Diabetes classification integral to hospital quality measurement, correct code capture
Association of Clinical Documentation Improvement Specialists, October 6, 2008
By James S. Kennedy, MD, CCS
Every year, the government updates the ICD-9-CM codes we use to report illnesses.
This year, effective October 1, ICD-9-CM introduces a bevy of new codes, including a new category for secondary diabetes mellitus and its complications.
Reporting these codes to specify the etiology of diabetes increases our patients’ illness severity in risk adjustment physician and hospital profiling algorithms.
Further, when we fail to document diabetic ketoacidosis (DKA) or hyperosmolarity of any cause as being present on admission (POA), Medicare will deem this a hospital-acquired condition (HAC), potentially disqualifying additional hospital reimbursement. Deep venous thrombosis, pulmonary embolus, and certain postoperative orthopedic wound infections are affected similarly.
Ask your coding or quality manager for more information on HACs so you can consider their POA status as you document them.
Definitions and criteria are crucial
- The clinical criteria for diabetes mellitus include:
- Two or more fasting blood glucoses greater than 126 mg/dl
- Symptoms of diabetes (e.g., polyuria or polydipsia) with a random blood glucose greater than 200 mg/dl
- A two-hour blood glucose greater than 200 mg/dl during an oral glucose tolerance test
- Criteria for gestational diabetes are stricter. Pregnant women require only a fasting blood glucose greater than 95 mg/dl at 24–28 weeks of gestation to make this diagnosis.
- Many physicians and ICD-9-CM categorize hyperglycemic states as follows:
- Hyperglycemia (code 790.29), impaired fasting glucose (code 790.21), or impaired glucose tolerance test (code 790.22) indicate only an elevated blood glucose. Although severe hyperglycemic states (glucose greater than 250 mg/dl) are associated with increased morbidity or mortality, code 790.29 does not have the same predictive power as uncontrolled diabetes or DKA in ICD-9-CM-based risk models.
- Type 1 diabetes (codes 250.x2 or 250.x3) is due to immunologic or idiopathic beta-cell destruction (absent C-peptide level) and can occur at any age.
- Type 2 diabetes (codes 250.x0 or 250.x1) is due to a range of insulin resistance and secretory defects (normal or elevated C-peptide levels). ICD-9-CM maps “maturity-onset diabetes of the young” to type 2 diabetes.
- Secondary diabetes (new codes 249.x0 or 249.x1) is due to external agents (e.g., chemicals or drugs) or other disease processes (e.g., Cushing’s disease or hemochromatosis). Diabetes due to pancreatectomy still maps to code 251.3.
- Gestational diabetes (code 648.8x) refers to diabetes with an onset that occurs during pregnancy. When diabetes is present prior to pregnancy, report code 648.0x with the appropriate diabetes code.
- Neonatal diabetes (code 775.1 refers to true diabetes in a newborn. Report code 775.0 for infants of diabetic mothers.
Note that the terms “IDDM” and “NIDDM” are obsolete and should be discarded. If we use these terms, expect coder queries for greater specificity.
‘Poorly controlled’ diabetes is ‘well controlled’
For codes 249.xx and 250.xx, the fifth digit indicates the type of diabetes and/or whether it is controlled or uncontrolled. We must document that the diabetes is “uncontrolled” or “out of control” for a coder to report it as uncontrolled.
When we document “poorly controlled diabetes,” ICD-9-CM interprets this as “well controlled” (see my letter to the editor in Annals of Internal Medicine, available at www.annals.org/cgi/reprint/145/5/394-a.pdf).
The criteria for uncontrolled diabetes is individualized by the treating physician; however a HgbA1C greater than 7% or multiple blood glucoses greater than 250 mg/dl usually indicate uncontrolled diabetes.
DKA generally is diagnosed when blood glucose levels of 250–600 mg/dl are associated with an elevated anion gap metabolic acidosis and ketonemia. Diabetic hyperosmolarity requires blood glucoses greater than 600 mg/dl (often greater than 1,000 mg/dl) with no acidosis or ketonemia.
Although uncontrolled diabetes does not affect most DRG-based hospital reimbursement (Maryland is an exception), DKA and diabetic hyperosmolarity serve as major complications and comorbidities when documented as POA.
Even though DKA can occur in type 2 diabetics, Coding Clinic requires coders to assign it as type 1 diabetes unless the physician specifies otherwise. Although Coding Clinic allows coders to assume that diabetes is uncontrolled in diabetic ketoacidosis, it does not assume the same for diabetic hyperosmolarity.
Even when the blood glucose is greater than 1,000 mg/dl, we must explicitly state that diabetes with hyperosmolarity is uncontrolled.
Long-term complications make a difference
Many diabetics develop long-term complications that, when documented and linked to their cause, affect risk-adjustment methodologies. Physicians commonly document neuropathy, nephropathy, and vasculopathy but fail to link these to their patients’ diabetes.
We need to remember that:
- Specifying all the underlying causes of the diabetes, such as diabetic angiopathy, diabetic neuropathy, pressure sores, or a combination thereof in diabetic ulcers, increases their severity.
- If a diabetic is taking an ACE-inhibitor or an angiotensin-receptor blocker, the coder needs to know whether there is a diabetic nephropathy as well as the stage of any chronic kidney disease (1–5).
- The role of a diabetic cardiomyopathy in causing diastolic heart failure adds specificity.
- Even when the diabetic has erectile dysfunction, the coder needs to know whether it is related to a diabetic vasculopathy or an autonomic neuropathy.
Thank you again for your attention to these matters and the support that you give your coding staff in their quest for data quality.
Editor’s note: Kennedy is a general internist and director at Brentwood, TN–based FTI Healthcare, which specializes in medical management, case management, clinical documentation, and quality reporting. You can contact him at 615/324-8676 or by e-mail at james.kennedy@ftihealthcare.com.
Most Popular
- Articles
-
- Note from Hugh
- CMS seeks comment on quality measures
- Note from the instructor: OIG report on usage of financial liability "G" modifiers
- Recent Recovery Auditor activity
- CMS releases new QAPI resources
- The week in Medicare updates
- HIPAA Q&A: Receiving faxed HEDIS requests
- Remind your workforce members to ’zip their lips’ when it comes to patient privacy
- Documentation of medical necessity drives successful RA appeals
- CMS says it's not too late to avoid payment adjustments
- E-mailed
-
- Note from the instructor: OIG report on usage of financial liability "G" modifiers
- Q/A: How do we report therapy G codes and modifiers for multiple therapies?
- HIPAA Q&A: Receiving faxed HEDIS requests
- CMS says it's not too late to avoid payment adjustments
- FDA makes new proposal related to C. diff and other threatening pathogens
- Demand a code for demand myocardial infarction
- Eyes see more ICD-10-CM codes because of laterality
- News: Study shows increase in observation services
- Product of the week: Optimizing PEPPER in the Audit Environment
- Dangers of reporting costs improperly
- Searched