCARE Research and Quality and Centers for Medicare and Medicaid (CMS) are actively engaged in ensuring that they ''get what they pay for'' by critically comparing patient safety indicators (PSIs) at academic medical centers. This information comes primarily from administrative databases that can be fraught with error and is heavily dependent on billing codes most often completed after patient discharge and before clinical review. Identificationofcodinganddocumentationerrorsisoften delayed resulting in rebilling, delayed reimbursement, and inaccurate United Healthcare Consortium (UHC) rankings. Given these significant implications, we sought to determine whether concurrent (before discharge) PSI evaluation would result in a more expeditious review, accurate patient safety reporting, and improved reimbursement.

We performed an Institutional Review Board-approved retrospective review of patient records that required a submission for rebilling to CMS or private insurance payors from January 2012 through December 2012 at our tertiary academic medical center. We compared billing accuracy before concurrent coding was initiated (January to June 2012) to after concurrent coding was initiated (July to December 2012). Before July 2012 (preC) all charts were coded after discharge and the chart was not billed until all coding was completed. After July 2012 (postC), coding was initiated while patients were still in the hospital allowing for real-time documentation clarification by the provider. These charts were not billed until after discharge and after coding was completed. The two groups were compared for total billing errors. We then evaluated the reasons for these errors and the effect on the amount and time to rebill.

There was a total of 700 charts reviewed during the timeframe of the study. There were 276 records reviewed in the preC versus 424 in the postC time periods. The vast majority of errors were the result of coding inaccuracies: 225 (81.5%) preC versus 365 (86.1%) postC (P 4 0.1117) followed by issues with documentation: 26 (9.4%) in preC versus 40 (9.4%) postC (P 4 1.0000). The other reasons for rebilling being required include identification of a present on admission diagnosis: six (2.2%) in preC vs 10 (2.4%) in postC (P 4 1.0000) and delay in physician re- sponse to a query: eight (2.9%) preC vs nine (2.1%) postC (P 4 0.6169). The modifications made to the patient's chart are listed in Table 1.

Total charges in the preC group were $3,782,024 versus $2,011,144 in the postC with recodes requiring rebilling at 21 (7.5%) preC versus four (0.9%) postC (P < 0.0001). The total amount rebilled was $231,723 (5.7%) preC versus $31,327 (1.6%) postC. Time from service to review (days) was 98.7 preC versus 52.3 postC (P < 0.0001). Time from review to rebill (days) was 30.7 preC versus 25.5 postC (P 4 0.8043). Time from service until rebill submitted averaged (days) 100.8 preC versus 54.0 postC (P 4 0.06).

Over the last decade, pay-for-participation has transformed into pay-for-performance such that high- performing providers receive financial incentives. However, it has continued to morph and now is more punitive with penalties for ''poor'' performance despite known inherent inaccuracies in the use of administra- tive data.1

One of the major challenges with these penalties is that performance is defined by metrics from the UHC and collected based on administrative data. Our error rates of over 80 per cent for both preconcurrent and concurrent coding demonstrate this well. Moreover, the slightly higher error rate in the concurrent group emphasizes that these errors are difficult to eradicate and perhaps more readily identified using concurrent coding.

In a previous study by the authors, we performed physician review of PSIs identified by codes. In this article, the overall positive predictive value of admin- istrative detection of PSIs through billing codes was 83 per cent with positive predictive values as low as 67 per cent for some PSIs.2 These findings are consistent with the literature that notes a positive predictive value for coding ranging from 29 to 83 per cent.3, 4

Despite all of the issues with the approach of using administrative data for safety monitoring, it is unlikely to change. The current setup of postdischarge review goes through thorough documentation review to en- sure accuracy of coding at odds with efficient billing. Many private insurers have a 72-hour limit from the time of discharge to bill a patient stay. This makes it very challenging for documentation nurses to complete their assessment and make changes to codes when necessary. In fact before concurrent coding, reviews took over three months to complete, thereby delaying rebills and consequently reimbursements. Another important finding is that the recodes requiring rebilling decreased significantly after postconcurrent coding, which should result in more accurate UHC reporting and more timely and appropriate reimbursements.

Based on our initial data, concurrent coding while patients are in the hospital can significantly decrease the time to review charts. It may be more sensitive as well identifying slightly more coding errors, but be- cause these are identified early, less rebilling is re- quired. As we move more toward a punitive system with lost reimbursements and fines for poor perfor- mance, concurrent coding may be a prudent approach to address these problems. Given that up to two per cent of Medicare reimbursement will be lost if metrics are not met, focusing resources on concurrent coding would be an excellent investment. Healthcare adminis- trators should support such efforts and trend data so that we can continue to optimize care and reimbursements in this new era of health care.

REFERENCES

1. Farmer SA, Black B, Bonow RO. Tension between quality measurement, public quality reporting, and pay for performance. JAMA 2013;309:349-50.

2. Ramanathan R, Leavell P, Stockslager G, et al. Validity of Agency for Healthcare Research and Quality Patient Safety In- dicators at an academic medical center. Am Surg 2013;79:578-82.

3. Borzecki AM, Kaafarani H, Cevasco M, et al. How valid is the AHRQ Patient Safety Indicator 'postoperative hemorrhage or hematoma'? J Am Coll Surg 2011;212:946-53.

4. Utter GH, Zrelak PA, Baron R, et al. Positive predictive value of the AHRQ accidental puncture or laceration patient safety in- dicator. Ann Surg 2009;250:1041-5.

THERESE M. DUANE, M.D.,* RAJESH RAMANATHAN, M.D.,[dagger] PATRICK LEAVELL, B.S.N., R.N.,[double dagger] CATHERINE MAYS, R.N.,[double dagger] DALE HARVEY, R.N.[double dagger]

From the *Division of Trauma, Critical Care & Emergency Surgery, the [dagger]Department of Surgery, and [double dagger]Performance Improvement, Virginia Commonwealth University Medical Center, Richmond, Virginia

Presented at the Annual Scientific Meeting and Postgraduate Course Program, Southeastern Surgical Congress, Savannah, GA, February 22-25, 2014.

Address correspondence and reprint requests to Therese M. Duane, M.D., 1200 E. Broad Street, P.O. Box 980454, Richmond, VA 23219. E-mail: [email protected].