Master malnutrition definitions, coding rules

CDI Blog - Volume 7, Issue 29

Malnutrition is at its most basic level any nutritional imbalance. While it can be overnutrition, such as being overweight, obese, or morbidly obese, providers more commonly equate malnutrition with undernutrition, which is a continuum of inadequate intake, impaired absorption, altered transport, and altered nutrient utilization.

Before 2012, no standard criteria existed for adult or pediatric malnutrition. Providers often equated low serum albumin or prealbumin with malnutrition, even if there was no weight loss or dietary invention. As a result, the Baltimore U.S. Attorney launched fraud and abuse investigations against Johns Hopkins Bayview, Good Samaritan Hospital, and Kernan Hospital, all in Baltimore, within the past five years.

Malnutrition is also underdiagnosed, given the lack of physician knowledge of standardized criteria and a dependence on the serum albumin or prealbumin as a clinical indicator, according to James S. Kennedy, MD, CCS, president of CDIMD in Smyrna, Tennessee.

The landscape changed dramatically in 2012 with the release of a consensus statement by The American Academy of Nutrition and Dietetics (the Academy) and the American Society for Parental and Enteral Nutrition (ASPEN) standardizing the criteria for adult malnutrition.

In their paper, the Academy and ASPEN stated that non-severe (moderate) and severe malnutrition in various inflammatory states should be diagnosed when criteria in at least two of the following six categories are identified:

  • Insufficient energy intake
  • Weight loss
  • Loss of muscle mass
  • Loss of subcutaneous fat
  • Localized or generalized fluid accumulation that may sometimes mask weight loss
  • Diminished functional status as measured by hand grip strength

The Academy and ASPEN do not differentiate between mild and moderate malnutrition in adults, classifying both as "nonsevere (moderate)", Kennedy says.

ICD-9-CM and ICD-10-CM classifies "non-severe" malnutrition as "unspecified," meaning the provider would have to document "moderate malnutrition" to obtain the ICD-9-CM or ICD-10-CM code specificity implied in this consensus statement.

ASPEN and the Academy further stated that serum albumin, pre-albumin, and similar biometrics are not useful indicators for malnutrition, given that serum levels of these proteins do not change with nutritional interventions, says William E. Haik, MD, FCCP, CDIP, director of DRG Review, Inc., in Fort Walton Beach, Florida.

"The reason being that they all are actually what we call acute phase reactants, meaning that they're falsely lowered in patients who have inflammation, even if that inflammation is chronic or very subtle," Haik says. "A patient with just merely starvation has inflammation, so the other tests we can do show inflammation."

Pediatric criteria broke new ground

Not to be outdone, an interdisciplinary ASPEN workgroup of physicians, nurses, dieticians, and pharmacists later released standardized pediatric malnutrition criteria in 2013. The American Academy of Pediatrics endorsed the criteria.

Unlike the adult criteria, ASPEN classified pediatric malnutrition according to the age-dependent World Health Organization (WHO) (children younger than 2) or CDC (age 2 to 20) anthropometric relationships to growth curves, known as Z-scores, and retained the ICD-9-CM classification of mild, moderate, and severe malnutrition.

"The pediatric criteria is ground-breaking," says Kennedy. "Coders, clinical documentation specialists, and pediatricians in children's hospitals now have a definitive reference whereby pediatric malnutrition can be diagnosed and, if present, documented and coded as to measure its impact on resource utilization and outcomes."

Compliance officers benefit also because Z-scores are not as subjective as some of the metrics in the adult criteria, Kennedy says.

Marasmus and kwashiorkor

In the criteria, neither the Academy nor ASPEN discussed other nutritional definitions, such as marasmus and kwashiorkor, which have challenging ICD-9-CM and ICD-10 codes that affect most severity and risk adjustment methodologies.

Marasmus is generally defined as generalized starvation with loss of body fat and protein. While not common in the United States, it may be seen in patients who have anorexia nervosa or other eating disorders.

Kwashiorkor is generally defined as selective protein malnutrition with edema and a fatty liver. It occurs mainly in connection with acute, life-threatening illnesses such as trauma and sepsis and with chronic illnesses that involve acute-phase inflammatory responses. The now calls kwashiorkor "severe acute malnutrition."

ICD-9-CM code 260 (kwashiorkor) has generated coding compliance scrutiny exhibited by articles in the California media and investigations by the Office of Inspector General (OIG).

The ICD-9-CM Index to Diseases fueled this controversy by classifying any documented protein malnutrition to kwashiorkor (an MCC in MS-DRGs) until advice from Coding Clinic for ICD-9-CM, Third Quarter 2009, p. 6, stated that documented mild or moderate protein malnutrition codes to its respective malnutrition code (a CC in MS-DRGs), not kwashiorkor.

Coding Clinic was silent on how to code "protein malnutrition" without a statement of its severity, leading some hospitals to continue to assign code 260 when the physician only documents protein malnutrition based on the code's listing in the ICD-9-CM Index to Diseases.

The ICD-10-CM Index to Diseases classified "protein malnutrition" only as E46, (unspecified protein calorie malnutrition), not as kwashiorkor nor with the severity levels associated with the protein-calorie malnutrition codes.

Given that kwashiorkor is rare in the United States and that this diagnosis is being monitored by the OIG, coders should not report ICD-9-CM code 260 (kwashiorkor) when a physician only state protein malnutrition, Haik says.

The physician should explicitly document the term "kwashiorkor" in the medical record before coders report 260, Kennedy adds. If coders are uncertain about whether the patient has kwashiorkor, they should query the provider.

Cachexia

Cachexia is a multifactorial syndrome characterized by severe body weight, fat, and muscle loss and increased protein catabolism due to underlying disease(s). Cachexia is considered the result of the complex interplay between underlying disease, disease-related metabolic alterations, and, in some cases, the reduced availability of nutrients.

 In one review, the criteria for cancer-related cachexia mirror the Academy/ASPEN criteria for malnutrition. Physicians will use these terms interchangeably, even though they are quite different.

Coders will more often see an actual diagnosis of mild, moderate, or severe malnutrition instead of cachexia, says Mindy Hamilton, RD, LD, a registered dietitian from Kansas City, Missouri. Cachexia may be present, but the patient could be a 90-year-old woman who lives alone, doesn't take in a lot of food, and still gets the nutrients that she needs, Hamilton says.

Cachexia by itself or due to cancer or malnutrition is classified by the ICD-9-CM Index as 799.4 (cachexia). If the physician only documents cachexia, coders must query to determine cachexia's etiology, because the ICD-9-CM Alphabetic Index considers cachexia to be integral to a number of other conditions when they coexist, including:

  • Hypopituitarism
  • Heart disease
  • Lead poisoning
  • Old age
  • Malaria
  • Tuberculosis

ICD-10-CM is similar, but differs in that a code for cachexia may not be used if marasmus is also documented.

ICD-9-CM classifies the term "wasting" to 799.4, and wasting due to malnutrition is classified as 261. The ICD-9-CM Index to Diseases also classifies the term "emaciation," an excessive leanness or wasting of the body, as 261 (a MCC in MS-DRGs). Coding Clinic for ICD-9-CM, First Quarter 2013, p. 13, states that "marasmus" be coded as 799.4, (a CC in MS-DRGs), unless the physician explicitly documents that the patient has malnutrition. Whether this concept applies in ICD-10-CM requires further clarification from Coding Clinic for ICD-10-CM.

While ICD-9-CM classified cachexia due to malnutrition as only 799.4, many coders may add an additional code for the documented malnutrition. Auditors may challenge this because the index classifies cachexia due to malnutrition as 799.4, and does not explicitly allow for the addition of a malnutrition code, Kennedy says.

On the other hand, if cachexia only coexists with (and is not "due to") malnutrition, coders may report an additional code for the documented malnutrition because the provider did not state that the cachexia was due to the malnutrition. (Coding Clinic, Second Quarter 2012, pp. 20-21)

Nutritional deficiency, starvation, and malnutrition

In coding nutritional deficiency states, coders must remember that ICD-9-CM and ICD-10-CM code assignment is based upon complete and consistent provider documentation of clinically valid conditions and proper use of the ICD-9-CM Index and Table to Diseases. Recovery auditors are authorized to challenge these codes if the circumstances do not support the documented diagnosis upon which the code is based.

Documentation of the term "deficiency" with various adjectives offer coding and compliance challenges. In the ICD-9-CM Index to Diseases, the term "deficiency" with various adjectives include:

  • Calorie, severe: 261, marasmus
  • Edema: 262, other severe protein-calorie malnutrition
  • Multiple, syndrome: 260, kwashiorkor
  • Nutrition, nutritional: 269.9, unspecified nutritional deficiency
  • Specified NEC: 269.8, other nutritional deficiency
  • Protein 260: kwashiorkor

"A dietary deficiency in and of itself is not coded as malnutrition unless ICD-9-CM specified adjectives with the deficiency are added to it," Kennedy says. "Even so, if only a specified nutritional deficiency coding to malnutrition is documented, given the Coding Clinic advice for emaciation referenced above, I would query the provider to ascertain the validity of the malnutrition code before it is submitted."

ICD-9-CM classifies starvation and food deprivation by itself with code 994.2 (effects of hunger). Coders can't assume a patient is malnourished just because the physician writes starvation, Kennedy says. The physician must define and document the malnutrition.

Coders can locate the appropriate code for the degree of malnutrition by first referencing the ICD-9-CM Index to Diseases, following the instructions in the Table to Diseases, and being aware of pertinent Coding Clinic advice.

If a physician documents "moderate-severe malnutrition," coders or CDI specialists must query the physician to determine which term applies, according to Coding Clinic, Third Quarter 2012, p. 10. If the answer is "severe," assign code 261. If the answer is "moderate," assign code 263.0 (malnutrition of moderate degree).

"We cannot code that documentation unless we query the physician," Kennedy says. The answer will affect the MS-DRG because 261 is an MCC, while 263.0 is only a CC.

Coding guidelines impact malnutrition-related code assignment

Coders also need to keep the ICD-9-CM Official Guidelines for Coding and Reporting in mind when coding malnutrition.

Physicians are allowed to think out loud and use terms like "probable," "suspected," "likely," "questionable," "possible," or "still to be ruled out" in their documentation, Kennedy says.

However, coders can only report these uncertain conditions if the physician documents them at the time of discharge (e.g. in the discharge summary, discharge note, or discharge order) and if the clinical circumstances or treatment reasonably support the coded diagnosis. Coding Clinic, Third Quarter 2005, p. 22 emphasized that admitting or interim notes cannot be used for uncertain diagnoses.

According to the ICD-9-CM Official Guidelines for Coding and Reporting, coders may report body mass index (BMI) and pressure ulcer stages based on medical record documentation from clinicians (e.g. nurses or dieticians) who are not the patient's provider (i.e., physician or other qualified healthcare practitioner legally accountable for establishing the patient's diagnosis). Coding Clinic, Fourth Quarter 2008, p. 19, however, states that a BMI documented by a dietician cannot be coded unless the provider documents a nutritionally-related diagnosis, such as obesity, overweight, underweight, or malnutrition, Kennedy says. These affirm that the BMI code meets the guidelines' definition of an additional diagnosis.

While coders may not code obesity, morbid obesity, or malnutrition from a dietician's note, given that dieticians are providing direct patient care and are expert in malnutrition's clinical criteria, some hospitals allow dieticians to add clinically valid nutritional diagnoses to the problem list. If that list is imported into a progress note and authenticated by a treating provider, the condition may then be coded. To ensure coding compliance, the provider should document the progress of the nutritional diagnosis with the dietary intervention and optimally add the nutritional diagnosis to the discharge summary.

Editor’s note: This article was originally published in the November issue of Briefings on Coding Compliance Strategies. Email your questions to Senior Managing Editor Michelle A. Leppert, CPC, at mleppert@hcpro.com.

Found in Categories: 
Clinical & Coding