Identify Causal Relationship Links in Diabetes
Small differences in documentation language play an important role in diagnostic coding.
A key concept for accurate diabetes coding is that of causal relationships. A causal relationship is a documented link between a disease (etiology) and a condition (manifestation) caused by that disease. For a coder to report a causal relationship between diabetes and a complication of diabetes, the provider must establish an unambiguous link in the documentation.
To illustrate causal relationships, consider these three assessments:
- Chronic kidney disease (CKD) stage 3
In this assessment, the provider does not document a causal link between the diabetes and the kidney disease. Correct code assignment is 250.00 Diabetes mellitus without mention of complication, type II or unspecified type, not stated as uncontrolled and 585.3 Chronic kidney disease, Stage III (moderate). You may not assume a causal relationship where documentation has not established one.
Report the diabetes as type 2, although this is not documented. This instruction is from the ICD-9-CM Official Guidelines for Coding and Reporting, and is a guideline based on the demographic that more than 90 percent of diabetic patients have type 2 diabetes.
- Diabetes and nephropathy
- CKD stage 3
In this second assessment, the provider does not successfully link the diabetes and the nephropathy, although it was likely the provider’s intent was to make that link. Correct coding in this instance includes 250.00 (diabetes), 583.9 Nephritis and nephropathy, not specified as acute or chronic, with unspecified pathological lesion in kidney (nephropathy), and 585.3 (stage 3 CKD). You may be tempted to report 581.81 Nephrotic syndrome in diseases classified elsewhere for the nephropathy, but this is a manifestation code that may be used only when a causal relationship has been established. The correct nephropathy code is found in the index under:
Nephropathy (see also Nephritis) 583.9.
Examples of acceptable causal links in diabetes include when a manifestation is preceded by the word “diabetic,” as in “diabetic retinopathy,” or followed by the phrases “of diabetes” or “due to diabetes.” “Diabetes and” does not establish a causal relationship, nor does the phrase “diabetes contributing to” because it only establishes a factor, not a cause.
These small differences of language are critical in diagnostic coding audits. Correct documentation of causal relationships does not increase the provider’s documentation burden; whenever possible, you should guide providers toward better documentation practices.
- Diabetic nephropathy
- CKD Stage 3
In this assessment, the provider clearly links the diabetes to the kidney disease, by stating “diabetic” nephropathy. Report codes 250.40 Diabetes with renal manifestations, type II or unspecified type, not stated as uncontrolled, 581.81 Nephrotic syndrome in diseases classified elsewhere, and 585.3 (Stage 3 CKD).
Hyperglycemia, as documented in all three assessments, is a symptom of diabetes and would not be separately reported. You cannot assume that high blood glucose means the patient has uncontrolled diabetes. This coding decision is supported in the index:
hyperglycemia – code to Diabetes, by type, with 5th digit for not stated as uncontrolled.
There are a few exceptions to the documentation rule for causal relationships in diabetes. Loss of protective sensation (LOPS), gangrene, and osteomyelitis may be reported as complications of diabetes even when no link is established in the documentation, according to Coding Clinic entries (LOPS, 4th Quarter 2009; gangrene, March 1986; osteomyelitis, 1st quarter 2004). Also, 2nd Quarter 2009 Coding Clinic states that diabetes “with” neuropathy establishes a causal link between the diabetes and the neuropathy. It would be unwise to extrapolate that rule across all diabetic complications, however.
New Rules on the Horizon
Don’t get too comfortable abstracting undocumented links between diabetes and neuropathy, osteomyelitis, gangrene, or LOPS because ICD-10-CM will start the rule-making all over again. An ICD-10-CM Coding Clinic entry from 4th Quarter 2013 is evidence that changes are already happening. It reads:
ICD-10-CM does not presume a linkage between diabetes and osteomyelitis. The provider will need to document a linkage or relationship between the two conditions before it can be coded as such.
Diagnosis coding is gaining more importance among professional coders as the number of risk-adjusted claims increases, year over year. Payment under risk adjustment is affected by these causal relationships, making it important for everyone to understand the causal relationship rules and code accurately for ICD-9-CM and ICD-10-CM. Following the guidelines, as well as the rules within Coding Clinic, will ensure accurate coding as you move from one code set to the other.
Sheri Poe Bernard, CPC, CPC-H, CCS-P, is an independent consultant specializing in risk adjustment coding quality and training, and author of Netter’s Atlas of Human Anatomy for CPT Surgery, to be published by the American Medical Association in early 2015. She is a member of the Salt Lake City, Utah, local chapter.