Home The Business of Medicine Beware of leading queries

Beware of leading queries

Some documentation efforts could lead to fraud

September 2011
physician documenting billing on computer

Published in the September 2011 issue of Today’s Hospitalist

HOSPITALS NOW ARE SCRAMBLING to maximize revenues “and pressuring physicians for thorough documentation to maximize billings. That’s led to the growth of an entire industry devoted to clinical documentation improvement.

Chances are you’re already fielding queries from these folks about clarifying diagnoses that you may or may not have documented. Making sure documentation is complete is even more important, now that insurers won’t pay to treat many conditions unless the chart reflects the fact that the condition was present on admission.

But some “documentation improvement specialists,” as they like to be known, can cross the line by sending physicians leading queries. Those in turn can lead to overpayments, fraud charges and greater potential liability for physicians. While documentation should certainly reflect the severity of a patient’s illness, doctors have to guard against documentation improvement efforts that go too far.

Reimbursement basics
Most hospitalists are familiar with the basics of inpatient Medicare reimbursement. A patient diagnosed with community-acquired pneumonia is assigned to a diagnosis-related group (DRG) for simple pneumonia (DRG 195). The Centers for Medicare and Medicaid Services (CMS) estimates that on average, a hospital treating a patient with simple pneumonia would receive about $4,472.

Question any written or verbal query that seems to lead you to a specific diagnosis

If, however, your documentation establishes complications or comorbidities for that patient, the patient would be assigned to DRG 194 “and the hospital would be reimbursed on average $6,398. If physicians can successfully make the case that the patient had major complications or comorbidities, reimbursement would jump even higher for DRG 193, which pays $9,325.

Aggregated over a year, documentation for higher-level DRGs can mean millions of dollars for a hospital. For an average 400-bed hospital, we estimate the improvement potential in revenue to be as much as $20 million per year.

Such high stakes have led to a proliferation of programs dedicated to improving clinician documentation. One survey released this year by HCPro suggested that nearly 80% of hospitals had such a clinical documentation improvement (CDI) program in place.

Clinical documentation specialists
The central figure in these programs is the clinical documentation specialist. Most are former nurses who have undergone extensive training on inpatient coding and reimbursement.

Hospitals also hire clinicians, typically an experienced physician with influence. The job of such specialists is to “assist” clinicians with documentation, and their primary tool is the written or verbal query.

The American Health Information Management Association (AHIMA), the guiding body for clinical documentation programs, defines a query as “a question posed to a provider to obtain additional, clarifying documentation to improve the specificity and completeness of the data used to assign diagnosis and procedure codes in the patient’s health record.”

AHIMA has also spelled out situations in which queries may be warranted. Examples include cases in which there are clinical indicators of a diagnosis but no documentation for that condition, or cases with no indication of present-on-admission status.

Problems arise when clinical documentation specialists go beyond clarification and ask questions that directly influence what clinicians document. This is called a leading query. During one conference this year, an advisory board member of the Association of Clinical Documentation Improvement Specialists, a professional association, defined “leading” as “when the provider is guided towards a particular conclusion/diagnosis by CDI or coding that results in an increase in reimbursement.”

Leading queries
What’s an example of a leading query? In a 2008 practice brief, AHIMA gave this example. An obtunded patient with a three-day history of nausea and vomiting was admitted. A chest X-ray revealed right lower lobe pneumonia, and the patient received clindamycin.

An example of a nonleading query might be: Can you further specify the etiology of the patient’s pneumonia? But here’s AHIMA’s example of a leading query: Is the patient’s pneumonia due to aspiration?

Such queries can get a hospital in trouble. The government went after one prominent hospital, alleging that it had filed false claims after clinical documentation specialists tried to claim such secondary diagnoses as malnutrition and respiratory failure. While the medical center denied those allegations, it settled the case in 2009 for $2.75 million.

Medicare recovery audit contractors (RACs) have recently started requesting query data, presumably to look for leading queries. To protect themselves, hospitals have put in place policies that make clear who to query, when to query and how to structure a query properly.

Perverse incentives
While hospitals have policies in place to protect themselves and provide guidance to clinical documentation specialists, rarely do they educate physicians.

In the case of a misleading query “when the clinical data do not support the documentation of a particular condition “we have our own ethical compass to guide us. But what about leading queries, where the violation is based on how the question is asked?

The risk for clinicians is further compounded by practices that many hospitals employ. One practice in particular involves measuring your response to queries ” referred to as query response rate or query agreement rate “and giving you a report card on those responses. Some groups even peg some portion of physicians’ quality bonus to their query response rate.

But such practices can create a perverse incentive by encouraging you to go along with queries as written. Also, because properly responding to queries can be time-consuming, a strong focus on query response rates creates even more of an incentive for busy clinicians to simply sign off on them. And the fact that many hospitalists are employed by their hospitals makes them easy targets for performance improvement initiatives in the form of a report card or incentives.

Risks to you
The risk to you is that by going along with a leading query or, in some cases, unknowingly accepting a misleading query, you are helping the hospital commit fraud in the eyes of Medicare.

What to do? Doctors should first educate themselves about AHIMA guidelines and be able to distinguish between leading and nonleading queries.

They should also always question any written or verbal query that seems to lead them to a specific diagnosis. And physicians should question any practice that encourages them to agree with queries they receive. Measuring or incentivizing query agreement rate is the wrong metric because it may encourage potentially fraudulent behavior.

Instead, “accurate documentation” is an appropriate metric to measure and even incentivize. When hospitals come up with metrics that target documenting accurately, those efforts invite full compliance.

kenjiKenji Asakura, MD, is a hospitalist at Swedish Medical Center in Seattle. Erik Ordal, MBA, is a Seattle-based independent consultant. Dr. Asakura and Mr. Ordal are also cofounders of ClinIntell, a company that markets clinical documentation improvement software.