Updated on: June 22, 2012

Moving From Coding to Payment Audits: 12 Easy Steps

By
Original story posted on: February 28, 2011

spencerCThe idiom "comparing apples to oranges" is a common response when two items that cannot be validly compared are, in fact, compared. The metaphor also may indicate that a false equivalence has been made between two items, such as when an apple is faulted for not being a good orange. Such is the case when counting errors in coding that result in a coding accuracy score (apple) to counting errors in coding that result in a payment error rate (orange).

 

The question is how can providers move from the first to the second? The place to start, of course, is to consider where the Centers for Medicare & Medicaid Services' (CMS) stand in the orchard.

 

Getting on the Same Page

 

The Centers for Medicare & Medicaid Services' (CMS) priority-which is, in fact, a legislative mandate from Congress-to reduce fraud, waste, and abuse in the Medicare program is at an unprecedented high. To achieve this goal, the agency has extended its tentacles to contractors who will assist with its goal to return $25 billion to the Medicare Trust Fund.  Listed below are the contractors who are now in operation and their missions.

 

  • Comprehensive error rate testing (CERT) program contractors issue reports about high-risk payment MS-DRGs.
  • Medicare fiscal intermediary (FI) and administrative contractors (MACs) conduct pre-bill and retrospective medical reviews for payment recoupment.
  • Recovery Audit Contractors (RAC) conduct retrospective audits for payment recoupment.
  • The TMF Health Quality Institute releases high-risk payment MS-DRGs through the Program for Evaluating Payment Patterns Electronic Report (PEPPER).
  • Program safeguard contractors (PSC) and zone program integrity contractors (ZPIC) recoup payment for errors related to fraudulent billing patterns.

 

There's one word in common in the missions of all of the above contractors and that word is "payment." In one way or another, they are all recouping payments, although their success can be attributed to a myriad of errors, including incorrect coding, MS-DRG assignment, and level of care setting (medical necessity of an inpatient versus outpatient setting); inappropriate billing for non-covered services; and double billing.

 

Looking at the orange in this way, it's clear to see that the metric of choice of Medicare's external regulatory agencies is improper payment reported as a paid claims error rate (that is, the financial error rate).

 

In contrast, hospitals and other healthcare organizations have been using an apple as their common metric over the past three decades. MS-DRG accuracy is the metric for measuring coding accuracy and an indirect measure of improper payment due to coding. Improvements in the coding process as well as implementation of costly controls, automation, and training efforts may or may not correct patterns of improper payment due to coding.

 

Time for a Change

 

Wouldn't a metric that results in findings that shows a direct relationship between incorrect code assignment and improper payment be more likely to result in corrective action with sustainable results reducing improper payment?

 

CMS's expectation for all healthcare providers is to "prevent..., detect ... and resolve improper payment." As stated above and repeatedly in Medicare messages, "payment" is the common denominator-not codes. Hospitals' and healthcare organizations' continued use of codes as the common denominator to perform measurements is outdated. It's time that they begin the transition to this new claims' audit strategy, which will align them with the method used by external regulatory agencies.

 

The 12 steps to this new strategy--an inpatient coding payment audit--are listed below. These also may be applied to audits performed in other areas including (but not limited to) outpatient coding, medical necessity, post-acute care transfer MS-DRGs, documentation, non-coverage, billing, and re-admissions.

 

Twelve Steps to Payment Audits

 

Step 1: Define the primary and the secondary purposes of the audit. Although there may be others, the most common primary purposes are the following:

 

  • Improper payment (underpayment and/or overpayment) due to incorrect coding and inadequate documentation as reported by a paid claims error rate; and
  • Correct code assignment as reported by MS-DRG accuracy.

 

Secondary purposes include conducting a baseline audit to compare progress over time and a follow-up audit to assess improvement from baseline.

 



 

Step 2: Select patient type. Will your audit focus on inpatient or outpatient records?

 

Step 3: Choose the payer or payers you want to audit. Options include Medicare, non-Medicare, and all federally funded programs.

 

Step 4: Determine exempt patient type(s). Inpatient prospective payment system audits may exclude exempt units such as rehabilitation, psychiatric, or skilled nursing facilities.

 

Step 5: Determine timeframe of audit. For example, you could review paid claims for the last fiscal year, last six months or last three months.

 

Step 6: Select patient sub-type. If you choose to do a representative sampling, the following could be reviewed:

 

  • All MS-DRGs billed;
  • All high-risk MS-DRGs billed (such as the DRGs posted by RAC, CERT, PEPPER);
  • Your hospital's top 20 high-volume MS-DRGs;
  • Your hospital's top 20 high-relative-weight MS-DRGs; or
  • Targeted sampling by some combination of the following:
    • Individual DRG (e.g., 871 or 177);
    • DRG grouping (e.g., 871, 872 or 177, 178, 179);
    • PEPPER DRG grouping (e.g., 871, 872, 689, 690 or 177, 178, 179, 193, 194, 195);
    • Major complication/co-morbidity (MCC) or complication/co-morbidity (CC); or
    • CC or without MCC/CC (underpayment).

 

Step 7: Determine exempt patient sub-type(s), such as one-day length of stay for discharge disposition 01 (home) or 04 (supportive care) for medical necessity DRGs, or exclude MS-DRGs with case counts under 10.

 

Step 8: Run Excel report to include full-sampling frame (total population). Include account number, medical record number, discharge date, and payer, and sort data fields by medical record number order (avoid bias).

 

Step 9: Determine audit type and sample size (using an automated sample-size selector tool). The audit types include the following: discovery, full, full operations-and-systems review, probe audit,

 

Discovery audit (50 records) is the initial audit conducted as a "test" sample to determine if there is financial risk. If the discovery audit results in greater than a 5 percent financial error rate, then a full audit is performed, and a full operations-and-systems review must be conducted.

 

A full audit is a follow-up audit to a discovery audit that has identified financial risk. It also can be used as a baseline audit to compare progress annually. (The number of records is based on desired confidence level and the margin of error.) A common confidence level applied is 95 percent and a +/- 5 percent margin of error, and the number of records may range from 138 to 374.

 

A probe audit (20 records), unlike a discovery audit, targets a stratified population. For example, it may include all of the following types of cases: pneumonia, excisional debridement, extensive and non-extensive operating room procedures unrelated to principal diagnosis, and sepsis.

 

If the financial error results of a probe audit are less than a 7 percent paid claims error rate, then monitoring should continue. If the paid claims error rate approaches 10 to 30 percent, then corrective action (including pre-payment review) may be warranted in a matching percentage. If payment error rates reach 50 percent or higher, then 100 percent pre-payment review may be required, and the facility's compliance and legal departments should be contacted.

 

 



 

Step 10: Select unbiased record sampling using a probability sampling method (automated tool), such as RAT-STATS(1) or the Research Randomizer(2).

 

  • Enter the total population from the sampling frame (see step 8), enter the number of records determined by sample size selector and audit type, and run the sampling tool, listing the random number in sequential order.
  • Match up the random number to the medical record number in the report (step 8). If replacement records are desired (for cases that do not meet audit parameters), then follow probability sample for replacements (spares).
  • Prepare pull list.

 

Step 11: Conduct audit after retrieving the following documents: remittance advice, UB-04, medical record, and data-collection tool. Validate the following components: MS-DRGs, principal diagnosis, MCC, CC, present on admission (POA), hospital-acquired condition (HAC), procedure impacting payment, and query validation.

 

Step 12: Collect and report data.

 

  • Collect payment MS-DRG changes due to code changes (overpayment plus underpayment).
  • Divide this by total net payments in the audit.
  • Calculate paid claims error rate (evaluate results to step 9) and determine corrective action.

 

Report payment audit results, benchmark to CMS contractor results, and conduct baseline and follow-up audits to track and trend progress over time. Forecast annualized improper payments based on baseline audit results (annual Medicare net inpatient revenue multiplied by the last year's payment audit results on the annual baseline audit). Tally improper payments from all audits and compare against forecasted improper payments. Re-adjust annual goals to reduce improper payments on an annual basis. Document all efforts to prevent, detect, and resolve improper payments in the claims process and report reductions in fraud, waste, and abuse to the hospital's governing board.

 


 

(1) This package of statistical software tools is designed to assist the user in selecting random samples and evaluating the audit results. It has been used by the Department of Health & Human Services Office of Inspector General since the early 1970s.

(2) This free service at www.randomizer.org offers software to those interested in conducting random assignment and random sampling.

 

To comment on this article please go to editor@racmonitor.com

 

"Embracing Healthcare Fraud Detection Technology to Improve Provide Payment Experience"

 

About the Author

 

Carol Spencer, RHIA, CHDA, CCS, is a senior consultant with Medical Learning, Inc. (MedLearn®) in St. Paul, Minn. She brings more than 20 years of experience in health information management, coding, teaching, data quality and revenue integrity. She is an accomplished local, regional, and national speaker and author covering topics such as recovery audit contractors (RACs), payment audits, MS-DRG reimbursement systems, ICD-9-CM coding, and is an AHIMA-approved ICD-10-CM/PCS Trainer. Ms. Spencer is a thought-leader in data analytics and an expert on compliance in coding, query, and clinical documentation improvement strategies.



Contact the Author

 

cspencer@medlearn.com


This email address is being protected from spambots. You need JavaScript enabled to view it.