December 19, 2011

Coding From Diagnostic Reports
For The Record
Vol. 23 No. 23 P. 31

In the inpatient setting, abnormal findings identified in diagnostic reports are not listed as secondary diagnoses unless the physician indicates their clinical significance. If findings are identified and further monitoring and testing is necessary and ordered to evaluate the condition or treatment is ordered, it is appropriate to ask the physician whether a corresponding diagnosis should be added (ICD-9-CM Official Guidelines for Coding and Reporting, effective October 1, 2011, pages 91-92).

Coders are not allowed to assign codes directly from impressions included on diagnostic reports, such as x-rays, MRI, CT scans, electrocardiograms, echocardiograms, and pathology, even if a physician has signed the diagnostic report. The diagnosis must be confirmed by the physician in the body of the medical record (eg, progress notes or discharge summary) before it can be coded. However, if the diagnostic report is adding specificity to an already-confirmed (physician-documented) diagnosis, then the coder may use the more specific code based on the diagnostic report without obtaining physician confirmation (AHA Coding Clinic for ICD-9-CM, 1999, first quarter, page 5).

The following are examples of findings identified on diagnostic reports and recommendations for follow-up:

• Elevated blood urea nitrogen, creatinine, and urine-specific gravity or receiving IV fluids is clinically significant and should be brought to the attention of the physician if no diagnosis has been documented.

• A hematocrit of 28%, even though asymptomatic and not treated, may have been evaluated by the physician with serial hematocrits. Because this is outside the range of normal laboratory values and has been further evaluated, it is significant enough to ask the physician whether an associated diagnosis should be documented.

• A routine preoperative x-ray on an elderly patient reveals collapse of the vertebral body. The patient was asymptomatic, and no further evaluation or treatment was carried out. This is a common finding in elderly patients and is insignificant for this episode.

• An echocardiogram showed mitral regurgitation. It was ordered to evaluate a murmur. The patient was referred to a cardiologist for follow-up. It is appropriate to ask if the diagnosis should be added to the final diagnostic statement.

Every record has to stand on its own individual merits. Before a condition can be coded, it must meet two requirements. One, the condition has to be documented by a physician in the body of the medical record, such as history and physical, consultant report, progress notes, or discharge summary. The second requirement is that it must affect patient care in terms of requiring one of these five criteria: clinical evaluation, therapeutic treatment, diagnostic procedure, extended the length of hospital stay, or increased nursing care and/or monitoring.

Laboratory Results
Additional diagnoses should not be arbitrarily added on the basis of an abnormal laboratory finding alone. To make a diagnosis on the basis of a single lab value or abnormal diagnostic finding is risky and carries the possibility of error. The physician must diagnose the patient.

A value reported either lower or higher than the normal range does not necessarily indicate a disorder. Many factors may influence the value of a lab study. These include the method used to obtain the sample (eg, a constricting tourniquet left in place for more than one minute prior to collecting the sample will cause an elevated hematocrit and potassium level), the collection device, the method used to transport the sample to the lab, the calibration of the machine that reads the values, and the condition of the patient. An example is a patient who, because of dehydration, may show an elevated hemoglobin level due to increased blood viscosity.

It is the physician’s responsibility to document the patient’s diagnoses. In the inpatient setting, a diagnosis based on an abnormal laboratory result or diagnostic test should not be determined by someone other than a physician. The physician must document the diagnosis in the medical record before it can be coded. In addition, it is not adequate for a physician to use only arrows (↑ or ↓) to indicate a diagnosis, even if treatment was given for that condition. For example, the physician documents in the progress notes, “↓Na. Decrease fluid intake. Change IV fluids.” In this example, hyponatremia could not be coded without the physician documenting “hyponatremia.” Query the physician regarding the patient’s specific diagnosis. In other words, it is not acceptable to code a diagnosis based on the physician’s up or down arrows or lab values. The physician must document the actual diagnosis (AHA Coding Clinic for ICD-9-CM, 2011, first quarter, pages 17-18).

Pathology Reports
Coding Clinic has clearly stated that in an inpatient setting, coders are not able to assign codes based on the pathology report without physician confirmation of the diagnosis. For example, breast cancer is documented, and the pathology shows mets to lymph nodes. Coders are not allowed to pick up a code for the lymph node mets until confirmed by the physician. In addition, if the physician documents “breast lump” and the pathology confirms it is breast cancer, coders cannot code “breast cancer” until the physician confirms this in the body of the record. In this example, the pathology is not providing specificity to an already confirmed diagnosis—it was providing a separate diagnosis, breast lump vs. breast cancer (AHA Coding Clinic for ICD-9-CM, 2008, third quarter, pages 11-12).

Cancer Staging Form
Coders are allowed to use a completed cancer staging form for coding purposes when it is signed by the attending physician (AHA Coding Clinic for ICD-9-CM, 2010, second quarter, pages 7-8).

Autopsy Reports
Coders can use the autopsy report for code assignment to provide greater detail and specificity. The coder can code diagnoses from the autopsy report as long as the diagnosis does not conflict with the rest of the medical record and as long as the diagnosis is also listed elsewhere in the medical record. If there are conflicting diagnoses or diagnoses listed only in the autopsy report, the attending physician needs to be asked for clarification before assigning a code for the diagnosis (AHA Coding Clinic for ICD-9-CM, 2001, first quarter, pages 5-6).

Coding from diagnostic reports is dependent on the physician documentation in the medical record and application of the Official Coding Guidelines for inpatient care. Also, use specific AHA Coding Clinic for ICD-9-CM and American Medical Association CPT Assistant references to ensure complete and accurate coding.

— This information was prepared by Audrey Howard, RHIA, of 3M Consulting Services. 3M Consulting Services is a business of 3M Health Information Systems, a supplier of coding and classification systems to more than 5,000 healthcare providers. The company and its representatives do not assume any responsibility for reimbursement decisions or claims denials made by providers or payers as the result of the misuse of this coding information. More information about 3M Health Information Systems is available at www.3mhis.com or by calling 800-367-2447.

ICD-10-PCS Coding Guidelines
The following is a review of the conventions guidelines found in the ICD-10-PCS Coding Guidelines available on the Centers for Medicare & Medicaid Services’ website.

• Seven characters are required for each ICD-10-PCS code. The position of each character (eg, axis of classification) contains similar identifying information about that procedure. For example, the fifth character identifies the procedural approach.

• There are 34 possible values in each of the seven character positions (0-9 and A-Z except I and O).

• The accurate values for each of the seven digit characters can be found only in the specific table for that procedure. The index can be useful for determining which table to use, but it is not necessary to use the index, as the user can proceed directly to the table to construct the procedure code.

• Each ICD-10-PCS code must contain a valid value in each of the seven characters in order to be a valid procedure code. If information is not readily available in the patient record, the physician must be queried for additional information.

• The first three characters of the procedure code are located in the upper or header portion of the table, with character values listed in columns within the table. Valid combinations of values are represented by rows within the table, and any combination not found within a single row is considered invalid.

• The use of the word “and” in a code description means “and/or.”

“It is the coder’s responsibility to determine what the documentation in the medical record equates to in the PCS definitions.”

— AH