Crucial coding rules for clinicians


Physicians are often perplexed and annoyed by the documentation specificity they're asked to provide. Due to complex, sometimes nonsensical coding rules, perfectly acceptable and clinically accurate terminology is occasionally inadequate to translate care into the precise ICD-10 codes that accurately reflect the patient's true condition.

Photo by Thinkstock
Photo by Thinkstock

More detailed, specific documentation allows the assignment of codes that correctly express the severity of illnesses, complexity of care, and overall use of resources for the admission. These factors then drive almost all measures of quality and performance and also determine reimbursement and pay-for-performance metrics.

The authoritative regulatory sources for coding and documentation that have the force of law include the ICD-10 coding classification itself and the CMS Official Guidelines for Coding and Reporting (OCG). These are updated each year on Oct. 1. The American Hospital Association's Coding Clinic provides additional advice for interpreting the authoritative sources but does not supersede them. There are an enormous number of coding rules and guidelines, but a working knowledge of just a few that are most pertinent for clinicians is a valuable asset.

Principal diagnosis. Selection of the principal diagnosis is crucial for correct diagnosis-related group (DRG) assignment and reporting of the patient's condition. The principal diagnosis that should be coded is sometimes different from what might be considered the clinical principal diagnosis. The principal diagnosis for coding and billing purposes is defined as the condition established “after study” to have been primarily responsible for prompting the hospital admission. The condition, or at least signs and symptoms that were later attributed to it, must have been present on admission and must have been the primary reason for admission.

As an example, consider a patient who is admitted for community-acquired pneumonia and 2 days later has a hemorrhagic stroke. The principal diagnosis should be pneumonia, no matter how long the patient remains in the hospital or what resources are required for management of the stroke.

Whenever a patient is converted to an inpatient from medical or surgical observation, the principal diagnosis should be the condition that necessitated inpatient admission, which may be different from the reason for the initial observation care. Always document clearly the specific reason for the change in admission status.

It is important to note that coders are responsible for determining the principal diagnosis based on physician documentation, complex coding rules, and a thorough review of the entire medical record. Physicians do not decide what diagnosis is assigned as principal diagnosis because most of us do not have the necessary knowledge of coding rules to do so.

Secondary diagnoses. Secondary diagnoses (also known as “additional” diagnoses) are defined as “all conditions that coexist at the time of admission, that develop subsequently, or that affect the treatment received and/or the length of stay.” The OCG states that all conditions (either present on admission or occurring during admission) are to be coded as secondary diagnoses if they require 1 or more of these 5 things: evaluation, treatment, diagnostic procedure, increased nursing care or monitoring, or extended length of stay.

Chronic conditions (including hypertension, congestive heart failure, asthma, chronic obstructive pulmonary disease, Parkinson's disease, and diabetes) that require no more than continuation of the patient's usual home medications meet the criteria and should be properly documented and coded even if no other attention or management is required. The clinical rationale for this coding rule lies in the fact that such chronic conditions increase risks, clinical decision making, nursing care, and length of stay. As an example, morbid obesity is always considered clinically significant since it requires “increased nursing care” at a minimum and often prolongs the length of stay.

Principal procedure. Although all procedures are usually coded, the principal procedure is defined as a procedure that is performed for definitive treatment rather than simply diagnostic or exploratory purposes or as a procedure necessitated by a complication. However, the procedure most closely related to the principal diagnosis, whether diagnostic or therapeutic, should take precedence.

There are a few peculiar exceptions. For example, whenever a diagnostic bone biopsy is performed in conjunction with a therapeutic kyphoplasty for osteoporotic compression fracture, the bone biopsy is assigned as the principal procedure.

Procedures performed in an operating room (denoted as OR procedures) have a major impact on the DRG assigned and the severity of illness classification and complexity of care, although again there are a few exceptions. Two procedures that are generally not performed in an operating room are still classified as significant OR procedures affecting the DRG: transbronchial biopsy conducted in an endoscopy suite, and excisional debridement of a wound, which may even be done at the bedside. Oddly, a water-jet excisional debridement is not considered an OR procedure, even when it requires general anesthesia in the operating room.

Present on admission (POA). A “present on admission” indicator (see Table) must be reported for each diagnosis code. For most purposes, POA indicators of “Y” (yes, present on admission) and “W” (unable to clinically determine) are considered equivalent. Likewise “N” (no, not present on admission) and indicator “U” (unknown due to insufficient documentation) are equivalent.

The POA status of a condition should never be left as unknown (indicator U) since it means that the clinician did not provide the needed documentation and the condition will be considered hospital-acquired (occurring after admission). A clinician should always use reasonable clinical judgment to decide if, in his opinion, a condition was or was not present on admission, or state that he is unable to clinically determine the POA status (indicator W) so that it is treated as present on admission, not hospital-acquired.

A condition is present on admission if it is present at the time the order for inpatient admission occurs. This includes all pre-existing conditions. Conditions that develop during any preadmission outpatient encounter—including in the ED, during observation care, or during ambulatory surgery—are considered present on admission. With rare exceptions, any condition identified in the initial history and physical exam is almost always treated as present on admission.

Diagnoses that were not present on admission (indicator N or U) are sometimes classified as serious complications of care. They may then adversely impact quality of care and other performance measures. Examples include pressure ulcers or injuries that result in a fracture occurring during hospitalization.

The POA status of any diagnosis should be easily ascertained from the usual medical record documentation as discussed above. In unclear cases, a clinician may need to specifically indicate whether in her opinion a condition was present on admission or not. Fortunately, CMS regulations impose no time limits on when a clinician may substantiate the POA status of a condition. It can be done on admission, at any time during admission, or even after discharge.

Any clinician involved in the care and treatment of a patient may document the POA status of a condition. The clinician is only required to use reasonable clinical judgment consistent with accepted standards of practice in making this determination and does not have to offer proof to substantiate it. A clinician may also indicate that she is unable to clinically determine the POA status of a condition (indicator W).

That covers the coding rules for principal and secondary diagnoses, principal procedure, and POA status. An upcoming edition of this column will focus on rules governing uncertain (presumed) diagnoses; documentation of signs, symptoms, and abnormal test results; whose documentation can be used for code assignment; and the regulatory standards for clinical validation (substantiation) of documented diagnoses.