Health Information Management

Understand potential pitfalls when codes misrepresent patient acuity and severity

JustCoding News: Inpatient, January 20, 2010

Want to receive articles like this one in your inbox? Subscribe to JustCoding News: Inpatient!

by Glenn Krauss, BBA, RHIA, CCS, CCS-P, CPUR, PCS, FCS, C-CDIS, CCDS

The incorporation of MS-DRGs into Medicare’s and other payers’ reimbursement methodologies introduced additional financial challenges to hospitals paid under the inpatient prospective payment system (IPPS).

Medicare recognized that with MS-DRGs would come improved documentation and coding patterns in hospitals, but that this would translate into an increase in hospitals’ reported patient severity through calculated case-mix that would not be truly representative of actual increases in patient severity.

A June 2009 MedPac commission letter highlights Medicare’s thoughts on this documentation and coding improvement phenomenon:

As expected, implementation of Medicare severity diagnosis related groups (MS-DRGs) in 2008 gave hospitals a financial incentive to improve medical record documentation and diagnosis coding to more fully account for each patient’s severity of illness. (For an explanation of how [documentation and coding improvement] occurs see the attachment to this letter). Documentation and coding improvements (DCI) improve documentation of patient severity, but they also increase reported case mix and payments without a real increase in patient severity or the resources hospitals must use to furnish inpatient care. Based on our analysis of hospital claims for fiscal year 2008, we concur with CMS’s conclusions that DCI occurred in 2008, and that under current law IPPS payment rates must be adjusted to offset its effects on reported case mix and payments. Our analysis indicates that downward payment adjustments will have to be taken over several years to recover past overpayments and to prevent overpayments from continuing in the future. 

DCI triggers argument that Medicare needs to recoup overpayments

At the onset of MS-DRGs, there was much discussion among various healthcare entities about the idea that DCI would likely play a role in case-mix increases that are not wholly explainable by legitimate increases in patient severity. Indeed, CMS found that DCI increased case-mix by 2.5% in 2008, which resulted in hospital overpayments of 1.9%.

CMS projected that by the end of 2009, hospitals’ DCI will have increased case-mix by a cumulative total of 4.8%. As a result, Medicare proposed an adjustment to the hospital market-basket update to DRG payments for 2010 in an effort to recoup past overpayments in fiscal years (FY) 2008 and 2009.

Based on objections from the American Hospital Association and other healthcare stakeholders, CMS did not include the payment adjustment proposal in the IPPS final rule for 2010.

Nevertheless, the contention that Medicare needed to recoup these DCI overpayments resurfaced with MedPac’s recommendations for 2011 Medicare payment rates, which the commission issued January 14. MedPac recommended that Congress give hospitals a fiscal year 2011 payment update equal to the rate of change in the market basket index, currently projected at 2.4%. The commission also recommended a reduction in the inpatient update by up to 2% in 2011, 2012, and 2013 to reflect changes in documentation and coding.

This was despite the 2008 hospital negative Medicare reimbursement margins of 7.2%, which CMS is slated to continue in 2009 and 2010.

Clinical scenario illustrates changes in documentation

While the provider community is naturally suspicious of Medicare’s contention of unwarranted increases in hospital case-mix due to DCI, many HM/coding and reimbursement consultants can attest to the fact that the MS-DRG system has altered certain coding patterns. Consider the following clinical scenario:

A demented patient is admitted to the hospital with a change in mental status from the nursing home. Initial thought is that the patient was dehydrated and may have an infection based on abnormal lab values, constitutional appearance, and response to treatment in the ED, mainly IV fluid boluses and a dose of IV antibiotics. Of note is the patient has a Foley in place for a neurogenic bladder, and the nursing home reports that the patient has been having decreased urine output that has worsened over the last few days. Physician orders strict Ins and Outs (I&O) measurements while the patient is hospitalized, as well as measurement of urine output. Patient responds nicely to medical management during hospitalization that consists of continued fluid resuscitation for dehydration, pre-renal azotemia and acute renal insufficiency. Infection was ruled out through urinalysis and blood cultures with lungs and chest x-ray not demonstrating a pneumonic process.

Coding sometimes overstates patient severity

Clinical documentation improvement (CDI) specialists query the attending physician regarding “acute renal insufficiency” versus “acute renal failure,” and the physician confirms the patient was in acute renal failure upon admission. Documentation in the progress notes and discharge summary reflect this diagnosis.

Final discharge diagnoses include the following:

  • Severe hypotension with dehydration, resolved
  • Congestive heart failure (CHF)
  • Acute renal failure (ARF), resolved
  • Hypertension, controlled
  • Diabetes type II, under good control
  • Cardiac arrhythmia
  • Hypothyroidism

Consider the accuracy of the following coding for this scenario:

  • Principal diagnosis: Hypotension (code 458.9)
  • Secondary diagnoses: ARF (code 584.9), hypertension (code 401.9), CHF (code 428.0), diabetes (code 250.00), cardiac arrhythmia (code 427.89), and hypothyroidism (code 244.9)
  • MS-DRG 314 (Other circulatory system diagnoses with MCC, relative weight: 1.7552 [FY 2009])

From a clinical coding and compliance standpoint, assignment of hypotension as the principal diagnosis of record is inaccurate and plays right into Medicare’s viewpoint of the need for documentation and coding adjustment to mitigate changes in coding patterns that don’t accurately reflect patient severity.

The most clinically accurate principal diagnosis to assign for this scenario is ARF, MS-DRG 684 (Renal failure without CC/MCC, relative weight: 0.7305 [FY 2009]).

Case study reveals potential for inaccurate coding

In contrast, compare the clinical presentation and scenario above with the following case study, in which there is missing clinical documentation that detracts from the reporting and capture of the patient’s true clinical acuity.

A patient is admitted to the hospital from the nursing home with pneumonia. This is the second admission for this patient, which is not surprising given her dysphagia as a sequel of a stroke experienced last year. She has a percutaneous endoscopic gastric (PEG) tube in place associated with the difficulty in feeding, per patient preference. Physician treats patient’s pneumonia as aspirational in nature and selects IV antibiotics accordingly. Patient responds well to treatment and is subsequently discharged on day five with an additional five-day course of antibiotics on board on the way home.

Final discharge diagnoses include:

  • Aspiration pneumonia 
  • CHF, chronic systolic 
  • Hypertension
  • Medical noncompliance
  • Obesity
  • Peripheral vascular disease

Consider the accuracy of the following coding for this scenario:

  • Principal diagnosis: Aspiration pneumonia (code 507.0)
  • Secondary diagnoses: CHF (code 428.0), chronic systolic (code 428.22); hypertension (code 401.9); medical noncompliance (code V15.81); obesity (code 278.00), and peripheral vascular disease (code 443.9)
  • MS-DRG 178 (Respiratory infections and inflammation with CC, relative weight: 1.4983 [FY 2009])

On face value, this coding appears to be accurate and compliant based on the patient’s clinical presentation and treatment, properly applying official guidance governing principal diagnosis assignment.

However, a quick review of the record beyond explicit physician documentation in the ER note, history and physical, progress notes, and discharge summary reveal a more complete clinical picture that illustrates how additional reporting of a diagnosis under active management can contribute to a higher severity MS-DRG assignment for this case.

Review of the medication administration record indicates that this patient received 40 MG IV Lasix on two separate occasions for three consecutive days, begging the question of whether the patient was experiencing an acute exacerbation of chronic systolic heart failure.

A review of the nursing notes reveals that the patient was complaining of worsening shortness of breath. Nursing personnel then contacted the physician, who ordered a chest x-ray and IV Lasix. Findings from the chest x-ray include cardiomegaly with increasing pulmonary vascular congestion in conjunction with patient’s compromising shortness of breath.

Pending clarification of whether the physician was managing an acute exacerbation of chronic systolic heart failure and provided the physician actually documents this in the record, the final code assignment would be as follows: 

  • Principal diagnosis: Aspiration pneumonia (code 507.0)
  • Acute on chronic systolic heart failure (code 428.23)
  • Hypertension (code 401.9)
  • Medical noncompliance (code V15.81)
  • Obesity (code 278.00)
  • Peripheral vascular disease(code 443.9)
  • MS-DRG 177 (Respiratory infections and inflammation with MCC, relative weight: 2.0393 [FY 2009])

Be aware of the pitfalls to avoid perpetuation of inaccuracies

It’s important for coders to recognize the potential for and the need to stop the perpetuation of increases in case-mix that unsubstantiated by increased patient acuity. Sometimes, we as coders try to convince ourselves of the “correctness” of principal diagnosis or secondary diagnosis selection by the effect code assignment will have upon reimbursement, not taking into account sound understanding and the clinical appropriateness of code assignment.

Coders need to refine their focus, which should include the following steps:

  • Review the record from a clinical point of view
  • Identify when a clinical query is relevant regardless of the effect on reimbursement
  • Follow through with a compliant, nonleading query
  • Assign diagnoses in congruence with clinical medicine and proper application of official coding guidelines and policies

By doing so we can avoid contributing to the problem Medicare cites as justification for continued adjustment based on the effect DCI has on patient severity reporting.

Editor’s note: Krauss is an independent coding consultant in Milton, WI. E-mail him at glennkrauss@earthlink.net.



Want to receive articles like this one in your inbox? Subscribe to JustCoding News: Inpatient!

Most Popular